international journal of renewable energy development int. journal of renewable energy development 1 (3) 2012: 99-105 p a g e | 99 © ijred – issn : 2252-4940 contents list available at ijred website, issn : 2252-4940 int. journal of renewable energy development (ijred) journal homepage: www.ijred.com the energy processing by power electronics and its impact on power quality j. e. rocha and w. d. c. sanchez * department of electrical engineering, federal university of technology paraná, brazil article history: received sep 19, 2012 received in revised form sep 24, 2012 accepted sep 30, 2012 available online abstract: this paper discusses the electrical architectures adopted in wind turbines and its impact on the harmonic flux at the connected electric network. the integration of wind electric generators with the power grid needs energy processing by power electronics. it shows that different types of wind turbine generator systems use different types of electronic converters. this work provides a discussion on harmonic distortion taking place on the generator side, as well as in the power grid side. keywords: grid connection, harmonic distortion, power electronics and converters, wind energy conversion systems, wind power, wind technology, wind turbines * corresponding author: email:walterdcsan@terra.com.br 1. introduction many improvements and innovations have been made since the wind turbines began to become important as an alternative methods of generating electricity from the 1980s. however, the basic electrical architectures are still present. an important innovation that has emerged in recent years was the direct drive and the use of generators with permanent magnets. direct drive is a transmission concept for wind turbines in which there is no gearbox, no speed multiplier. durability of these gearboxes has been a serious problem for a long time. conventional technologies such as synchronous generator, asynchronous generator and doubly-fed induction generator are commonly used. each of these technologies requires a specific electrical connection to the grid in order to process energy. controlling power flow into the utility grid is required to ensure compatibility operation. the grid interface is provided through the use of electronic power converters to control the energy and adapt it to the conditions required by the electrical system. even the induction generators which are connected directly to power grid need an electronic converter to start the connection process to the utility grid. in this case, it’s used a soft-starter converter in which the output voltage rises smoothly and a successful startup occurs. using power electronic in the various electrical architectures can increase performance and reduces the size and cost of the system. however, electronic converter may cause a degradation of power quality. the aim of this work is to discuss circuit topologies adopted in wind turbines and its impact on the harmonic flux at the grid side converter and at the machine side. the paper also discusses how to mitigate harmonic distortions in the point of connection to the grid in order to achieve regulatory goals. 2. harmonics several static converter topologies are applied to wind energy processing solutions. they are used to the regulation of power flow too. all power electronic converters generate harmonic currents, which will be injected into the utility grid, causing distortion of the utility voltage waveform. interestingly enough, the technology that allows for more efficient control of electrical energy, switch-mode power conversion technology, is also responsible for a negative impact on power quality. regulatory standards for utility system are critical to maintaining the quality of ac power distribution citation: rocha je and sanchez bwdc (2012) the energy processing by power electronics and its impact on power quality. int. journal of renewable energy development 1(3): 99-105 p a g e | 100 © ijred – issn : 2252-4940 systems. different measurement considerations and methodologies are used for determining harmonics compliance. total harmonic distortion is an important index widely used in defining the level of harmonic content in electrical facilities. the total harmonic distortion quantifies the thermal effect of all the harmonics. it is the ratio of the rms value of all the harmonics to the fundamental (iec 61000-2-2). harmonics distort the sine wave which becomes apparent when a distorted current or voltage waveform is mathematically analyzed. through fourier analysis, an arbitrary periodic function can be divided into a number of sine waves. a non-sinusoidal signal can be decomposed in a sum of some of its multiple in frequency called harmonics. a distorted current or voltage waveform consists of the fundamental frequency and harmonics. most harmonic analysis algorithms are based on fft to obtain the voltage and current frequency spectrum from discrete time samples. fft is a very efficient algorithm for evaluating fourier transform. this algorithm becomes considerably more efficient when a large number of data in a discrete sequence need to be transformed into frequency domain. it is the objective of the electric utility to supply its customers with a sinusoidal voltage of fairly constant magnitude. the generators that produce the electric power must generate a very close approximation to a sinusoidal signal. grid connected wind turbines are equipped with power converter systems that produce besides the fundamental signal also the harmonic signal. power electronic converters in wind energy systems generate harmonic currents and harmonic voltages, which are injected into power system. harmonic filters reduce distortion by diverting harmonic currents in low impedance paths. these filters can be either active or passive in nature. passive filters use passive components in their design. they are considered passive because they do not use any active components like semiconductors. modern active filters are superior in filtering performance, smaller in physical size, and more flexible in application, compared to traditional passive filters using capacitors, inductors and/or resistors. active filters use sophisticated electronics and power igbts to inject equal and opposite harmonics onto the power system to cancel those generated by converters. different topologies of the interfacing converter can be used in wind energy conversion to minimize the current harmonic content. most of the topologies without isolation have an intermediate dc link. it means that on the generator side there is a rectifier while on the grid side is placed an inverter. some low pass filter can be added to the inverter to fulfill standard requirements. the overall performance of an interface converter is highly affected by the properties of the combination of the generator side and grid side converter. 3. electronic converters electronic converters can be used to convert power supplies from one voltage and frequency to another. a common example is the system that first converts the alternating current (ac) to direct current (dc) and then returns the dc back to ac. a power electronic device for converting alternating current to direct current is the rectifier. a power electronic device that converts direct current to alternating current is called inverter. the conversion is performed with semiconductor switching devices such as diodes and transistors. variable speed wind turbines have some advantages over fixed generation speed, but in this case, the frequency is variable. for this reason, it is necessary to rectify the ac signal coming from the synchronous generator and operate an inverter to synchronize its frequency, phase and amplitude with the utility. the inverter synthesizes an alternating current sine wave from the direct current voltage by the switching frequency. the pwm drives the gates of the power transistors in the inverter. control is by a voltage source pwm inverter that uses igbts for power transistors. the igbts are configured in a full h-bridge with freewheeling diodes. figure 1 shows a three-phase rectifier topology, a dc link and a frequency inverter. the freewheeling diodes are not shown in the figure 1. the ability to produce near sinusoidal waveforms around the fundamental frequency is dictated by the modulation technique controlling when the power transistors are on and off. the topology shown in figure 1 is responsible for keeping the frequency at a fixed value. a fixed frequency output waveform is generated by the inverter to synchronize with the utility. for three-phase spwm, three modulating signals that are 120 degrees out of phase with one another are used in order to produce out of phase voltages. the rectifier used in figure 1 is a simple diode bridge rectifier with capacitor voltage filter. this type of rectifier inherently draws non-sinusoidal current from the generator. fig. 1 ac to ac converter topology int. journal of renewable energy development 1 (3) 2012: 99-105 p a g e | 101 © ijred – issn : 2252-4940 there is a new topology called active three-phase rectifier that shape the three-phase input current to be sinusoidal and to be in phase with the generator output voltage. in this new topology, also called back-to-back pwm-vsi, the diodes are replaced by six controlled switches to shape the input current. the advantage of this new topology is to reduce alternating current harmonic content and improve power factor to generator side. this topology is shown in figure 2. the capacitor decoupling between the grid inverter and the generator inverter offers separate control of the two inverters, allowing compensation of asymmetry both on the generator side and on the grid side, independently [1]. figure 3 shows the harmonic spectrum of a six pulse diode bridge rectifier. the spectrum is the distribution of the amplitudes of the various harmonics as a function of their harmonic number, often illustrated in the form of a histogram. the amplitude of a harmonic is often expressed with respect to that of the fundamental. in the example of figure 3, the fifth harmonic current has a magnitude close to 50% over the value of the amplitude of the fundamental current. it is shown that as the order of harmonic increases, the current amplitude decreases. the amplitude of the harmonics may vary depending on the value of the capacitance in dc link and the power being transferred through the rectifier. also, the value of inductance that exists between the rectifier and the source influences the amplitude of the harmonics. fig. 2 the back-to-back pwm-vsi topology fig. 3 harmonic spectrum of a six pulse diode rectifier the main effects of current harmonics within the armature winding of the generator are additional losses in the stator and in the rotor (copper and iron), and pulsating mechanical torque. this causes a reduced efficiency and mechanical fatigue due to vibrations. harmonics increase the rms value of the current for an equal active power generated. it is relatively common practice to “derate” (reduce the output of) generators when supplying nonlinear loads to minimize the effects of harmonic heating. many of these problems can be resolved, however, using low pass filters at the input of the rectifier. the system can use active igbt rectifier in place of conventional diode rectifier. an active rectifier by its construction produces extremely low levels of harmonics. this ensures a total harmonic distortion of less than 5%. the last stage of the topology shown in figure 1 represents the frequency inverter. the dc to ac power conversion is the result of power switching devices, which are commonly fully controllable semiconductor power switches like igbt’s. the output waveforms are therefore made up of discrete values, producing fast transitions rather than smooth ones. the ability to produce near sinusoidal waveforms around the fundamental frequency is dictated by the modulation technique controlling when the power switches are on and off. modulation technique commonly used includes the spwm technique. three-phase inverter is used in applications that require sinusoidal voltage. figure 4 shows an example of a sine wave generated by pwm technique. three equal waveforms are produced but are phase shifted by 120 degrees. the pwm waveform has harmonics of several orders in the voltage waveform. the other two phases are identically generated but 120 degrees apart in phase. the line to line voltage will not have any triplen harmonics. pwm techniques produce an ac output waveform from a dc source. functions of inverters for wind energy systems are power conversion from variable dc voltage into fixed ac voltage and frequency, and output power quality assurance with low total harmonic distortion. fig. 4 pwm sine wave citation: rocha je and sanchez bwdc (2012) the energy processing by power electronics and its impact on power quality. int. journal of renewable energy development 1(3): 99-105 p a g e | 102 © ijred – issn : 2252-4940 fig. 5 the effect of switching frequency in the sine wave the frequency is controlled by controlling the period of the pwm and that is built up with a microprocessor system and synchronized with the utility frequency. the rms value of output voltage can be controlled by varying the width of the pulse. a little longer pulse width will result in a little higher rms output voltage. the harmonic content at the inverter output has to be effective reduced to admissible values. by controlling the switches to produce pwm, and filtering the output, a nearly sinusoidal line current waveform is obtained. different output filter topologies are commonly used to interface inverter to the grid, namely the l, lc and the lcl filter. the use of the filter coupling the inverter to the grid reduces the high frequency pollution of the grid that can disturb electric system. the harmonic attenuation with using filter permits the use of a lower switching frequency to obtain harmonic limits, increasing efficiency. figure 5 shows the topology required for the power stage of a three-phase inverter and different output waveforms. different modulation frequencies were used to generate the waveforms shown in figure 5. in this figure, the first waveform has a large ripple. in the second waveform, the switching frequency increased and there was a decrease in the ripple. in the third waveform, the switching frequency is higher and the ripple is no longer noticed. in order to generate more precise sinusoidal ac current waveforms and keeping the size of the lc filter small, high modulation frequencies can be used. 4. filters a sinusoidal pulse width modulation is shown in figure 6a and a sinusoidal waveform adapted to connect with the grid can be seen in figure 6b. those waveforms were acquired with an oscilloscope. the pwm waveform shown in figure 6a was filtered before the connection with the grid, in other words, the second waveform, figure 6b, was originated by the first one, figure 6a. fig. 6 sinusoidal pulse width modulation (a) and sinusoidal filtered voltage (b) fig. 7 lc-type low-pass filter in accordance with figure 6, the voltage waveform modulated through the frequency inverter is not adapted to connect with the grid. as a result, a series inductor is placed between the grid and the inverter in order to permit the instantaneous differences of voltage between the both sides. this inductor also dampens the sudden current variations caused by that instantaneous voltage differences. in addition to the inductor, is placed a capacitor to storage energy and to control the current flux to the electrical system. the capacitor effect is to attenuate any suddenly variation of voltage waveform. the emission of harmonic and interharmonic currents from wind turbines with directly connected induction generators has been expected to be negligible in service. wind turbines connected to the grid through power converters however emit harmonic and/or interharmonic currents and contribute to the voltage distortion. inverters based on new technologies have a limited emission of harmonics at lower frequencies compared to the converters used in the first generation of variable speed wind turbines. instead they produce interharmonics at higher frequencies which are easier to filter than at lower frequencies [2]. an lc filter is used to filter the frequencies of high order generated by the inverter. figure 7 shows an lctype low-pass filter usually employed to eliminate the switching frequency component of output harmonics. int. journal of renewable energy development 1 (3) 2012: 99-105 p a g e | 103 © ijred – issn : 2252-4940 a low pass filter only allows low frequency signals from zero hertz to its cut-off frequency to pass while blocking those any higher. in other words, the filter practically maintains the amplitude of the fundamental voltage component generated by the inverter. however, it attenuates the amplitude in high frequencies. depending on the characteristic of the grid, the harmonics can be filtered through the leakage inductance of the transformer and the intrinsic capacitance of the cables. an alternative to the filter previously described is the use of a shunt lc filter tuned in the switching frequency. the advantage of this solution is that this filter would need to be sized only for the rms value of the harmonic current [3]. 5. power quality regulation the energy processing through electronic converters adds advantages to the whole system, but it affects the power quality. this does not necessarily affect the operation of the electrical system. the important thing is that the waveform parameters must be in compliance with the indexes recommended by the energy regulatory authority. the brazilian electricity regulatory agency determines, through indices defined by resolutions, when a voltage waveform is in accordance with the prescriptions for a good operation of the grid. those power quality indices can be found in the distribution procedures of electric energy in the national electrical system [4] and in the management of performance indices of basic system and its components [5]. international standards can be used as reference when national standards or recommendations do not try a specific problem. a recommendation used internationally is the “recommended practices and requirements for harmonic control in electrical power systems” [6]. this recommendation defines maximum values for each harmonic voltage in percentage of the fundamental voltage. the values of the harmonic currents are defined too. the acceptable value for the voltage and current harmonics depends on the system characteristics. an additional recommendation used specifically in wind turbines is the “european standard for testing of wind turbines” [7]. in this recommendation is defining a set of requirements to ensure good power quality in the connection of a wind turbine system with the power grid. 6. wind turbine grid connection there are different technologies to carry out the connection between the wind generator and the electric system. each technology seeks to adapt to the conditions of variation in wind speed, the power available on the shaft, and the requirements for constant frequency and voltage in the power output. fixed speed wind turbines cannot use the full potential of wind energy and therefore are giving way to wind turbines with variable speed control [8]. the asynchronous generators, also called induction generators, are used when the turbine speed will be maintained constant. figure 8 shows a schematic diagram of a wind turbine using an induction generator. this system has a simple construction, low cost, low maintenance and dispenses of synchronism devices. there are two basic needs in this system: thyristors are used to promote a soft starter providing the attenuation of the high starting currents, and the requirement of reactive power to create a magnetic flux necessary to operate the induction generator. this reactive power is usually supplied through capacitor banks, so avoiding the low power factor system. the electronic converter with thyiristors is used in the starting and stopping of the turbine, remaining inactive during steady state operation. this electronic converter is disconnected through a bypass during steady state operation. harmonics are generated during both the starting and stopping operation and that can affect the system operation. the turbine speed can be kept constant by controlling the pitch angle. another way to maintain the speed constant is through the special construction of the blade so that it reaches the stall condition. some wind turbine systems classified as asynchronous generators type use two generators with different number of poles or a generator with two windings in the stator that permit the operation of the system in two different speeds. a second system used is the electric generator with wound rotor, also called the doubly fed asynchronous generator. in this case, the turbine rotor operates with variable speed set to the real situation of the wind speed. the stator winding is directly connected to the grid while the rotor winding is connected to an electronic converter and this is connected to the grid. figure 9 shows a schematic diagram of a wind turbine using a doubly fed asynchronous generator. in this system, slip control of the asynchronous machine allows the rotor to vary its speed without varying the frequency of the voltage waveform in the stator. the slip control is done through an electronic converter that changes the slope of the torque curve of the generator. the transformer shown in figure 9 is required to adjust the level of the output voltage of the converter with the voltage level of the stator. fig. 8 wind turbine with asynchronous generator citation: rocha je and sanchez bwdc (2012) the energy processing by power electronics and its impact on power quality. int. journal of renewable energy development 1(3): 99-105 p a g e | 104 © ijred – issn : 2252-4940 fig. 9 wind turbine with doubly fed asynchronous generator fig. 10 wind turbine with synchronous generator the electronic converter operates in steady state, so harmonics are continually generated. however, as the largest portion of power goes directly to the grid by the stator, a smaller proportion of harmonics is generated. the doubly fed asynchronous generator requires more maintenance because it has collector rings. these collector rings are used to connect the rotor winding to the external circuit. another system that allows using variable speed is the one that uses the synchronous generator. in this system, the rotor operates at a variable speed adjusted to the actual situation of wind speed. the frequency generated depends on the wind turbine rotation. therefore, the generated voltage must be rectified and then to pass for a frequency inverter. the voltage and frequency are adjusted to values of the grid through a frequency converter. figure 10 shows a schematic diagram of a wind turbine using a synchronous generator. an lc low-pass filter or a shunt filter is used at the converter output. all power produced by the synchronous generator passes through the electronic converter while only a small portion of the energy produced by the doubly fed asynchronous generator goes through its electronic converter. therefore, the wind turbine with synchronous generator requires a larger filter for harmonic attenuation. currently, there are permanent magnet synchronous generators that require no gearbox (speed amplifiers) because they have a high number of poles. in this case, the generator is directly driven by the rotor of the wind turbine. this increases system reliability because the gearbox is one of the points of failure. the drawback of this type of generator is its large diameter and weight. the increase in weight of the permanent magnet generator is partially compensated because there is no need for the multiplier making the system more compact. 7. conclusions the availability of electric power within acceptable parameters of power quality is essential to modern society. many modern facilities are sensitive to poor power quality. in this paper, the authors reviewed the application of electronic converters for conditioning power produced by wind turbines. the concept of harmonic voltage and current was clarified as well as the consequence of waveform distortions in the electrical system. some standards and recommendations, both national and international, were cited. the basic principle of sinusoidal waveform pwm was discussed. it was verified the necessity of using an lc filter to attenuate the distortion of voltage and current waveform. several electronic converter topologies used in wind turbines were shown. the discussion showed that the variable speed wind turbines require the use of frequency converters and these converters cause injection currents with harmonic content at the grid connection point. the fixed speed wind turbines produce harmonics only in the start and end stages of operation. these wind turbines systems do not produce harmonics in steady state. abbreviations igbt – insulated gate bipolar transistor. pwm – pulse width modulation. spwm – sinusoidal pulse width modulation. dfig – doubly-fed induction generator. fft – fast fourier transform. vsi – voltage source inverter. references [1] hansen lh, helle l, blaabjerg f, ritchie e, munk-nielsen s, bindner h, sørensen p, bak-jensen b (2001) conceptual survey of generators and power electronics for wind turbines. risø national laboratory, roskilde, denmark. [2] mittal r, sandhu ks, jain dk (2011) grid voltage control of inverter interfaced wind energy conversion system (wecs). international journal of environmental science and development 2(5). [3] araújo sv, antunes flm, sahan e a. engler b (2007) analysis and design of a lcl-filter for a npc inverter in offshore wind power conversion system. apresentado no 9° congresso brasileiro de eletrônica de potência – cobep 2007, blumenau, brasil. [4] agência nacional de energia elétrica (2008) procedimentos de distribuição de energia elétrica no sistema elétrico nacional prodist – módulo 8 – qualidade da energia elétrica. . aneel resolução nº 345. [5] operador nacional do sistema elétrico (2008) gerenciamento dos indicadores de desempenho da rede básica e de seus componentes submódulo 2.8, ons. [6] ieee recommended practices and requirements for harmonic control in electrical power systems, ieee standard: 519-1992. int. journal of renewable energy development 1 (3) 2012: 99-105 p a g e | 105 © ijred – issn : 2252-4940 [7] european standards (2008) measurement and assessment of power quality characteristics of grid connected wind turbines. en 61400-21. [8] dos reis fs, alé jav, adegas fd, tonkoski jr. r, slan e k. tan s (2006) active shunt filter for harmonic mitigation in wind turbines generators. 37th ieee power electronics specialists conference, jeju, korea. international journal of renewable energy development int. journal of renewable energy development 5 (3) 2016: 179-185 p a g e | 179 © ijred – issn: 2252-4940, october 15th 2016, all rights reserved contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: http://ejournal.undip.ac.id/index.php/ijred potential effect and analysis of high residential solar photovoltaic (pv) systems penetration to an electric distribution utility (du) jeffrey t. dellosa college of engineering and information technology (ceit), caraga state university, philippines abstract. the renewable energy act of 2008 in the philippines provided an impetus for residential owners to explore solar pv installations at their own rooftops through the net-metering policy. the net-metering implementation through the law however presented some concerns with inexperienced electric du on the potential effect of high residential solar pv system installations. it was not known how a high degree of solar integration to the grid can possibly affect the operations of the electric du in terms of energy load management. the primary objective of this study was to help the local electric du in the analysis of the potential effect of high residential solar pv system penetration to the supply and demand load profile in an electric distribution utility (du) grid in the province of agusan del norte, philippines. the energy consumption profiles in the year 2015 were obtained from the electric du operating in the area. an average daily energy demand load profile was obtained from 0-hr to the 24th hour of the day based from the figures provided by the electric du. the assessment part of the potential effect of high solar pv system integration assumed four potential total capacities from 10 mega watts (mw) to 40 mw generated by all subscribers in the area under study at a 10 mw interval. the effect of these capacities were measured and analyzed with respect to the average daily load profile of the du. results of this study showed that a combined installation beyond 20 mwp coming from all subscribers is not viable for the local electric du based on their current energy demand or load profile. based from the results obtained, the electric du can make better decisions in the management of high capacity penetration of solar pv systems in the future, including investment in storage systems when extra capacities are generated. keywords: residential solar pv system, solar photovoltaic (pv) system penetration, net-metering, energy demand, load profile article history: received july 15th 2016; received in revised form sept 23rd 2016; accepted oct 1st 2016; available online how to cite this article: dellosa, j. (2016) potential effect and analysis of high residential solar photovoltaic (pv) systems penetration to an electric distribution utility (du). int. journal of renewable energy development, 5(3), 179-185. http://dx.doi.org/10.14710/ijred.5.3.179-185 1. introduction the global energy demand situation is expected to increase by 56% between the years 2010 to 2040. the us energy information agency reports that among this increase is from the fast growing economies that includes countries in asia, particularly china and india (useia 2013). while renewable energy is one of the two (the other one being the nuclear power) fastest-growing energy source, fossil fuels remain as the leading supply for the world energy requirement through 2040 (useia 2013). in the power sector, coal emerges as the fuel of choice for electricity generation primarily because of its abundance and affordability. it is expected that the use of coal in electricity generation from today to the year 2035 will increase from less than 33% to 50% share (iea 2013). the power sector in the philippines faces the same dilemma as with other asian countries. rene almendras, then secretary of the department of energy (doe), reported on march 2012 that there is a total deficit of 170mw for the peak demand and reserve requirements in mindanao (almendras, 2012, doe 2008). the country’s doe has approved in the last two years four coal-fired power plants in mindanao for installations totaling to 1,300mw capacity to address the lack of electric power supply in mindanao, from the total 4,552mw of the sixteen approved coal-fired power plants nationwide (green peace international 2014). if this strategy by the doe which is to install more coalfired power plants in the years ahead, it has been * corresponding author: +63-917-6349497 email: jtdellosa@gmail.com citation: dellosa, j. (2016), potential effect and analysis of high residential solar photovoltaic (pv) systems penetration to an electric distribution utility (du). int. journal of renewable energy development, 5(3), 179-185, doi: 10.14710/ijred.5.3.179-185 p a g e | 180 © ijred – issn: 2252-4940, october 15th 2016, all rights reserved estimated that 90% of the total electricity generated would be coming from coal-fired power plants in the year 2020 (adb 2009). the ever-increasing use of coal as the preferred fuel for electric power generation in the country and abroad however presents disturbing environmental concerns in terms of the carbon emission to the environment (kolhe 2015, munnik et. al. 2010, grean peace int’l., 2005). carbon dioxide (co2) is one of the three green house gases (ghgs) that have increased in the atmosphere since the pre-industrial times and is verified to be the cause of climate change (ipcc 2013, ipcc 2014, usepa 2012). many countries, including countries in asia, are the most vulnerable and have already felt the ill-effects brought about by the climate change, especially the effects of extreme weather events such that of super typhoon yolanda that hit the philippines in 2013 (köppinger 2014). republic act 9154 or otherwise known as the renewable energy act of 2008 was passed into law in the philippines on december 2008. it is the first law in the country that introduced non-fiscal incentives which allows residential owners to implement or generate their own electricity from renewable energy (re) sources through the net-metering policy (dellosa 2015). the implementation of net-metering allowing the interconnections of solar photovoltaic (pv) installations, however, have yet to fully progress in the country, except for areas in metro manila. several electric cooperatives in the country have yet to implement the new policy pending more understanding on the effects of high degree integration of solar pv systems to their operations, especially on the supply and demand as reflected in their daily energy load profile. renewable energy production is considered to play a significant role to lower the preference on the use of coal as fuel for electricity generation around the world (zwickel et. al. 2012). renewable energy generation suppresses the increasing amount of co2 in the atmosphere with the use of these environment-friendly energy sources (latour et al. 2013). solar photovoltaic (pv) power, one of the renewable energy sources, harness available sunlight and convert them to usable electricity. in many countries around the world, solar pv power has already been established as the leading renewable technology that is becoming a primary source of electricity (latour et al., 2013, panzer et. al. 2016). many developing countries including the philippines, however, have a lot to progress with regards to renewable energy generation, especially at the residential levels. but the country finally produced a law on renewable energy known as the “renewable energy act of 2008” which encourages home and business owners to engage in renewable energy generation in the form of wind, solar and even biomass among others (legarda 2008). electric distribution utilities (dus) were required by law to make the renewable energy systems from its qualified subscribers to be connected to their power system (legarda 2008). even with the re law passed by the government, the expectation that it will progress dramatically was not realized and a lot of issues with regards to the re implementations are still being clarified (liss 2013). the rules enabling the interconnections of the solar pv and other re systems were finally approved only on may 2013 (ducut 2013). many electric dus in the country have yet to integrate solar pv systems to their own power systems, since, one of the major issues that confront these electric dus is the uncertainty and unknown implications of the re systems integration . the electric du in the province of agusan del norte has yet to determine the implications of the solar pv systems integration (damiel, 2014). while the philippines is lagging behind in the analysis and investigation of pv grid integration, studies on the potential effects of the high pv systems integration were carried out extensively in different locations around the world (tselepis and neris, 2006, katiraei et. al. 2007, rabbee et. al. 2013, jimenez et. al. 2006, byrne et. al. 2016, elliot et. al. 2014, mather et. al. 2016, thoma et. al. 2004). this study, therefore, was carried out to assess & analyze and determine the potential effect of the residential solar pv system integration with high combined capacities to the energy demand and load management of the electric du in the province of agusan del norte. 2. methodology 2.1 energy consumption profile to obtain a realistic effect of the high solar pv system penetration coming from residential owners, data were needed on the energy consumption profile from the local electric du. consultation meetings were made with some members of the management personnel of the local electric du to obtain the needed information such that the potential effect of the solar pv systems integration on the utility’s energy demand (or load) profile can be analyzed. amongst the critical information sought were: 1. energy consumption per subscriber classification and 2. their actual average energy demand load profile showing their base load, intermediate and peak loads. the energy consumption per subscriber classification will determine what percent of the contribution residential users provide with respect to the total energy demand. the data obtained were summarized and displayed with the provision of a table. also, the data obtained on int. journal of renewable energy development 5 (3) 2016: 179-185 p a g e | 181 © ijred – issn: 2252-4940, october 15th 2016, all rights reserved the actual average energy demand per hour and its profile was plotted from 0-hr to the 24th hour. 2.2 analysis on high solar pv capacities in this study, it was assumed that there will be combined capacities of 10 mega watt-peak (mwp) to 40 mwp of total solar pv systems coming from residential owners, at a 10 mw interval. the selection of the 40 mwp as the maximum total solar pv system capacity is based on the initial information that the usual or average hourly demand from subscribers is at 40 mwp. however, it is not clear what particular hour of the day and for how long the 40 mw demand does exist. meanwhile, the 10 mwp interval will provide enough differences to distinguish the effect of one capacity to the next capacity. the consideration of each capacity i.e. a 10 mwp solar pv system, is assumed to have its peak at 12 nn where this is the hottest hour of the day in the country. at this hour, it is assumed that the solar pv system will harvest 100% of the 10 mw capacities. the solar pv system will begin to generate electricity from 6am, which is the usual time of sunrise in the country. the system stops harvesting at 6pm when the sun is no longer available. it was also assumed that the total generated energy for a day of any of the four defined solar pv system capacities is defined below: teh = spvsc (mw) x 6 (hr) (1) where:  teh is the total energy harvested (mw-hr),  spvsc is the solar pv system capacity. from the given formula in (1), it can be computed that a 10 mwp solar pv system capacity will have a 60 mwhr total energy harvested in a given day. on the other hand, the 20 mwp, 30 mwp and 40 mwp systems will have total generated outputs of 120 mw-hr, 180 mw-hr and 240 mw-hr respectively. using the microsoft excel tools, graphs were generated over the 24-hour period reflecting the energy demand from the subscribers, the different solar pv system capacity from all subscribers and the net load profile. the net load profile reflects the net amount of energy from the subscribers’ total demand and the contribution of the solar pv systems. the generated graphs were then analyzed as to how much of the peak and base load demand especially during peak hours were curtailed. the information obtained was then used to analyze the potential capacity reduced from diesel and coal-fired power plant. consultations were also made with the management of the local electric du and their technical personnel on their feedback on the different generated outputs from the effect of the 10 mwp to 40 mwp combined solar pv system capacity integration. 3. results and discussions 3.1 data collected from the electric du data on the hourly average consumption from all subscribers from the electric du were obtained and can be observed from table 1. table 1 actual average hourly energy use or demand in agusan del norte hour energy demand (in mw) equivalent time of the day 1 31.41 1 am 2 30.36 2 am 3 29.23 3 am 4 28.65 4 am 5 28.86 5 am 6 30.45 6 am 7 28.89 7 am 8 28.53 8 am 9 35.29 9 am 10 40.50 10 am 11 44.44 11 am 12 46.09 12 nn 13 44.72 1 pm 14 47.96 2 pm 15 48.00 3 pm 16 46.50 4 pm 17 42.59 5 pm 18 46.36 6 pm 19 50.19 7 pm 20 48.20 8 pm 21 46.93 9 pm 22 44.05 10 pm 23 39.40 11 pm 24 35.07 12 mn source: electric distribution utility in agusan del norte province it can be noted in table 1 the highest energy consumed by all subscribers is at 19th hour or at 7 p.m. the peak energy consumption on the average was at 50.19 mw on this given hour, while the least energy consumption on the average was at seven to eight in the morning. based from the table, it was determined that indeed, the average energy consumption per hour by all subscribers is at 39.27 mw. it was confirmed by the management of the electric du that energy consumption begins to increase from 8am onwards due to several establishments that are opening and are using air-conditioning (ac) system. it was also confirmed that around seven in the evening, most if not thousands of residential owners are back at home and uses ac systems along with their television citation: dellosa, j. (2016), potential effect and analysis of high residential solar photovoltaic (pv) systems penetration to an electric distribution utility (du). int. journal of renewable energy development, 5(3), 179-185, doi: 10.14710/ijred.5.3.179-185 p a g e | 182 © ijred – issn: 2252-4940, october 15th 2016, all rights reserved sets and other electrical systems, hence the high energy demand during these hours. fig. 1 below is the load profile plotted using the microsoft excel software obtained from the electric du based from the numbers as provided in table 1. fig. 1 average energy consumption of all subscribers in the province of agusan del norte philippines. shown in fig. 1 also is the base load of the electric du at 30 mw. it is also shown the peak load at 50 mw. based on the discussion with the management of the electric du, the base load requirements are being supplied primarily by both geothermal and hydroelectric power plants, all renewable energy sources. however, the intermediate and peak load demands were supplied by coal-fired and diesel power plants based upon the discussion with the management of the electric du. this means that whenever the demand from all subscribers increases beyond the 30 mw, the energy requirement will be supplied by nonrenewable energy sources such as the diesel and coalfired power plants. this means that beyond the base load supply of the electric du, the use of coal-fired and diesel power plants are prevailing. fig. 2 shows the contribution of the residential to the total annual energy demand from all subscriber type in agusan del norte in one year. the residential consumption accounts for 44% and is the highest among all categories. this simply means that residential owners can play a very significant role in putting renewable energy sources to the grid and prevent further reliance of energy sources which are from diesel and coal-fired power plants which are served during the intermediate and peak hours. demand from industrial and the rest of the subscribers having a total of 42% share can be augmented as well, if industrial companies in the province such as plywood factories will invest in solar pv system or other forms of renewable energy. energy consumption from the commercial establishments having a 14% share are equally capable of similar reductions with re sources deployments. 119,561,937 44% 36,840,799 14% 114,601,018 42% annual energy consumption (in kwhr and %) residential commercial industrial and others fig. 2 actual energy consumption of the different type of subscribers in agusan del norte philippines in one year. 3.2 impact of high solar pv penetration figure 3 shows an output of a 30 mwp solar pv system. it is assumed that the system begins harvesting at 6 am and ends at 6 pm and the peak reaches 30 mw at noon time for all combined residential systems with 100% system efficiency. the total harvest was approximated at 180 mwhr in a day using equation 1. 0.00 10.00 20.00 30.00 40.00 50.00 60.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 m e g a w a t t ( m w ) hour simulated solar pv system output fig. 3 generated total output of 30 mwp solar pv systems. fig. 4 shows the generated output of the four assumed pv system sizes from 10 mwp to 40 mwp systems, which is the subject of this study. shown are the peak outputs at 100% efficiency. each of these graphs were then used to determine the impact of each capacity to the existing and actual energy demand or load profile of the local electric du as shown in fig. 1. 0.00 20.00 40.00 60.00 1 3 5 7 9 11 13 15 17 19 21 23 m e g a w a tt ( m w ) hour simulated solar pv systems output profile 10 mwp profile 20 mwp profile 30 mwp profile 40 mwp profile fig. 4 generated output of 10, 20, 30 and 40 mwp solar pv systems. base load intermediate and peak load int. journal of renewable energy development 5 (3) 2016: 179-185 p a g e | 183 © ijred – issn: 2252-4940, october 15th 2016, all rights reserved fig. 5 shows what it would look like when a 10 mwp combined capacity of solar pv system from residential owners affect the existing load profile of the local electric du. a 10 mwp capacity would stave off partially the intermediate and peak load (orange graph) with the resulting net output in the yellow graph. the impact of this combined capacity can be viewed as moderately significant. however, discussions with representatives from the local electric du viewed this as favorable to their operations as there is a gradual decrease of load from 6 am to 8 am and a gradual increase from 8 am onwards. electric du also preferred the absence of steep changes of load, downward or upward changes, from one hour to the next hour. this is due to the cost implications and the supply from grid suppliers to the demand of the subscribers according to the management of the local electric du. from equation 1, the total potential load reduction during the day with a 10 mw system is at 60 mw-hrs. this would reduce the demand at 2.5 mw per hour on the average (60 mw-hr/24 hours in a day). 0.00 10.00 20.00 30.00 40.00 50.00 60.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 m e g a w a t t ( m w ) hour net load with 10 mwp solar pv output actual load solar output (mw) net load profile fig. 5 generated total output of 20 mwp solar pv systems 0.00 10.00 20.00 30.00 40.00 50.00 60.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 m e g a w a t t ( m w ) hour net load with 20 mwp solar pv output solar output actual load net load profile fig. 6 generated total output of 20 mwp solar pv systems fig. 6 shows the potential output of a 20 mwp combined capacity integrated to the actual energy load profile of the electric du. it can be observed from the graph that a huge quantity of the intermediate and peak load was reduced from 8 am towards 4 pm. this would translate to the reduction of demand from coal-fired and diesel power plants to supply beyond the energy requirement beyond the base load during the day. the effect of the 20 mwp combined capacity would have a stable load period from 8 am towards 1 pm and slowly ramps up towards the intermediate and peak load from 1 pm and onwards. according to representatives from the local electric du, the resulting load profile with 20 mw is likewise acceptable and favorable with regards to the energy demand and load management. the resulting figure did not show steep changes of energy demand. with a 20 mwp capacity, the total potential load reduction during the day would be around 120 mw-hrs as obtained using equation 1. this would also translate to a reduction of demand from its residential subscribers at 5 mw per hour on the average. fig. 7 and fig. 8 show generated outputs with 30 mwp and 40 mwp solar pv system capacities respectively. 0.00 10.00 20.00 30.00 40.00 50.00 60.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 m e g a w a t t ( m w ) hour net load with 30 mwp solar pv output actual load solar output net load profile fig. 7 generated total output of 30 mwp solar pv systems 0.00 10.00 20.00 30.00 40.00 50.00 60.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 m e g a w a t t ( m w ) hour net load with 40 mwp solar pv output actual load solar output net load profile fig. 8 generated total output of 40 mwp solar pv systems citation: dellosa, j. (2016), potential effect and analysis of high residential solar photovoltaic (pv) systems penetration to an electric distribution utility (du). int. journal of renewable energy development, 5(3), 179-185, doi: 10.14710/ijred.5.3.179-185 p a g e | 184 © ijred – issn: 2252-4940, october 15th 2016, all rights reserved unlike the two previous generated outputs, the 30 mwp capacity staved off significant portions of the intermediate and peak load but at the same time clipped the demand at the base load during the day. the same thing is evident in fig. 4 where the base load was curtailed and resulted to a net demand of 5 to 8 mw between the hours of 10 am to 1 pm. while the results shown in fig. 7 and fig. 8 should be favorable having renewable sources supplying the intermediate and peak loads including the base load, however, for the local electric du this is not their preferred situation. the presence of steep changes of demand from 6am to 10pm and then from 1pm onwards revealed undesirable load management scenario according to them. for the 30 mwp and 40 mwp systems, harvests of 180 mw-hr and 240 mw-hr are expected throughout the day, using the formula in equation 1. these are based on the assumptions that all the systems are operating at 100% efficiency. it was also assumed at the same time that the weather during the day is favorable with the sun available from sunrise to sunset. these assumptions can be considered appropriate to measure the worst-case scenario how combined capacities impact the energy demand profile of the local electric du. based from the discussions with the management of the local electric du, the preferred and acceptable load profile is up to a combined capacity of 20 mwp from all residential subscribers in terms of non-complex energy demand and load management. the presence of steep changes from the 30 mwp and 40 mwp systems discouraged the electric du because of its potential difficulties in energy sourcing from its suppliers. however, if the combined capacities will indeed exceed 20 mwp in the future, the local electric du will prepare proactive actions to facilitate the renewable energy sources integration to their grid. 7. conclusion the objective of this paper was to investigate the potential effect of the high solar pv system penetration to the local electric du in the province of agusan del norte through the conduct of analysis using actual data on the energy demand and load profile of the electric du. it was determined from this study that a combined capacity of 20 mwp from all residential subscribers was the preferred choice of the local electric du with regards to how the energy demand and sourcing is concern. the absence of the steep changes from this load profile was the preferred advantage. however, the electric du confirmed that proactive actions will be planned to accommodate additional renewable energy sources to be integrated in the grid if it exceeds the combined capacity of 20 mwp. acknowledgments i would like to thank the management of the agusan del norte electric cooperative, inc. (aneco), the local electric du in the province of agusan del norte, for generously providing the needed actual data used during the analysis of this study. special thanks to arch. horacio b. santos, general manager, engr. darwin damiel, head of the technical services division, mr. rhenie tagoloan and mr. melquisedic bompat (accounting department) who are all from aneco. references almendras, j. (2012) towards a sustainable and competitive power sector. department of energy, philippine government. asian development bank (2009) the economics of climate change in southeast asia: a regional review. byrne, r.h., concepcion, r., neely, j., wilches-bernal, f., elliot, r., lavrova, o.& quiroz, j. (2016) small signal stability of the western north american power grid with high penetrations of renewable generation. proceedings of the photovoltaic specialists conference. oregon, usa. damiel, d. (2014) discussions on net-metering. agusan del norte electric cooperative, inc. (aneco) dellosa, j. t. (2015). financial payback of solar pv systems and analysis of the potential impact of net-metering in butuan city, philippines. in environment and electrical engineering (eeeic), 2015 ieee 15th international conference on (pp. 1453-1458). department of energy (2008) philippine energy situation. energy situationer. ducut, z. (2013) rules enabling the net-metering for renewable energy. energy regulations commission (erc), philippines. elliot, r., byrne, r., ellis, a. & grant, l. (2014) impact of increased photovoltaic generation on inter-area oscillations in the western north american power system. proceedings of the 2014 ieee pes general meeting. national harbor, md, usa. green peace international (2005) the environmental impacts of coal. green peace international (2014). greenpeace statement ahead of the state of the nation address: pnoy’s legacy: more coal plants than those built by all his predecessors combined? inter-government panel on climate change (2013) climate change 2013: the physical science basis. inter-government panel on climate change (2014) climate change 2014 synthesis report, summary for policy makers. international energy agency (2013) south east asia energy outlook, special report. jimenez, h., calleja, h., gonzales, r., huacuz, j. & lagunas, j. (2006) the impact of photovoltaic systems on distribution transformer: a case study. journal of energy conversion and management, 47, 311-321. katiraei, f., mauch, k. & dignard-bailey, l. (2007) integration of photovoltaic power systems in high-penetration clusters for distribution networks and mini-grids. international journal of distributed energy resources, 3(3). kolhe, m., khot, p. (2015) impact of the coal industry on environment. international journal of advanced research in computer science and management studies, 3(1), 66-73. köppinger, p. (2014) climate report 2014 energy security and climate change worldwide. latour, m., fontaine, b., masson, g., rekinger, m., theologitis, i. & papoutsi, m. (2013) global market outlook for photovoltaics 2013-2017. epia, 5-6. legarda, l. (2008) republic act 9513: renewable energy act of 2008. liss, b. (2013) accompanying the philippines on the road towards sustainable energy supply. gesellschaft fur internationale zusammenarbeit (giz). mather, b., cheng, d., seguin, r., hambrick, j. & broadwater, r. (2016) photovoltaic (pv) impact assessment for very high penetration levels. ieee journal of photovoltaics, 295-300. int. journal of renewable energy development 5 (3) 2016: 179-185 p a g e | 185 © ijred – issn: 2252-4940, october 15th 2016, all rights reserved munnik, v., hochmann, g., hlabane, m. & law s. (2010) the social and environmental consequences of coal mining in south africa. environmental monitoring group, cape town, south africa and both ends, amsterdam, the netherlands, january 2010, http://www.bothends.org/uploaded_files/uploadlibraryitem/1ca se_study_south_africa_updated.pdf panzer, c., balabanov, t., pabon, v., ulreich, s., krämer, l., chaturvedi, p., diodato, a., khatib, h., gulczynski, d., miga-papadopol, p., day, b., camacho, m., auchariyamet, s., eakponpisan, p., shastri, a., echinope, m. and olson, s. (2016) variable renewables integration in electricity systems: how to get it right, world energy perspectives, renewables integration, world energy council. https://www.worldenergy.org/wpcontent/uploads/2016/09/variable-renewables-integration-inelectricity-systems-2016-how-to-get-it-right-_-full-report-1.pdf. rabbee, f., wadud, a.m.a, zaman, m.t. and rahman, m.r (2013) renewable energy: an ideal solution of energy crisis and economic development in bangladesh. global journal of researches in engineering electrical and electronics engineering, volume 13(5), 19-27. thoma, m., laukamp, h., meyer, t. & erge, t. (2004) impact of a large capacity of distributed pv production on the low voltage grid. proceedings of the 19th european photovoltaic solar energy conference. paris, france. tselepis, s. & neris, a. (2006) impact of increasing penetration of pv and wind generation on the dynamic behaviour of the autonomous grid of the island of kythnos, greece. 3rd european conference on pv hybrids and mini-grids, aix en provence. united states energy information administration (2013) international energy outlook 2013. united states environmental protection agency (2012) the emissions & generation resource integrated database for 2012 (egrid2012) technical support document. zwickel, t., edenhofer, o., pichs-madruga, r., sokona, y., seyboth, k., matschoss, p., kadner, s., eickemeier, p., hansen, g., von stechow & c., schlömer, s. (2012) special report on renewable energy sources and climate change mitigation. international journal of renewable energy development international journal of renewable energy development 11(1) 2022: 103-110 |103 ijred-issn: 2252-4940.copyright © 2022. the authors. published by cbiore contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: https://ijred.undip.ac.id efficiency improvement of ground-mounted solar power generation in agrivoltaic system by cultivation of bok choy (brassica rapa subsp. chinensis l.) under the panels manoch kumpanalaisatit, worajit setthapun, hathaitip sintuya, surachai narrat jansri* asian development college for community economy and technology (adicet), chiang mai rajabhat university, mae rim district, chiang mai province, 50180, thailand abstract. an agrivoltaic system is a combination of solar power generation and crop production that has the potential to increase the value of land. the system was carried out at a 25-kw photovoltaic (pv) power plant located at the asian development college for community economy and technology (adicet), chiang mai rajabhat university, thailand. the growth and yield of bok choy (brassica rapa subsp. chinensis l.) and the solar power output were investigated and compared with the control. moreover, the efficiency of the agrivoltaic system was evaluated. the results indicated that the average intensity of solar radiation of 569 w/m2 was obtained. the highest power generation was recorded in the pv with crop production of 2.28 kw. furthermore, the control plot of crop production at 35 days provided higher growth than bok choy plots under solar panels of 2.1 cm in plant height, 6 in leaf number, 2.2 cm in leaf length and 0.2 cm in leaf wide. high-yield of bok choy was also obtained in the control plot of 17.31 kg. although the yield of bok choy is extremely low, possibly because of light intensity, crop cultivation under solar panels could reduce the module temperature to less than the pv control of 0.18 °c, resulting in increased voltage and power generation by around 0.09 %. therefore, an agrivoltaic system is another option for increasing revenue and land equivalent ratio in solar power plants focusing only on electricity generation. however, suitable crops for the space under pv panels should be investigated further. keywords: agrivoltaic system, ground-mounted solar power plants, solar power generation, solar power output, bok choy. article history: received: 1st sept 2021; revised: 7th oct 2021; accepted: 12th oct 2021; available online: 20th oct 2021 how to cite this article: kumpanalaisatit, m., setthapun, w., sintuya, h., jansri, s.n. (2022) efficiency improvement of ground-mounted solar power generation in agrivoltaic system by cultivation of bok choy (brassica rapa subsp. chinensis l.) under the panels. int. j. renew. en. dev., 11(1), 103-110. https://doi.org/10.14710/ijred.2022.41116 1. introduction presently, solar energy can be converted into thermal or electrical energy. solar energy is the cleanest and most abundant renewable energy source available at present, which can be converted into electricity directly by using photovoltaic (pv) systems (missoum et al., 2021). such systems usually consist of photovoltaic modules, batteries, charge controllers, inverters, load controllers, circuit breakers and wiring (wasfi 2011). in 2018, the global capacity of photovoltaic power plants was over 500 gw (jäger-waldau 2018), most of which comprised groundmounted photovoltaic power plants with a capacity of 471 gw. moreover, the remainder is separated by 27.9 gw of solar roof tops (europe 2018) and 1.1 gw of solar floating (gamarra et al. 2019). the total land-use requirement for ground-mounted photovoltaic power plants was 1,448,612 ha (ong et al. 2013). in line with the global trend, ground-mounted photovoltaic power plants are still favored in thailand. solar power plant installations cover an area of * corresponding author: surachai_nar@g.cmru.ac.th approximately 9,020 ha, which can presently generate total solar power at 2,819 mw per day (chimres et al. 2016). by 2036, the thai government plans to increase the number of solar farms to generate increased capacity at more than 6,000 mw (department of renewable energy development and energy efficiency 2015). the expansion of solar power generation will inevitably require more land area, which may affect not only the land use competition, but also food security in the future. while solar power plants have been promoted in thailand and other countries, the areas under the solar panels have not been fully utilized because they focus only on solar power generation. therefore, a combination of solar power generation and food production to increase the land benefit, also known as an agrivoltaic system, has been studied for its potential advantages (valle et al. 2017; dupraz et al. 2011). in addition, annual income could be increased by using agrivoltaic systems (li et al. 2017; mohammadi et al., 2017). planting crops for agrivoltaic systems involves co-production between the solar power generation and crop production in a single space at the research article citation: kumpanalaisatit, m., setthapun, w., sintuya, h., jansri, s.n.. (2022). efficiency improvement of ground-mounted solar power generation in agrivoltaic system by cultivation of bok choy (brassica rapa subsp. chinensis l.) under the panels. int. journal of renewable energy development, 11(1), 103-110, doi: 10.14710/ijred.2022.41116 104 | ijred-issn: 2252-4940.copyright © 2022. the authors. published by cbiore same time. shade-tolerant plants have been suggested for planting under solar panels (beck et al. 2012). a study by malu et al. (2017) indicated that grape farms deploying the agrivoltaic systems may increase the economic value by more than 15 times compared to conventional farming. moreover, trypanagnostopoulos et al. (2017) found that the plant growing results under shading effect were satisfactory as they were at the same level with a reference greenhouse without a pv-covered roof. there are presently 2 guidelines on employment of agrivoltaic systems which are land utilization under pv panels of fixed pv systems without agricultural plans and planned pv system installation with agricultural plans. it was found that fixed pv system installation for electricity generation as the only purpose was with the environment under pv panels that did not sufficiently suit cultivation. hence, the number of these studies on land utilization under pv panels of pv systems are still very few studies. consequently, in this work, the efficiency improvement of ground-mounted solar power generation through plant cultivation under solar panel were investigated. 2. materials and methods 2.1 ground-mounted pv power plant the study of the agrivoltaic system was carried out at a 25-kw pv power plant located at the asian development college for community economy and technology (adicet), chiang mai rajabhat university, thailand. this pv power plant was designed and installed in the purpose of generating electricity only. the plant consists of 5 pv arrays in a 25 kwp ac micro grid-connected pv system as shown in figure 1. the system is connected to a distribution system in the chiang mai rajabhat university grid. there are 225 amorphous pv modules installed at 2.0 and 0.8 m from the ground on the north and the south direction, respectively, with the total pv panels surface area of 352.15 m2. the modules were installed with a solar charge controller, 40 deep-cycle batteries and a grid-interactive inverter. the system in order to distribute the power to chiang mai world green city (cwgc), consisting of six households, an office, a restaurant, a coffee shop and a convenience store. 2.2 crop plants the condition under the ac micro grid-connected pv system is not suitable for general crop production, however shade-tolerant crops could grow. therefore, bok choy (brassica rapa subsp. chinensis l.) was selected for investigated in this study. the reason was that it is a biennial plant that can grow best in partial shade or receive 3-5 h of sun light exposure each day (chintu, 2021). moreover, it can tolerate temperatures as high as 35°c and as low as -3°c and prefers slightly acidic (ph 5.5 to 7.0) sandy soil rich in nutrients (liu et al. 2019). all plants at the control plot of the experiment were under direct sun light by around 10 h (07.00 a.m. – 05.00 p.m.) while the plants under solar panels by around 20 % were under direct sun light for 3 h in the morning (07.00 a.m. – 10.00 a.m.) fig. 1 25-kw ac pv farm at adicet international journal of renewable energy development 11(1) 2022: 103-110 |105 ijred-issn: 2252-4940.copyright © 2022. the authors. published by cbiore 2.3 soil preparation for crop production under the pv panels the structure of the soil under the solar panels is sandy loam mixed with rock and is of low nutrient value. therefore, soil under solar panels was amended by organic matter. the soil under 4 pv panels and a control (outside the pv panels) was prepared and ploughed. from the photo in figure 2, the plot should be way around 1 x 15 cm (15 m2). then, it was mixed with an organic compost (pork manure mixed with rice husk at a ratio of 7:3) of 100 kg per bed and released for 30 days. the nutrition of soil was analyzed with commercial soil test kit. results indicated that soil had a slightly acidic ph of 6.2. total nitrogen, phosphorus and potassium in soil were report to be 20, 20 and 400 ppm, respectively. after that, bok choy seedlings, 25 days after germination, were transplanted into each plot at 15 cm between the plants and rows, providing a density of 132 plants/plot. they were watered twice a day (07.00 a.m. and 05.00 p.m.) with 500 l of water per day. 2.4 monitoring and evaluation 2.4.1 environment and electricity the solar radiation (08.00 a.m. – 05.00 p.m.), temperature of solar panels, voltage and the current were monitored and recorded during the same day of crop monitoring. moreover, the surface area of the solar panels was also measured. solar power generation was then evaluated and compared with control, as in equation 1 (ekpenyong et al. 2013). (1) 2.4.2 vegetative growth and yield of crops various vegetative parameters including the height of plants, the number of leaves, the size of leaves and the weight of plants were monitored. the plant height was measured from the soil line to the shoot apex. simultaneously, the leaf number per plant was counted followed by measuring the length and width of the leaves. all the various vegetative parameters were recorded every 7 days, starting from the 7th day after transplanting to the harvest date. at 35 days after transplanting, five plants per plot were harvested, weighed, and recorded. after that, they were dried at 60°c with a hot air dryer for 48 h or to constant weight. subsequently, the dry weight of samples was determined for percentage of biomass (kongake et al. 2014), as in equation 2. finally, the total crop yield was also harvested and weighed. the experimental data was statistically analyzed using microsoft excel. analysis of variance (anova) was carried out on every measured parameter to determine the significance of differences between the means of each plot. means for each parameter were separated by duncan’s multiple range test (dmrt) at p ≤ 0.05 (kongake et al. 2015). (2) 3. results and discussion 3.1 environment and electricity the planting period was conducted from july10th to august 14th, 2018, which was a monitoring period for the solar radiation, temperature, voltage and current. meanwhile, the surface area of the solar panels was measured from the top side of panels that faced to the sunlight. 3.1.1 solar radiation the solar radiation indicated that the average solar radiation intensity of 569 w/m2 was obtained. from the graph in figure 3, the highest intensity recorded is below 1,000 w/m2 during daytime from 11.30 a.m. – 12.00 p.m. 3.1.2 temperature the temperature under the solar panels, including with the control, showed that their temperatures were higher than the ambient temperature as in figure 4. however, the average temperature of all solar panels with crop production was lower than the control around by 0.18°c. the result was consistent with barron-gafford et al. (2019). fig. 2 experimental layout and plots for crop production citation: kumpanalaisatit, m., setthapun, w., sintuya, h., jansri, s.n.. (2022). efficiency improvement of ground-mounted solar power generation in agrivoltaic system by cultivation of bok choy (brassica rapa subsp. chinensis l.) under the panels. int. journal of renewable energy development, 11(1), 103-110, doi: 10.14710/ijred.2022.41116 106 | ijred-issn: 2252-4940.copyright © 2022. the authors. published by cbiore 3.1.3 solar power generation pv no.1 and 3 could generate power at around 2.28 and 2.12 kw, respectively, which was higher than the control. the control could generate power of approximately 2.06 kw, which was similar to pv no.2. the lowest power generation was recorded in pv no.4 (~1.87 kw), as in figure 5. the experimental design is not homogeneous since pv no. 4 was covered the shading from trees and a mound of dirt near the array after 03.00 p.m. therefore, pv no.4 generated less electricity than the control. according to the suggestion of sathyanarayana et al. (2015), the area of the shade had a significant effect on the i-v characteristics as well as power output. fig. 3 solar radiation intensities fig. 4 solar panel and ambient temperature fig. 5 solar power generation international journal of renewable energy development 11(1) 2022: 103-110 |107 ijred-issn: 2252-4940.copyright © 2022. the authors. published by cbiore table 1 growth of plants days vegetative growth plot f test cv (%) 1 2 3 4 control 7th the height of plant (cm) 14.3a 13.1ab 12.4b 13.7ab 13.31b * 6.8 the leaf number 6 5 5 6 5 ns 11.7 the left width (cm) 3.9a 3.5ab 3.3ab 3.7a 2.9b * 15.2 the leaf length (cm) 10.4a 9.3ab 8.7bc 9.0ab 7.2c ** 12.8 14th the height of plant (cm) 16.7 16.1 15.4 16.2 16.8 ns 8.1 the leaf number 8 7 8 8 9 ns 8.1 the left width (cm) 4.4 3.9 4.3 4.7 5.0 ns 17.4 the leaf length (cm) 12.1a 10.0b 11.3a 12.0a 11.5a ** 8.0 21st the height of plant (cm) 18.7a 17.8ab 16.1b 17.4ab 19.3a * 7.9 the leaf number 9ab 8b 10a 9ab 10a * 12.1 the left width (cm) 4.6b 4.1b 4.8b 4.8b 5.8a * 15.2 the leaf length (cm) 13.2a 10.3b 11.7ab 12.7a 13.0a ** 9.2 28th the height of plant (cm) 20.6 20.2 17.1 20.1 21.4 ns 13.6 the leaf number 11b 10b 12ab 11ab 14a * 13.6 the left width (cm) 5.3b 4.4b 5.3b 5.4b 7.2a ** 19.1 the leaf length (cm) 13.8ab 11.9b 12.1b 14.3ab 15.9a * 14.1 35th the height of plant (cm) 21.5 21.7 20.8 21.6 23.5 ns 10.7 the leaf number 15b 11c 14bc 13bc 19a ** 15.7 the left width (cm) 5.6b 4.8b 5.9ab 6.8ab 8.0a * 24.1 the leaf length (cm) 13.9b 12.3b 13.7b 14.5b 17.9a ** 14.5 note: a, b and c-significantly different at p ≤ 0.05 by duncan's multiple range test, *-significant at the 0.05 level, **-significant at the 0.01 level, ns-non significant, cv-coefficient of variation 3.2 vegetative growth and yield similar to the date of monitoring and evaluation for environment and power generation, the height of plants, leaf number, leaf size and weight of plants were analyzed. 3.2.1 height of plants the mean height of bok choy plants was not influenced by the conditions under pv panels. the results as in table 1. show that the average height of bok choy in the control plot after 14 days until harvesting was from 16.80 to 23.5 cm, respectively, which was not significant difference to all the crops in the plots under pv panels. moreover, the stems of the bok choy under the solar panels were thin and smaller than the control plot as in figure 6. the reason was that the bok choy planted under the solar panels received less light intensity. the result was consistent with the experiment by nguyen et al. (2019), in which spinach (spinacia oleracea) farming under lower light intensity resulted in a lower height than suitability. 3.2.2 leaf numbers the shading of the solar panels could reduce solar intensity of 1,200 µmol/m/s (rittiram et al. 2019) for shining on the area under the solar panels at lower than 74 µmol/m/s, which was less than the requirement of bok choy by around 65 %. consequently, the bok choy in the plot under pv panels gave lower leaf numbers than the control as shown in the results in table 1. according to the citation: kumpanalaisatit, m., setthapun, w., sintuya, h., jansri, s.n.. (2022). efficiency improvement of ground-mounted solar power generation in agrivoltaic system by cultivation of bok choy (brassica rapa subsp. chinensis l.) under the panels. int. journal of renewable energy development, 11(1), 103-110, doi: 10.14710/ijred.2022.41116 108 | ijred-issn: 2252-4940.copyright © 2022. the authors. published by cbiore results of nguyen et al. (2019), the low leaf number of spinach plants was obtained under low light intensity. 3.2.3 leaf sizes the length and width of the leaves were found to be significantly influenced at 35 days. the longest and widest leaves were recorded in the control as in table 1. it was observed that the size of the leaves in the control was larger than all bok choy plots under pv panels because the control received sufficient light intensity. the result was consistent with petchthai et al. (2017), who studied the effect of light intensity and light-exposure duration on the growth and quality of lettuce. their experiment found that high light intensity provided the size of the canopy wider and greater leaf size than the lower one. 3.2.4 weight of plants there was significant effect of different plots under pv panels and the control on fresh and dry weight of bok choy. similar to the reasons above for vegetative parameters, the control provided the highest shoots and fresh root weight, shoots and root dry weight, and total fresh weight and dry weight as in table 2. moreover, the highest total yield per plot was recorded in the control, consistent with the experiment by weiguo et al. (2012). they investigated the effects of different light intensities on anti-oxidative enzyme activity, quality and biomass in lettuce. the results showed that high light intensity could affect fresh weight better than low light intensity because the light intensity directly affected the rate of photosynthesis. fig. 6 bok choy plot under solar panels and the control plot table 2 fresh and dry weights of bok choy at harvest date plot no. fresh weight dry weight total total yield (kg) shoots (g) roots (g) shoots (g) roots (g) shoots (g) roots (g) 1 51.8b 5.6b 1.9b 0.4 57.3b 2.2b 2.31 2 26.7b 2.8b 5.6b 3.3 29.5b 8.8ab 1.05 3 49.0b 6.1b 2.4b 0.4 55.1b 2.9b 1.50 4 36.3b 3.8b 1.9b 0.4 40.1b 2.4b 1.16 control 233.9a 23.3a 16.7a 2.82 257.2a 19.5a 17.31 f test ** ** ** ns ** * cv (%) 8.1 8.6 8.1 8.6 12.7 12.7 international journal of renewable energy development 11(1) 2022: 103-110 |109 ijred-issn: 2252-4940.copyright © 2022. the authors. published by cbiore table 3 efficiency of the agri-voltaic system power generation (kw) crop yield (kg) control pv eff measurement 2.06ab pv no.1 2.28a 2.31 pv no.2 2.05ab 1.05 pv no.3 2.12ab 1.50 pv no. 4 1.87c 1.16 control pot 17.31 f test ** cv (%) 5.04 table 4 correlations between power generation and crop yield variables power generation (kw) crop yield (kg) power generation (kw) 1.000 0.850* crop yield (kg) 0.850* 1.000 note: *-p < 0.05 3.3 efficiency of the agrivoltaic system the combination between pv power generation and crop production or agrivoltaic systems as in table 3 indicated that the power generation were found to be significantly influenced. the highest power generation was recorded in pv no.1 compared to the control. the pv no.3 could generate power higher than the control but there was nonsignificant. the correlation of pv power generation and crop production was significant correlated as in table 4 which the experimental result showed that the highest power generation was generated with the highest crop production under the pv panel. the result indicated that planting crops under the pv panels could reduce the temperature under the pv panels. on the other hand, the control could generate power similar to pv no.2 and was significantly higher than pv no.4 due to the geographical characteristic problem. although the effect from light intensity caused reduced yield of bok choy under solar panels, steam caused by evaporation of water from the soil surface and by transpiration of plant under the solar panels could reduce module temperature. the decrease of pv panel temperature could promote the voltage and increase power generation by around 0.09 %, which was similar to the results of boonsri et al. (2017), peng et al. (2017) and jatoi et al. (2021). 6. conclusion to the point, solar power generation with crop production could promote power generation, but the shading of solar panels interfered with crop production. consequently, the use of agrivoltaic systems should be encouraged. not only because the crop production under the pv panels could promote power generation, but crop cultivation under solar panels is another option to increase revenue and land equivalent ratio in solar power plants which designed and installed for generating electricity only. for solving the problems mentioned above, shade-tolerant crops will be considered for planting under the solar panels. they consist of leafy vegetables (lettuce, celery, spinach and spring onion), root and tuber vegetables (ginger, galangal, sweet potato and carrots) and solanaceous crop (chili). however, in assessing the physical properties of plants, root vegetable cultivation under solar panels can damage to the structure of ground-mounted photovoltaic power plants. nevertheless, leafy vegetables and solanaceous crop as well as crop production conditions should be investigated further. acknowledgments we would like to express our thanks to the asian development college for community economy and technology (adicet), chiang mai rajabhat university, thailand; energy policy and planning office, ministry of energy, thailand, office of naval research, united states and institute of research and development chiang mai rajabhat university, thailand for supporting this research. author contributions: kumpanalaisatit, m., jansri, s.n.: conceptualization, methodology, formal analysis, writing— original draft, jansri, s.n.: supervision, resources, project administration, jansri, s.n., kumpanalaisatit, m., setthapun, w., and sintuya, h.; writing—review and editing, project administration, validation. all authors have read and agreed to the published version of the manuscript. funding: this research was funded by policy and planning office, ministry of energy, thailand, office of naval research, united states and institute of research and development chiang mai rajabhat university, thailand. conflicts of interest: the authors declare no conflict of interest. references barron-gafford, g. a., pavao-zuckerman, m. a., minor, r. l., sutter, l. f., barnett-moreno, i., blackett, d. t., & macknick, j. e. (2019). agrivoltaics provide mutual benefits across the food–energy–water nexus in drylands. nature sustainability, 2, 848-855; https://doi.org/10.1038/s41893-019-0364-5. beck, m., bopp, g., goetzberger, a., obergfell, t., reise, c., & schindele, s. (2012). combining pv and food crops to agrophotovoltaic–optimization of orientation and harvest. in 27th eupvsec, frankfurt, germany, 4096-4100, 24-28 september 2012; doi:10.4229/27theupvsec2012-5av.2.25. boonsri, s., sangsuwan, s., & hindee, i. (2017). the study temperature reduction technique of solar panels using different water-cooling system (in thai). in 4th national cenference kamphaeng phet rajabhat university, kamphaeng phet, thailand, 391-399, 22 december 2017. https://doi.org/10.1038/s41893-019-0364-5 https://www.eupvsec-proceedings.com/proceedings?paper=18297 citation: kumpanalaisatit, m., setthapun, w., sintuya, h., jansri, s.n.. (2022). efficiency improvement of ground-mounted solar power generation in agrivoltaic system by cultivation of bok choy (brassica rapa subsp. chinensis l.) under the panels. int. journal of renewable energy development, 11(1), 103-110, doi: 10.14710/ijred.2022.41116 110 | ijred-issn: 2252-4940.copyright © 2022. the authors. published by cbiore https://research.kpru. ac.th/sac/fileconference/10742018-05-01.pdf. accessed on 25 august 2021. chimres, n., & wongwises, s. (2016). critical review of the current status of solar energy in thailand. renewable and sustainable energy reviews, 58, 198-207; https://doi.org/ 10.1016/j.rser. 2015.11.005. chintu, d. (2021). bok choy cultivation: plant care, soil, light and water requirements, fertilizer needs, harvesting and much more. https://krishijagran.com. accessed on 25 august 2021. department of renewable energy development and energy efficiency. (2015). alternative energy development plan: aedp 2015. minister of energy, thailand. dupraz, c., marrou, h., talbot, g., dufour, l., nogier, a., & ferard, y. (2011). combining solar photovoltaic panels and food crops for optimising land use: towards new agrivoltaic schemes. renewable energy, 36, 2725-2732; https://doi.org/ 10.1016/j.renene.2011.03.005. ekpenyong, e. e., & anyasi, f. i. (2013). effect of shading on photovoltaic cell. iosr journal of electrical and electronics engineering, 8(2), 1-6. http://www.iosrjournals.org. accessed on 25 august 2021. europe, s. p. (2018). global market outlook for solar power 2018– 2022. solar power europe, brussels. gamarra, c. & ronk, j. (2019). floating solar: an emerging opportunity at the energy-water nexus. texas water journal, 10(1), 32-45; https://doi.org/10.21423/twj.v10i1.705 0. jäger-waldau, a. (2018). snapshot of photovoltaics− february 2018. epj photovoltaics, 9(6), 1-6; https://doi.org/10.1051/ epjpv/2018004. jatoi, a. r., samo, s. r., & jakhrani, a. q. (2021). performance evaluation of various photovoltaic module technologies at nawabshah pakistan. international journal of renewable energy development, 10(1), 97-103. https://doi.org/10.14710/ ijred.2021.32352 kongake, s., kulisara, t., & natthawut, h. (2014). effect of nutrient solution on growth of lettuce (lactuca sativa l.) cultivated under hydroponic system (in thai). science and technology, 22(6), 828-836. https://li01.tcithaijo.org/index. php/tstj/article/view/22433/19209. accessed on 25 august 2021. kongake, s., praneet, j., & wipaporn, s. (2015). effect of growing methods on growth and pigment concentrations of leaf lettuce (lactuca sativa var. crispa l.) (in thai). phranakhon rajabhat research journal: science and technology, 10(1), 83-95. https://li01.tci-thaijo.org/index.php/prrjscitech/articl e/view/41964. accessed on 25 august 2021. li, c., wang, h., miao, h., & ye, b. (2017). the economic and social performance of integrated photovoltaic and agricultural greenhouses systems: case study in china. applied energy, 190, 204-212; https://doi.org/10.1016/j.apenergy.2016.12.121. liu, h., & liu, g. (2019). bok choy, an asian leafy green vegetable emerging in florida. edis, 6, 5; https://doi.org/10.32473/edishs1337-2019. malu, p. r., sharma, u. s., & pearce, j. m. (2017). agrivoltaic potential on grape farms in india. sustainable energy technologies and assessments, 23, 104-110, https://doi.org/ 10.1016/j.seta.2017.08.004. missoum, m., & loukarfi, l. (2021). investigation of a solar polygeneration system for a multi-storey residential building-dynamic simulation and performance analysis. international journal of renewable energy development, 10(3), 445-458. https://doi.org/10.14710/ijred.2021.34423 mohammadi, m & yavari, g. (2017) factors affecting the use and development of solar energy in iran's agricultural sector. international journal of renewable energy development, 6(1), 45-53. http://dx.doi.org/10.14710/ijred.6.1.45-53 nguyen, t. p. d., tran, t. t. h., & nguyen, q. t. (2019). effects of light intensity on the growth, photosynthesis and leaf microstructure of hydroponic cultivated spinach (spinacia oleracea l.) under a combination of red and blue leds in house. international journal of agricultural technology, 15(1), 75-90. http://www.ijat-aatsea.com. accessed on 25 august 2021. ong, s., campbell, c., denholm, p., margolis, r., & heath, g. (2013). land-use requirements for solar power plants in the united states, national renewable energy lab. golden, co, united state. peng, z., herfatmanesh, m. r., & liu, y. (2017). cooled solar pv panels for output energy efficiency optimisation. energy conversion management, 150, 949-955, https://doi.org/ 10.1016/ j.enconman.2017.07.007. petchthai, p., & thongket, t. (2017). effect of light intensity and light-exposure duration on growth and quality of lettuce (in thai). songklanakarinjournal of plant science, 4(3), 54-59. http://natres.psu.ac.th/sjps/index.php/journal/article/view/28 0/222. accessed on 25 august 2021. rittiram, j., & tira-umphon, a. (2019). effects of light intensity on growth and yield of lettuce in plant factory system (in thai). khon kaen agriculture journal, 47(6), 1243-1250; doi:10.144 56/kaj.2019.113. sathyanarayana, p., ballal, r., sagar, p. l., & kumar, g. (2015). effect of shading on the performance of solar pv panel. energy and power, 5(1a), 1-4; doi:10.5923/c.ep.201501.01. trypanagnostopoulos, g., kavga, a., souliotis, μ., & tripanagnostopoulos, y. (2017). greenhouse performance results for roof installed photovoltaics. renewable energy, 111, 724-731; doi: 10.1016/j.renene.2017.04.066. valle, b., simonneau, t., sourd, f., pechier, p., hamard, p., frisson, t., & christophe, a. (2017). increasing the total productivity of a land by combining mobile photovoltaic panels and food crops. applied energy, 206, 1495-1507; https://doi.org/10.1016/j.apenergy.2017.09.113. wasfi, m. (2011). solar energy and photovoltaic systems. journal of selected areas in renewable and sustainable energy, 1-8. weiguo, f., pingping, l., yanyou, w., & jianjian, t. (2012). effects of different light intensities on anti-oxidative enzyme activity, quality and biomass in lettuce. horticultural science, 39, 129134; https://doi.org/10.17221/192/2011-hortsci. © 2022. the authors. this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://research.kpru.ac.th/sac/fileconference/10742018-05-01.pdf https://research.kpru.ac.th/sac/fileconference/10742018-05-01.pdf https://doi.org/%2010.1016/j.rser.%202015.11.005 https://doi.org/%2010.1016/j.rser.%202015.11.005 https://krishijagran.com/ https://doi.org/10.1016/j.renene.2011.03.005 https://doi.org/10.1016/j.renene.2011.03.005 https://doi.org/10.1016/j.renene.2011.03.005 http://www.iosrjournals.org/ https://doi.org/10.21423/twj.v10i1.705%200 https://doi.org/10.21423/twj.v10i1.705%200 https://doi.org/10.1051/epjpv/2018004 https://doi.org/10.1051/epjpv/2018004 https://doi.org/10.14710/%20ijred.2021.32352 https://doi.org/10.14710/%20ijred.2021.32352 https://li01.tcithaijo.org/index.%20php/tstj/article/view/22433/19209 https://li01.tcithaijo.org/index.%20php/tstj/article/view/22433/19209 https://li01.tci-thaijo.org/index.php/prrjscitech/articl%20e/view/41964 https://li01.tci-thaijo.org/index.php/prrjscitech/articl%20e/view/41964 https://doi.org/10.1016/j.apenergy.2016.12.121 https://doi.org/10.32473/edis-hs1337-2019 https://doi.org/10.32473/edis-hs1337-2019 https://doi.org/10.1016/j.seta.2017.08.004 https://doi.org/10.1016/j.seta.2017.08.004 https://doi.org/10.14710/ijred.2021.34423 http://dx.doi.org/10.14710/ijred.6.1.45-53 http://www.ijat-aatsea.com/ https://doi.org/%2010.1016/%20j.enconman.2017.07.007 https://doi.org/%2010.1016/%20j.enconman.2017.07.007 http://natres.psu.ac.th/sjps/index.php/journal/article/view/280/222 http://natres.psu.ac.th/sjps/index.php/journal/article/view/280/222 https://www.bing.com/search?q=https%3a%2f%2fag2.kku.ac.th%2fkaj%2fpdf.cfm%3ffilename%3d13_19_62.pdf%26id%3d3963%26keeptrack%3d5&form=prthth&pc=eupp_hcte&httpsmsn=1&msnews=1&refig=7412c29a75fc490ba22d914d6626a721&sp=-1&pq=https%3a%2f%2fag2.kku.ac.th%2fkaj%2fpdf.cfm%3ffilename%3d13_19_62.pdf%26id%3d3963%26keeptrack%3d5&sc=0-75&qs=n&sk=&cvid=7412c29a75fc490ba22d914d6626a721 https://www.bing.com/search?q=https%3a%2f%2fag2.kku.ac.th%2fkaj%2fpdf.cfm%3ffilename%3d13_19_62.pdf%26id%3d3963%26keeptrack%3d5&form=prthth&pc=eupp_hcte&httpsmsn=1&msnews=1&refig=7412c29a75fc490ba22d914d6626a721&sp=-1&pq=https%3a%2f%2fag2.kku.ac.th%2fkaj%2fpdf.cfm%3ffilename%3d13_19_62.pdf%26id%3d3963%26keeptrack%3d5&sc=0-75&qs=n&sk=&cvid=7412c29a75fc490ba22d914d6626a721 http://article.sapub.org/10.5923.c.ep.201501.01.html https://ideas.repec.org/a/eee/renene/v111y2017icp724-731.html https://doi.org/10.1016/j.apenergy.2017.09.113 https://doi.org/10.17221/192/2011-hortsci microsoft word ogunjuyigbe et al_183_190.doc int. journal of renewable energy development 7 (2) 2018: 183-190 p a g e | © ijred – issn: 2252-4940, july 15th 2018, all rights reserved 183 contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: http://ejournal.undip.ac.id/index.php/ijred hydrokinetic energy opportunity for rural electrification in nigeria ogunjuyigbe ayodeji samson olatunji, ayodele temitope raphael* and ibitoye tahir yomi power energy machine and drives (pemd) research group, electrical and electronic engineering department, faculty of technology, university of ibadan abstract. this paper is part of the ongoing research by the power, energy, machine and drive (pemd) research group of the electrical engineering department of the university of ibadan. the paper presents various sites with possible hydrokinetic energy potential in nigeria with the aim of quantifying their energy potential for rural electrification application. overview of hydrokinetic technology is also presented with the view of highlighting the opportunities and the challenges of the technology for rural electrification. a case study of using hydrokinetic turbine technology in meeting the energy demand of a proposed civic center in a remote community is demonstrated. some of the key findings revealed that nigeria has many untapped hydrokinetic potential site and if adequately harnessed can improve the energy poverty and boost economic activities especially in the isolated and remote rural communities, where adequate river water resource is available. the total estimated untapped hydrokinetic energy potential in nigeria is 111.15mw with the northern part of the country having 68.18mw while the southern part has 42.97mw. the case study shows that harnessing hydrokinetic energy of potential site is promising for rural electrification. this paper is important as it will serve as an initial requirement for optimal investment in hydrokinetic power development in nigeria. keywords: hydrokinetic, hydro-turbine, nigeria, rural electrification article history: received november 16th 2017; received in revised form april 7th 2018; accepted april 15th 2018; available online how to cite this article: olatunji, o.a.s., raphael, a.t. and yomi, i.t. (2018) hydrokinetic energy opportunity for rural electrification in nigeria. int. journal of renewable energy development, 7(2), 183-190. http://dx.doi.org/10.14710/ijred.7.2.183-190 * corresponding author: tayodele2001@yahoo.com 1. introduction in recent time, there is noteworthy interest in hydrokinetic potential globally for power generation using ‘zero head’ or hydrokinetic turbines, which require no dams. nigeria is blessed with copious renewable energy sources (res) that is harmless, clean, and affordable (sambo 2009). this res can be optimally used in solving the challenges of increasing energy demand, cost and adverse effects of fossil fuels. in addition, it can help to combat the increasing health hitches that arise from the use of other conventional energy sources. renewable energy technologies offer many environmental benefits over conventional energy sources (kaygusuz and kaygusuz 2002). the use of electrical energy plays a vital role in economic growth and development of the people’s standard of living. it has been reported by bertsch (2015) that though one-third of the world’s population have access to moving water but not to electricity. majority of the rural residents in nigeria have low standard of living and have limited access to modern amenities like (road, school, communication, etc). despite the efforts in the direction of rural electrification, progress and success rates remain low because of harsh operating conditions, limited resources as well as poor planning (anyi et al. 2010). the living standards of the rural nigerians can be improved by the provision of electricity which is often require to power their loads such as lighting, communication (radio and television), cooling, etc (anyi et al. 2009). rural electrification can be addressed using techniques such as diesel generators, grid-extension and/or renewable energy systems (such as wind, solar, hydro, etc)(ayodele 2014). however, gridextension to rural areas is expensive due to low consumption and poor load factors. this is definitely an unappealing option since most rural residents in nigeria are poor and thus unable to afford electrical services (zomers 2003, ayodele et al. 2016). hydrostatic and hydrokinetic methods are the main methods of harnessing energy from water. hydrostatic method is the traditional way of generating electricity by storing water in reservoirs to create a water head and extracting the potential energy of water through suitable turbine (khan et al. 2008). in hydrokinetic method, the kinetic energy of the flowing rivers is captured and research article citation: olatunji, o.a.s., raphael, a.t. and yomi, i.t. (2018) hydrokinetic energy opportunity for rural electrification in nigeria. int. journal of renewable energy development, 7(2), 183-190, doi.org/10.14710/ijred.7.2.183-190 p a g e | © ijred – issn: 2252-4940, july 15th 2018, all rights reserved 184 converted into electricity by relatively small scale turbines known as hydrokinetic turbines with no head (kusakana 2015). hydrokinetic energy technologies have some advantages over the traditional hydropower system. some of these advantages are: minimum amount of civil work is required; there is no need for acceptable water head and easy installation in free flowing water (khan et al. 2008). as a result of the simple design of the hydrokinetic power system and the natural endowment, there are theoretically huge numbers of potentials sites in nigeria as compared to small hydropower generation. the cost of generating hydrokinetic is lower than that of small hydropower (kusakana 2015). hydrokinetic technology is more economical compared to some other sources of renewable technologies (solar, wind, etc).this paper therefore present the potential of utilizing hydrokinetic technology in nigeria for rural electrification in places where adequate water resource are available. 2. hydrokinetic technology hydrokinetic technology is currently receiving global attention with the intention of overcoming the problems associated with the traditional hydropower generation. hydrokinetic technologies convert kinetic energy from flowing water into electricity, or other forms of energy (güney and kaygusuz 2010). hydrokinetic turbine (hkt) is connected to a generator which will be turn by the blades of the turbine to produce electricity directly from the flowing water in a river stream. the following advantages of the technology are identified over the conventional hydropower: no construction of dam is needed and hence problem associated with massive construction of reservoir is removed; hydrokinetic turbine does not require head which makes the technology suitable for any site, the direction of river flow does not change unlike the wind and solar technology which require additional component to track the direction of the wind/solar. in hydrokinetic technology, no destruction of land and re-settlement as the technology basically rely only on existing kinetic energy in the water stream. 2.1 hydrokinetic turbine the operation principles of the hydrokinetic turbines are similar to that of wind turbines. they are designed to extract the kinetic energy of flowing water, driving a generator to produce electricity. hydrokinetic turbine systems are practically zero-head turbines and are not constrained in a confined environment which makes them different from small hydro turbine systems. small hydro turbines make used of a penstock that transports pressurized water from an elevated hydraulic head. as a consequence of the confined flow and the associated turbine designs, the efficiency of small hydro turbines can be approximately 90% (khan et al. 2009). the most common hydrokinetic turbine designs are vertical and horizontal axis. although, they both execute similar function of extracting kinetic energy from the rivers and ocean, each of the turbines has its own design and performance advantages. the turbines can be mounted in three ways either as bottom structure mounted, floating structure mounted and near surface structure mounted (ogunjuyigbe et al. 2015). the various types of hydrokinetic turbines as illustrated in figure 1 are classified into three major groups: horizontal axis, vertical axis and cross-flow turbines. the turbines under various classifications are depicted in figures 2 and 3. hydrokinetic turbines operate at lower energy extraction efficiency when compared with small hydropower turbines (faure et al. 1986). the propeller type of turbine has theoretical efficiency limit of 59.3% (betz limit) (gorban et al. 2001). in practice hydrokinetic turbines operate with efficiencies between 16% and 42%, depending on the hydraulic conditions of the turbine (faure et al. 1986, gorban et al. 2001). hydrokinetic turbine horizontal axis cross flow vertical axis sc-darrieus (straight blade) h-darrieus (straight blade) darrieus (curved blade) gorlov (helical blade) savonius(straight/skewed) straight axis inclined axis solid mooring buoyant mooring non-submerged generator submerged generator figure 1 .classification of hydrokinetic turbines (khan et al. 2009) int. journal of renewable energy development 7 (2) 2018: 183-190 p a g e | © ijred – issn: 2252-4940, july 15th 2018, all rights reserved 185 (a) inclined axis (b) solid mooring (c) non-submerged (d) submerged figure 2. horizontal axis hydrokinetic turbines (vermaak et al. 2014) (a)in-plane axis (b)squirrel cage darrieus (c) h-darrieus (a) darrieus (b) gorlov (c) savonius figure 3. vertical axis hydrokinetic turbines (vermaak et al. 2014) vertical axis hydrokinetic wind turbines have the following advantages: it is not dependent on the direction of current, the blades are very easy to construct and the span of the blade can be easily increased. however, the efficiency is slight lesser than horizontal turbine and the fixed blade does not start naturally. the horizontal axis wind turbine has the advantages of being more efficiency and the technology is more matured compared to the vertical turbine (coiro 2007). however, the turbine depends on the direction of stream current and it requires complex mechanism for blade rotation 3. opportunities and challenges of harnessing hydrokinetic energy in nigeria this section presents the opportunities and the possible potential of hydrokinetic energy for electricity generation in rural community in nigeria. 3.1 opportunities of hydrokinetic the demand for renewable energy is expected to increase significantly in the nearest future as it is expected that more rural community will be electrified. according to united states energy information administration; the world electricity consumption will likely increase by 73% between 1999 and 2020 especially from sub-sahara africa where the population is fast growing. hydrokinetic energy has great potential to bridge this gap in nigeria as there are presently abundant flowing river in the rural villages across the country that are yet to be tapped. exploitation and utilization of this ample potential in nigeria would provide opportunity in boosting the socio-economic activities in the local communities. hydrokinetic technology can offer opportunity for cheap and affordable rural electrification for meeting the basic needs which can improve the quality of life of the rural dwellers. the bts station which houses the communication equipment can also be powered by hydrokinetic power thereby providing good communication that can reduce ruralurban migration. similarly, the small scale business such as “gari” processing industries can be enhanced by supply from hydrokinetic source, thereby empowering the rural dwellers and improving their economic values. furthermore, social life will be boosted which can promote unity within the community as community halls and central buildings can be powered using this technology 3.2 possible challenges of utilizing hydrokinetic technology in nigeria despite the great opportunities and high enthusiasm in this field, skepticism on technological feasibility still exists. many technologies’s related questions are being put forward continuously in addition to fundamental inquiries (availability of resources, area of application). the possible challenges in developing hydrokinetic technology in nigeria are briefly highlighted in the succeeding sub-section. 3.2.1 resource assessment one of the possible problems militating against hydrokinetic development in nigeria is lack of adequate information due to limited research on hydrokinetic resource assessment in the country. there is need to identify resourceful sites to harness hydrokinetic energy. this would necessitate an investigation of macro and micro scale site assessment, determination of annual energy yield and analysis of river characteristics. presently, this is lacking for hydrokinetic energy as most of the assessments done in nigeria are in support of citation: olatunji, o.a.s., raphael, a.t. and yomi, i.t. (2018) hydrokinetic energy opportunity for rural electrification in nigeria. int. journal of renewable energy development, 7(2), 183-190, doi.org/10.14710/ijred.7.2.183-190 p a g e | © ijred – issn: 2252-4940, july 15th 2018, all rights reserved 186 conventional small hydropower generation (manohar and adeyanju 2009) and not for hydrokinetic generation. hence, lack of databases for hydrokinetic potential in the country is a major challenge. 3.2.2 climatic condition the main drawback of hydrokinetic technology is its resource-dependent power output and its reliance on weather and climatic condition (chen et al. 2007). nigeria has two main seasons: the wet season between april –september and the dry season between october march. thus, it cannot always produce sufficient energy to match the fluctuating load requirements continuously throughout the year without the use of energy storage system (kusakana 2015) especially in the dry season. this often form one of the reasons the investors are shying away from such technology. similarly, the advantages of hybridization with other res may be employed to overcome this limitation. 3.2.3 economics viability the most prevailing factor that hinders the success of most energy technologies is the associated cost of installation. this includes: the capital cost, operations and maintenance cost, scalability, simplicity of design, applications diversity, material and labor engagement. several other factors that may affect the cost indirectly are system reliability, societal acceptance and system performance (khan et al. 2008). due to this various economic and social constraint in addition to the small population in the rural communities, government and private company are always reluctant to invest in this kind of adventure as it always considered being uneconomical due to the small population. 4. estimation of nigerian hydrokinetic energy potential the kinetic energy of flowing water can be determined from the density of the water, the velocity at which the water flows and the cross-sectional area at which the energy will be extracted. the hydrokinetic energy available in a mass of flowing river can be determined using: 31 2 p avr= (1) but q va= (2) therefore, the hydrokinetic potential expressed in (1) can be re-written as: 21 2 p qv= (3) where, p is the hydrokinetic power/available kinetic power; ρ is the water density (1000kgm-3), a is the cross sectional area of extraction (m2), q is the discharge rate (m3s-1), v is the water flow velocity (m/s). hydrokinetic power potential is often reported as a power density which is the power normalized to a unit area. 21 2d p p v a r= = (4) also, hydrokinetic energy density (energy potential) can be written as: d de e t= ´ (5) the power and energy generated by a given hydrokinetic turbine are given as follows, respectively: dhkt pp p a c´ ´= (6) hkt hkte p t= ´ (7) where pc is the coefficient of performance of the hydrokinetic turbine, v is the water velocity (m/s) and t is the duration of power supply (hours) figure 4. map of nigeria showing some major rivers (f.a.o 2012) when considering flows in rivers, one can make the reasonable assumption that the density (ρ) remains constant, even with changes in temperature. the discharge rate, area and velocity remain the only variables parameters required to determine the kinetic power. the calculation of the kinetic power depends on either the discharge rate or cross section area which is either the river cross-sectional area or the area of the device (turbine) that will be used to extract the kinetic energy. for the purposes of this study and to generalize the power potential over such a large land mass, the river cross sectional area will be used. figure 4 illustrates the spread of rivers across the country designating the extent of hydrokinetic potential in the country int. journal of renewable energy development 7 (2) 2018: 183-190 p a g e | © ijred – issn: 2252-4940, july 15th 2018, all rights reserved 187 table 1. some rivers in nigeria and their average discharge (zarma 2006) location river average discharge (m3/s) danko jebba zungeru ii zungeru i shiroro zurubu gwaram izom gudi kafancha kurra ii richa ii richa i mistakuku kombo kiri kramti beli garin deli sarkin dango gembu kasimbila markurdi lokoja onisha ifon ikom afikpo niger niger kaduna kadunai kaduna kaduna jamaare gurara mada kongum sanga daffo mosari kurra gongola gongola kam taraba taraba suntai donga kastina ala kastina ala benue niger niger osse cross cross 1650 1767 343 294 55 55 75 55 41.5 2.2 5.5 4.0 6.5 2.0 128 154 80 266 323 20 45 170 740 3185 6253 6635 80 759 1621 the annual rainfall ranges between 500 and 1000 mm in the arid north and between 1,500 and 4,000 mm in the southern region, the country experiences flowing rivers and streams for two-thirds of the year (f.a.o 2012).river niger and benue are the major rivers in nigeria with average discharge of 5589 m3/s and 3400 m3/s respectively (olatunde and oladele 2013). the list of some flowing rivers in nigeria and their average discharge and their specific locations is presented in table 1 5. case study in this section, a case study is presented for a typical isolated rural community with about 10 households located close to river in the southern part of nigeria. hydrokinetic energy is offered to meet the energy demand of proposed civic center for the remote village. the main occupation of the duelers are fishing and farming. the civic center would serve as a central point where villagers would charge phones and their rechargeable electronics. the center is also expected to serve as common meeting point where the inhabitants can socialize. the daily load profile depicting the energy requirement of the civic center is depicted in figure 5 with average power demand of about 1 kw with a peak load of 1.38kw. the kinetic energy of the flowing rivers is captured and converted into electricity by relatively small-scale turbines. the average water velocity of the river close to the community is about 1.2m/s it was observed to flow with almost constant velocity throughout the day as depicted in figure 6 2 4 6 8 10 12 14 16 18 20 22 24 0 200 400 600 800 1000 1200 1400 time (hour) lo ad d em an d (w ) figure 5. load profile for the proposed civic center 5 10 15 20 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 time of the day(hour) w at er v el oc ity (m /s ) figure 6. water velocity for the available river close to the village citation: olatunji, o.a.s., raphael, a.t. and yomi, i.t. (2018) hydrokinetic energy opportunity for rural electrification in nigeria. int. journal of renewable energy development, 7(2), 183-190, doi.org/10.14710/ijred.7.2.183-190 p a g e | © ijred – issn: 2252-4940, july 15th 2018, all rights reserved 188 in order to ensure smooth supply of energy throughout the day, the hydrokinetic turbine is made to supply electrical power to the load through the dc bus consisting of battery. hence the total energy supply to the loads come from hybrid of hydrokinetic turbine and the battery as depicted in figure 7 hydrokinetic turbine inverter battery dc/dc converter ac/dc converter load dc bus ac bus figure 7. hybrid configuration of hydrokinetic and battery the total energy supply to the load at time t (i.e. energy from battery and hydrokinetic turbine) can be written as: ( ) ( ) ( )hkt batte t e t e t= + (8) the required storage capacity of the battery system in ampere-hour (ah) can be determined using: d l batt batt inv s a e m dod u ´ = h ´h ´ ´ (9) where da is the battery autonomous days (i.e. the maximum number of days the battery can supply continuous energy without a recharge by hydrokinetic energy source), dod is maximum permissible depth of discharge of the battery and us is the system voltage in volts, batth is the efficiency of inverter (%), invh efficiency of inverter (%), le is the total energy demand. the battery charges when the power from hydrokinetic source is greater than the total load demands. the excess power is ultilised to charge the battery. however, power is drawn from the battery (discharging) whenever the load demands is greater than the power output of the hydrokinetic generator (i.e power deficit). the state of charge of the battery at time (t) is given as: ( ) ( ) ( 1).(1 ) ( ) .lhkt b inv e t soc t soc t e t æ ö = -s + hç ÷ hè ø , charging (10) 1 1 l hkt inv e (t ) soc(t ) soc(t ).( ) e (t) æ ö = -s + -ç ÷ hè ø , discharging (11) the basic parameters used in simulating the battery storage are shown in table 2 table 2 battery storage parameters for the simulation parameter da (days) dod su (volt) batth invh battm (ah) values 2 60% 12 0.9 0.9 350 6. result and discussion this section presents the results quantifying the available hydrokinetic energy potential in various part of nigeria. it also presents the results showing the potential of using hydrokinetic turbine in meeting the electrical energy needs of rural community with good water resources. 6.1 hydrokinetic potential in nigeria the hydrokinetic power potential of the flowing rivers depicted in table 1 is evaluated using (4) given the average velocity of flowing river in nigeria as 3m/s. the sites are classified into two (the northern and southern), this is to allow us understand the hydrokinetic potential in the northern and southern part of the country. the estimation hydro kinetic potential in the northern nigeria is depicted in table 3 while that of southern nigeria in shown in table 4. table 3 estimated hydrokinetic power (mw) and energy potential (mwh) of rivers located in northern nigeria location river hydrokinetic power density (mw/m2) hydrokinetic energy density (mwh/m2) gwaram jamaare 0.3375 8.1 kafancha kongum 0.0099 0.2376 kombo gongola 0.576 13.824 kiri gongola 0.711 17.064 beli taraba 1.197 28.728 garkin suntai 0.09 2.16 danko donga 0.2025 4.86 gembu kastina ala 0.765 18.36 zurubu kaduna 0.2475 5.94 zungeru kaduna 1.5435 37.044 shiroro kaduna 1.323 31.752 donko niger 7.425 178.2 jebba niger 7.9515 360 kasimbila kastina ala 3.33 79.92 makurdi benue 14.3325 343.98 lokoja niger 28.1385 675.324 total 68.18 1805.4936 table 4. hydrokinetic power (mw) and energy (mwh) potentials of some rivers located in southern nigeria location river hydrokinetic power density (mw/m2) hydrokinetic energy density (mwh/m2) onitsha niger 29.857 716.568 ifon ose 0.36 8.64 ikom cross 3.4153 81.9672 afikpo cross 7.2945 175.068 ajura ogun 0.23967 5.75208 apoje osun 1.8025 43.26 total 42.97 1031.255 from table 3, it is observed that river niger (lokoja) and benue has the highest potential for hydrokinetic power in the northern part of nigeria with power density values of 28.43 mw/m2 and 14.33 mw/m2, respectively. the rivers also have energy density of 675.324 mwh/m2 int. journal of renewable energy development 7 (2) 2018: 183-190 p a g e | © ijred – issn: 2252-4940, july 15th 2018, all rights reserved 189 and 343.98 mwh/m2. the total hydrokinetic power and energy potential in the north was estimated to be 68.18 mw/m2 and 1805.4936 mwh/m2, respectively. table 4 reveals that river niger (onitsha) and cross (afikpo) has the highest power potential for hydrokinetic power in southern nigeria with the value of 29.86 mw/m2 and 7.29 mw/m2, respectively. the total power and energy potential of hydrokinetic power generation in the southern part is estimated as 42.97 mw/m2 and 1031.255 mwh/m2. this demonstrates that the country has an estimated total power of 111.15 mw/m2 untapped hydrokinetic potential spread across the nation. this can be used to help meet some of the energy needs of some rural villages. 6.2 meeting load demand of an isolated community using hybrid of hydrokinetic and battery. load audit of a proposed civic center was conducted for a typical village with the intention of meeting the electrical requirement of the building using the available water resource within the community. the average load requirement of the center was determined to be about 1kw. the center is thought to improve the social life of the villagers by providing the central charging point for rechargeable electronics and lighting. based on the load requirements of the civic center and the water velocity of the river, the battery sized was determined to be 350ah and the size of the hydrokinetic turbine was evaluated to be 2.5kw assuming the coefficient of performance cp of 35%. the simulation result showing the load profile of the civic center, hydrokinetic power generation using the 2.5kw turbine (determined using equation 5) and the battery state of charge (determined using equations 6-7 and equations 9 are depicted in figure 8. 2 4 6 8 10 12 14 16 18 20 22 24 200 400 600 800 1000 1200 1400 time (hour) p ow er (w ) load profile battery soc hydrokinetic turbine figure 8. power output of hydrokinetic turbine, state of charge of battery and the load profile of the civic center the figure reveals that the hydrokinetic turbine generates nearly constant power of about 1.34kw throughout the 24 hours period. this is enough to meet the average power demand (about 1kw) of the civic center but will not be able to meet the peak load demand (1.38kw). the constant power from the turbine was as a result of nearly constant water flow velocity of about 1.2m/s throughout the entire period. the figure also reveals that the soc of the battery has a negative correlation with the load profile. this is expected because, whenever there is increase in load demand, more energy is required from the battery to meet the load demand thereby putting the battery in the state of discharge. this is indicated in the figure with the negative trend of the soc of the battery each time there is an increase in load profile. similarly, the battery is charged whenever there is decrease in the load demand. figure 9 depicts the combination of the battery and the hydrokinetic turbine in meeting the electrical power demands of the civic center. the figure reveals that the hybrid of the battery and hydrokinetic turbine generates peak power of 1.6 kw. this indicates that at any period of time, the hydride configuration will effectively meet the peak power demand (1.38kw) of the civic center. the configuration consisting of a 350ah battery and 2.5kw hydrokinetic turbine 2 4 6 8 10 12 14 16 18 20 22 24 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 time (hour) p ow er (w ) battery +hydrokinetic load profile figure 9. power output of hybrid of hydrokinetic turbine and battery in meeting the load requirement of civic center 7. conclusion this paper has investigated the opportunity of hydrokinetic power generation in nigeria as means of rural electrification. the gaps which can hinder the development of hydrokinetic technology in nigeria for rural electrification were identified. also, the potential of some rivers across the country being used for hydrokinetic power generation were explored. this study found that there is huge hydrokinetic potential with an estimated value of 111.15 mw which are untapped in nigeria. the hydrokinetic potential in northern part of the country is estimated to be 68.18 mw while that of the southern part is 42.97 mw. a case study conducted using hydrokinetic turbine and battery reveals that the technology is feasible for rural community. this paper is useful as it reveals the hydrokinetic energy potential in nigeria. it also showcases the potential of using hydrokinetic technology to meet the electrical energy need of isolated communities. furthermore, it can also serve as first hand useful information for investors and project developers who are interested in rural electrification. future research will compare the hydrokinetic technology to other renewable energy generators in term of cost of energy, life cycle cost and reliability in meeting the energy needs of isolated rural communities. reference anyi, m., ali, s. & kirke, b., (2009). remote community electrification. international conference on renewable energies and power quality, (icrepq'09). valencia,spain, 15-17. citation: olatunji, o.a.s., raphael, a.t. and yomi, i.t. (2018) hydrokinetic energy opportunity for rural electrification in nigeria. int. journal of renewable energy development, 7(2), 183-190, doi.org/10.14710/ijred.7.2.183-190 p a g e | © ijred – issn: 2252-4940, july 15th 2018, all rights reserved 190 anyi, m., kirke, b. & ali, s., (2010). remote community electrification in sarawak, malaysia. renewable and sustainable energy reviews, 35 (7), 609–613. . ayodele, t.r., (2014). feasibility study of stand-alone hybrid energy system for rural electrification in nigeria: the case study of ala-ajagbusi community. international journal of renewable energy resources, 4, 1-12. ayodele, t.r., ogunjuyigbe, a.s.o. & babatunde, j.b., (2016). sustainable electricity generation in rural communities using hybrid energy system: the case study of ojataye village. international journal of renewable energy, 11 (1), 43-56. bertsch, d.j., (2015). hydrokinetic energy: trying to navigate the energy and wave law frame work to develop new renewable energy technology. available from: . http://www.elizabethburleson.com/hydrokineticenergyderek bertsch.pdf, accessed 15th may 2015., available from: http://www.elizabethburleson.com/hydrokinetic energyderekbertsch.pdf. chen, f., duic, n., aives, l.m. & carvalho, m.g., (2007). renewable energy solutions for islands. renew sustainable energy reviews, 11 (8), 1888-1892. coiro, d.p., (2007). experiments on horizontal and vertical axis water turbines for harnessing marine currents: technological and economical aspects. www.dpa.unina.it/adag/. available from: www.dpa.unina.it/adag/. f.a.o, (2012). food and agriculture organization of the united nations, hydrology and inland water resources of nigeria, retrieved august 2012 http://www.fao.org/docrep., 1-50. faure, t.d., pratte, b.d. & swan, d., (1986). the darrieus hydraulic turbine-model and field experiments. fourth international symposium on hydro power fluid machinery: presented at the winter annual meeting of the american society of mechanical engineers. anaheim, california: american society of mechanical engineers, 123-127. gorban, a.n., gorlov, a.m. & silantyev, v.m., (2001). limits of the turbine efficiency for free fluid flow. journal of energy resources technology, transactions of the asme, 123 (2-4), 311-317. güney, m.s. & kaygusuz, k., (2010). hydrokinetic energy conversion systems. renewable sustainable energy reviews, 14, 2996–3004. kaygusuz, k. & kaygusuz, a., (2002). renewable energy and sustainable development in turkey. renewable energy 25, 431–453. khan, m., j., bhuyan, g., iqbal, m.t. & quaicoe, j.e., (2008). river current energy conversion systems: progress, prospects and challenges. renewable and sustainable energy reviews, 12, 2177-2193. khan, m.j., bhuyan, g., iqbal, m.t. & quaicoe, j.e., (2009). hydrokinetic energy conversion systems and assessment of horizontal and vertical axis turbines for river and tidal applications: a technology status review. applied energy, 1823 – 1835. kusakana, k., (2015). feasibility analysis of river off-grid hydrokinetic systems with pumped hydro storage in rural application. energy conversion and management, 96, 352362. manohar, k. & adeyanju, a.a., (2009). hydropower energy resources in nigeria. journal of engineering and applied sciences, 4 (1), 68-73. ogunjuyigbe, a.s.o., ayodele, t.r. & ibitoye, t.y., (2015). review of hydrokinetic energy potential in nigeria. first faculty of technology conference. university of ibadan, nigeria, 129-134. olatunde, s.o. & oladele, o., (2013). eco-partitioning and indices of heavy metal accumulation in sediment and tilapia zillii fish in water catchment of river niger at ajaokuta, north central nigeria. international journal of physical sciences, 8 (20), 1111-1117. sambo, a.s., (2009). strategic developments in renewable energy in nigeria. vermaak, h.j., kusakana, k. & s.p., k., (2014). status of microhydrokinetic river technology in rural applications. renewable and sustainable energy reviews, 29, 625 633. zarma, i.h., (2006). hydropower resources in nigeria. 2nd hydropower for today conference of the international centre for small hydropower (icshp), hangzhou, china. canada, 3-7. zomers, a., (2003). the challenges of rural electrification. energy for sustainable development 7(1), 69–76. international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (2), 375-380 |375 https://doi.org/10.14710/ijred.2023.49909 issn: 2252-4940/© 2022.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id qpva-based electrospun anion exchange membrane for fuel cells asep muhamad samsudin a,b* and viktor hackera ainstitute of chemical engineering and environmental technology (ceet), graz university of technology, austria bdepartment of chemical engineering, diponegoro university, indonesia abstract. the anion exchange membrane is one of the core components that play a crucial and inseparable role in alkaline anion exchange membrane fuel cells. anion exchange membranes (aems) were prepared from quaternary ammonium poly(vinyl alcohol) (qpva) by an electrospinning method. qpva was used both as material for electrospun fiber mats and as filler for the inter-fiber void matrix. the objective of this work is to investigate the influence of the inter-fibers void matrix filler concentration on the properties and performance of eqpva-x aems. ftir spectra were used to identify the chemical structures of the aems. the primary functional groups of pva and quaternary ammonium-based ion conducting cation were detected. the surface morphology of qpva nanofiber mats and eqpva-x aems was observed using sem. electrospun nanofiber structures of qpva with an average size of 100.96 nm were observed in sem pictures. the ion exchange capacity, swelling properties, water uptake, and oh− ions conductivity were determined to evaluate the performance of eqpva-x aems. by incorporating the qpva matrix of 5 wt.% concentration, the eqpva-5.0 aems attained the highest ion exchange capacity, water uptake, swelling properties, and oh− conductivity of 0.82 mmol g−1, 25.5%, 19.9%, and 2.26 m⋅s cm−1, respectively. electrospun qpva aems have the potential to accelerate the development of alkaline anion exchange membrane fuel cells. keywords: anion exchange membranes, fuel cells, qpva, electrospinning, nanofibers @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 30th oct 2022; revised: 2nd january 2023; accepted: 29th january 2023; available online: 11th february 2023 1. introduction the decline of energy reserves and the occurrence of ecological damage encourage researchers to develop renewable energy sources that are more environmentally friendly, efficient, and sustainable. efforts to produce electrical energy from mechanical, thermal, and chemical energy have continued in the last few decades (ayaz et al., 2022). among the developed new energy sources, the fuel cell is regarded as eco-friendly alternative energy that can replace conventional fossil fuels with high efficiency of energy conversion and almost no emissions. anion exchange membrane fuel cells (aemfcs), as one fuel cell category, are getting immense attention because of their advantages. the advantages include the opportunity to utilize less expensive transition metals instead of a costly catalyst of platinum group metals (pgm) and a faster oxygen reduction process. due to the counter-direction between the fuel and oh− ions, aemfcs also benefit from reduced fuel crossover and reduced corrosion concerns in alkaline environments (iravaninia & rowshanzamir, 2015; samsudin et al., 2022). anion exchange membranes (aems) are part of a fuel cell that has an essential role in hydroxide conduction, electron inhibition, and gas barrier (hagesteijn et al., 2018; ramaswamy & mukerjee, 2020). despite their positive points, the development of aemfcs faces several difficulties. since hydroxide ions have lower mobility than hydrogen ions, the ionic conductivity of aems tends to be lesser than that of proton exchange membranes (pems) (das et al., 2022). the toxic and expensive solvents are also an issue in the synthesis of some * corresponding author email: asep.samsudinl@live.undip.ac.id (a.m. samsudin) aems. in addition, complex routes and costly equipment have also become a concern (wang et al., 2013). polymers are the backbone material used in the manufacture of aems. to date, many types of polymers have been developed for aems, ranging from poly (aryl ether) based aems (e.g., poly (ether sulfone) (du et al., 2022; wang et al., 2022), poly(2,6-dimethyl-1,4-phenylene oxide)(becerraarciniegas et al., 2019; mayadevi et al., 2022) and aryl-ether free aems (e.g., polybenzimidazole (guo et al., 2022; g. zhang et al., 2022) and aliphatic-based aems (e.g., poly (vinyl alcohol) (huang et al., 2022; samsudin & hacker, 2019, 2021). poly(vinyl alcohol) (pva) is a synthetic polymer that possesses scentless, flavorless, biocompatible, and biodegradable characteristics. due to its beneficial attributes, pva is frequently employed as a backbone polymer for aems development. due to its hydrophilicity, it exhibits a high water uptake and possesses exceptional film-forming characteristics. furthermore, the availability of reactive functional groups and lesser fuel crossover are favorable for chemical crosslinking and other modifications that improve the properties of the membrane (aslam et al., 2018; ding & qiao, 2022; samsudin et al., 2022; susanto et al., 2016; zhang et al., 2013). various techniques and methods for membrane preparation of functionalized polymers have been introduced. solution casting is a membrane casting technique that has been widely used because it is simple, easy, and versatile (samsudin et al., 2022). apart from solution casting, another method that is starting to attract attention is electrospinning. this technique employs a high-voltage source to induce an electric field from research article https://doi.org/10.14710/ijred.2023.49909 https://doi.org/10.14710/ijred.2023.49909 mailto:asep.samsudinl@live.undip.ac.id https://orcid.org/0000-0002-0131-5667 https://orcid.org/0000-0001-5956-7579 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.49909&domain=pdf a.m.samsudin et al int. j. renew. energy dev 2023, 12(2), 375-380 |376 issn: 2252-4940/© 2023. the author(s). published by cbiore the spinneret to the collector. a taylor cone emerges at the spinneret's edge at a particular electric field intensity. after intensity overcomes polymer drop surface tension, an electrified solution jet is released from the taylor cone. the solution jet evaporates and solidifies in the collector, forming nanofibers (sood et al., 2016). electrospinning is favorable because it facilitates the formation of interlinked structures, which improves oh− transfer. additionally, electrospinning is effective for achieving a uniaxial alignment of nanofiber-formed polymer chains that has the potential to strengthen the membrane (fennessey & farris, 2004; sood et al., 2016; tamura & kawakami, 2010). despite the many benefits, the fabrication of anion exchange membranes by electrospinning is still limited. yang et al. prepared electrospun aems, which utilized a combination of poly(vinyl alcohol) and chitosan. they investigated the impact of various crosslinking times on the characteristics of the membrane (yang et al., 2018). gong et al. compared imidazolium-functionalized polysulfone (impsf) aems manufactured by the solution casting and electrospinning methods. the results depict that electrospun aems produce higher conductivity and lower swelling properties than cast membranes (gong et al., 2016). du et al. fabricated quaternized poly(2,6-dimethyl-1,4-phenylene oxide) electrospun (qpponf)/poly(vinyl alcohol) anion exchange membrane. by varying the ratio between qpponf and pva, it was found that the addition of the ratio of qpponf to pva increased the ion conductivity of aems (du et al., 2020). most developments in the field of electrospun anion exchange membranes have concentrated on nanofiber mats, while studies on matrix fillers are still rare. in this work, qpvabased electrospun nanofiber aems were prepared by an electrospinning method. a commercially available qpva, namely gohsenxtm k-434, was used for both the fibers and the matrix filler between the fibers. the objective of this work is to study the influence of different concentrations of qpva as interfibers void matrix fillers on the aems properties. 2. experimental 2.1 preparation of electrospun aems 12 wt.% of qpva solute was prepared by dissolving qpva (gohsenxtm k-434, 85.5-88.0% hydrolyzed, 18-22 mpa.s, obtained from mitsubishi chemical corporation) in ultrapure water (upw, resistivity ~18 mω.cm) with constant stirring at 80–90 °c overnight. the chemical structure of gohsenxtm k-434 is depicted in figure 1. a quantity of the qpva solution was then inserted into a 10 ml size spinneret needle syringe. a horizontal movable spinneret and a drum collector which a distance of 10 cm were used to increase the dimensional homogeneity of the membrane. then, 20 kv high voltage was introduced between the spinneret edge and the aluminium foilcoated drum collector. the electrospinning was carried out with a polymer flow rate of 0.5 ml/h at room temperature. the relative humidity was set in the range of 50−60%. the qpva fiber mats were heated at 130 °c for one hour to induce physical crosslinking between qpva polymer chains. subsequently, they were soaked in a cross-linker solution composed of 2.5 wt.% glutaraldehyde, 0.2 wt. % hydrochloric acid in acetone to promote chemical crosslinking. to produce dense aems (qpva-x), qpva fiber mats were then submerged in various concentration of gohsenxtm qpva solution at room temperature. then, crosslinking was repeated for aems to increase crosslinking degree qpva chains. the identities of the qpva aems were determined using table 1. the aem's preparation processes are shown in figure 2. fig. 1 the structure of gohsenxtm k-434 table 1 the aems samples composition membrane samples fiber (qpva) (wt.%) matrix filler concentration (qpva) (wt.%) cqpva 12 eqpva-2.5 12 2.5 eqpva-5.0 12 5.0 eqpva-7.5 12 7.5 eqpva-10.0 12 10.0 2.2 ftir analysis an ir-bruker alpha spectrometer was used for the ftir study to determine the major functional group of the membranes. ftir analysis was conducted at a wavenumber of 400–4000 cm−1 and a resolution of 4 cm−1. the ir spectra of the aems were displayed as absorbance versus wavenumber graphs. 2.3 sem analysis sem analysis (zeiss supra 55vp) was conducted to study the morphology of the electrospun qpva aems. the measurement was performed at a voltage of 15 kv. the nanofiber size distribution of the electrospun mats was determined using imagej software on sem results. 2.4 ion exchange capacity (iec) the iec of eqpvax aems was measured by back titration. firstly, the membranes were weighed and then soaked for 24 h in 1 m koh solution to change the aems into oh− form. after removing the koh residue using ultra-pure water for 24 h, the aems samples were then soaked for 24 h in 0.1 m hcl solution. the titration was accomplished using 0.1 m koh solution. iec was calculated using formula 1 (samsudin et al., 2022) as follows: iec = (𝐕𝐕𝐛𝐛−𝐕𝐕𝐦𝐦).𝐂𝐂𝐇𝐇𝐂𝐂𝐇𝐇 𝐰𝐰𝐝𝐝 (1) where, vb, vm, chcl, and wd are the consumed koh volumes of the 0.1 hcl solution without membrane samples, the consumed koh volumes of aems, the hcl concentration, and the dry weight of aems, respectively. 2.5 swelling properties the aems' swelling properties were assessed by measuring water uptake (wu) and swelling degree (sd). the wu was measured by determining the weight difference of the aems after submerging them in water. on the other hand, the sd was evaluated by comparing the volume of aem due to water immersion for 24 h in rt. formula 2 and 3 (movil et al., 2015) were used for calculating the wu and sd as follows: wu = ww − wd wd x 100% (2) sd = vw − vd vd x 100% (3) where, ww, wd, vw, and vd are the wet weight, dry weight, wet volume, and dry volume of aems, respectively. a.m.samsudin et al int. j. renew. energy dev 2023, 12(2), 375-380 |377 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 2 schematic of the preparation procedure 2.6 ion conductivity electrochemical impedance spectroscopy was utilized to evaluate ionic conductivity (σ). the gamry reference 600 potentiostat was used in conjunction with a standard four-probe conductivity clamp (scribner associates, usa). the impedance of oh− form aems was measured between 0.1 hz and 10 khz frequency and with 50 mv voltage. formula 4 (feketefoldi & cermenek, 2016) was used for calculating the σ as follows: σ = d rm.t.w (4) where d, rm, t, and w are the distance of electrodes, the impedance of membranes, the thickness of wet aems, and the width of the membranes, respectively. 3. results and discussion 3.1 chemical structure of aems ftir spectroscopy was used to verify the chemical composition of eqpva-x aems. figure 3 displays the ftir spectra of eqpva-x aems. the absorption peaks at 3378 and 1022 cm−1 appear to be caused by the –oh and c–o stretch in the pva polymer backbone. the peak in the bending vibration at 2940 cm−1 attributable to the existence of the c–h group. the stretching vibration of the chemical bond c=o was indicated by the intensity at the wavenumber of 1734 cm−1. the intensity at 1434 cm−1 and 1376 cm−1 occur on account of the presence of the ch3 bend and ch2 bend, respectively. a wavenumber of 1240 cm−1 belongs to the c–o–c bond stretching vibration, which indicates the establishment of covalent bonds between – oh groups of qpva and –cho groups from ga. fig. 3 ftir spectrum of eqpva-5.0 fig. 4 chemical structure of crosslinked qpva the intensities at 1098 cm−1 and 841 cm−1 correspond to the c– n stretch and n–h bend of ion-conducting cation groups in qpva. the chemical structure of crosslinked qpva is illustrated in figure 4. 3.2 morphology figure 5a shows the surface morphology of the eqpva nanofiber aems from the sem analysis. it was seen that the nanofibers formed well with no beads. the size distribution of the nanofiber mat is presented in figure 5b. the qpva mats fibers possess a size distribution of 69–⁠179 nm and a mean fiber diameter of 100.96 nm, identifying them as nanofibers (patel et al., 2018). inter-fiber void space, visible as pores between fibers, is observed in the membranes. this inter-fiber void space of the membrane should be occupied with a matrix filler in order for it to be utilized in fuel cells. fig. 5 a) sem image of qpva nanofibers, b) qpva nanofibers size distribution, c) sem image of eqpva-5 aem a.m.samsudin et al int. j. renew. energy dev 2023, 12(2), 375-380 |378 issn: 2252-4940/© 2023. the author(s). published by cbiore this matrix filler can prevents fuel and oxidant gas transport through the membrane (i.e., gas crossover). the permeability of fuel through aems should be prevented for fuel cells. since this crossover process may lead to voltage loss on account of the mixed potential caused by the penetrated fuel oxidation. additionally, fuel crossover could cause peroxide and excess heat generation, which can degrade the fuel cell (francia et al., 2011; t. huang et al., 2022; inaba et al., 2006). figure 5c shows the eqpva-5.0 aems. since this membrane is derived from matrix addition to the eqpvc nanofibers, we can see that the inter-fiber voids are filled with the matrix while maintaining the nanofiber structure. 3.3 ion exchange capacity (iec) iec can be described as the capability of the aems functional groups to carry out ions displacement, which is integrated and loosely attached to its polymer backbone chain structure by oppositely charged ions in the adjacent solution (elumalai et al., 2018). iec demonstrates the quantity of functional groups or active sites in an anion exchange membrane that is accountable for ion exchange or facilitates the transfer of hydroxide (kumar et al., 2018). iec can be expressed as milliequivalent or millimoles of anionic-exchange groups per gram of the dry membrane (meq g−1 or mmol g−1). figure 6 depicts the ion exchange capacity (iec) of eqpvax aems in hydroxide form of aems at 30°c. the iec of eqpva2.5 aems is 0.46 mmol g−1. the ion exchange capacity increases by around 78% after enhancing the concentration of the qpva matrix to 5 wt.% (eqpva-5.0), which is the highest iec value. when the concentration of the qpva matrix is enhanced to 7.5 and 10 wt.%, the iec decreases to 0.79 and 0.41 mmol g−1. the decrease of iec at higher concentrations of the qpva matrix is possibly owing to the qpva matrix high viscosity, which causes an impediment for the matrix filler solution to infiltrate the interfiber space of the nanofiber mats. consequently, the number of ion-conducting cations in the electrospun qpva aems decreases, followed by the decline of iec. 3.4 swelling properties the existence of water in the anion exchange membranes is significant in the process of conducting hydroxide ions. the ion movement process in aems is highly reliant on the membrane's hydration level (λ, i.e., the water molecules number per oh–), the dispersion and distribution of water, and the solvation of oh− ions (zelovich et al., 2019). water clusters are able to act as anion transport channels within the aems, improving hydroxide conductivity (zhang et al., 2013). . fig. 6 ion exchange capacity of eqpva-x aems. fig. 7 sweling properties of eqpva-x aems. at low water content or low hydration level and in alkaline conditions, oh− ions can react with side cation charge groups, which lessen the iec of aems since only free ions play a part in the conductivity. furthermore, the degradation of the cation charge groups will reduce the performance and lifetime of the aems (tomasino et al., 2022). in addition to the lack of hydration, excessive water content is also avoided in aems. this excessive water content can induce severe swelling, which can cause instability in the membrane dimensions causing mechanical degradation. moreover, the excessive water content may dissolve some of the charged cations bound in the polymer backbone, lowering the iec and causing the anionic conductivity to decrease (vandiver et al., 2014; zheng et al., 2018). the water uptake (wu) and swelling degree (sd) of eqpvax aems are depicted in figure 7. the aem with the lowest qpva matrix concentration (eqpva-2.5) has a water uptake of 17.6 wt.%. by incorporating qpva 5 wt.% (eqpva-5.0), the wu increases by 45% to 25.5 wt. %. however, by further increasing the concentration of the matrix to 7.5 (eqpva-7.5) and 10.0 wt.% (eqpva-10.0), the wu decrease to 20.3 and 16.9 wt.%. the results of the wu are in accordance with the iec values. change in water uptake is significantly connected to the amount of cation groups attached to the aems, which in the case of these eqpva-x membranes, are quaternary ammonium groups from qpva (samsudin et al., 2021). the swelling degree measurement demonstrates a similar tendency to water uptake. the higher the water uptake indicates the higher the water content in the membrane, which leads to swelling formation and results in changes in the dimensions of the membrane. 3.5 hydroxide conductivity hydroxide conductivity is the most crucial characteristic of aems, owing to the principal role of aems as oh− ions conductors. the conductivity of aems is very reliant on the wu and iec of the aems. the high hydrophilicity of the anion exchange membranes is resulted from the high density of cation charge groups within the aems, which provide sufficient anionic conductivity (ayaz et al., 2022). figure 8 exhibits the hydroxide conductivity of eqpva-x aems. the eqpva-x aems demonstrate oh− conductivities in the range of 0.71–2.26 ms cm−1 at 30 °c. generally, the aems with the highest iec and wu also possess the highest hydroxide conductivity, which can also be observed in this work. the eqpva-5.0 exhibits the highest value of wu and iec, which also demonstrates the highest oh− conductivity of 2.26 ms cm−1. the grotthuss mechanism describes how the hydroxide ions migrate along the water molecules chain via the formation and breaking of hydrogen bonds (li et al., 2020). accordingly, as wu increases, water content rises as well, improving the conductivity of hydroxide ions. a.m.samsudin et al int. j. renew. energy dev 2023, 12(2), 375-380 |379 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 8 hydroxide conductivity of eqpva-x aems. high hydroxide conductivity has a favorable impact on power density and is accountable for reduced power/ohmic losses (cermenek et al., 2018). the higher the ionic conductivity, the more hydroxide ions are transported through the membrane so that more hydroxide ions react with the fuel to produce more electrons. the increase in electrons leads to an increase in the current density, which follows equation 5 (kang & cannon, 2015). 𝐼𝐼 = 𝜎𝜎𝜎𝜎 (5) where i is current density, σ is conductivity and e is electric field. since power density (p) is the multiplication of current density (i) and voltage (v) according to equation 6, an increase in current density will give a proportional increase in power density (o’hayre, 2017). 𝑃𝑃 = 𝑉𝑉𝐼𝐼 (6) this is consistent with a study conducted by samsudin et al. (2021, 2022) that membranes of different conductivity will produce different power densities in the same fuel cell and operating conditions. 4. conclusion anion exchange membranes (aems) composed of gohsenxtm k-434 quaternary ammonium poly(vinyl alcohol) (pva) as material for electrospun fiber mats and inter-fiber void filler have been prepared by the electrospinning method. ftir spectra recognized the primary functional groups of membranes. sem images display the electrospun nanofibers structures of eqpva with a mean size of 100.96 nm and the surface morphology of eqpva-5.0 dense aems. by incorporating the qpva matrix of 5 wt.% concentration, the eqpva-5.0 membrane achieved the highest iec, water uptake, swelling degree, and hydroxide conductivity of 0.82 mmol g−1, 25.5%, 19.9%, and 2.26 ms cm−1, respectively. acknowledgments the authors thank the austrian science fund (fwf) for providing financial support for the study under project number i 3871-n37. additionally, the authors acknowledge kemdikbud (indonesia) and oead (austria) for the iasp scholarship. author contributions: a.m.s.: conceptualization, methodology, formal analysis, writing—original draft, writing—review and editing, v.h.; writing—review and editing, supervision, resources, project administration. all authors have read and agreed to the published version of the manuscript. funding: this research was funded by austrian science fund (fwf) under project number i 3871-n37. conflicts of interest: the authors declare no conflict of interest. references aslam, m., kalyar, m. a., & raza, z. a. (2018). polyvinyl alcohol: a review of research status and use of polyvinyl alcohol based nanocomposites. polymer engineering and science, 58(12), 2119– 2132. https://doi.org/10.1002/pen.24855 ayaz, s., yao, z. y., chen, y. j., & yu, h. y. (2022). preparation of poly(arylene ether ketone) based anion exchange membrane with pendant pyrimidinium and pyridazinium cation derivatives for alkaline fuel cell. journal of membrane science, 659(june), 120778. https://doi.org/10.1016/j.memsci.2022.120778 becerra-arciniegas, r. a., narducci, r., ercolani, g., antonaroli, s., sgreccia, e., pasquini, l., knauth, p., & di vona, m. l. (2019). alkaline stability of model anion exchange membranes based on poly(phenylene oxide) (ppo) with grafted quaternary ammonium groups: influence of the functionalization route. polymer, 185, 121931. https://doi.org/10.1016/j.polymer.2019.121931 cermenek, b., ranninger, j., & hacker, v. (2018). alkaline direct ethanol fuel cell. in ethanol: science and engineering (pp. 383– 405). elsevier. https://doi.org/10.1016/b978-0-12-8114582.00015-8 das, g., choi, j.-h., nguyen, p. k. t., kim, d.-j., & yoon, y. s. (2022). anion exchange membranes for fuel cell application: a review. polymers, 14(6), 1197. https://doi.org/10.3390/polym14061197 ding, c., & qiao, z. (2022). a review of the application of polyvinyl alcohol membranes for fuel cells. ionics, 28(1), 1–13. https://doi.org/10.1007/s11581-021-04338-w du, s., li, s., xie, n., xu, y., weng, q., ning, x., chen, p., chen, x., & an, z. (2022). development of rigid side-chain poly(ether sulfone)s based anion exchange membrane with multiple annular quaternary ammonium ion groups for fuel cells. polymer, 251(april), 124919. https://doi.org/10.1016/j.polymer.2022.124919 du, x., zhang, h., yuan, y., & wang, z. (2020). semi-interpenetrating network anion exchange membranes based on quaternized polyvinyl alcohol/poly(diallyldimethylammonium chloride). green energy and environment. 6(5), 743-750 https://doi.org/10.1016/j.gee.2020.06.015 elumalai, v., ganesh, t., selvakumar, c., & sangeetha, d. (2018). phosphonate ionic liquid immobilised sba-15/speek composite membranes for high temperature proton exchange membrane fuel cells. materials science for energy technologies. https://doi.org/10.1016/j.mset.2018.08.003 feketefoldi, b., & cermenek, b. (2016). chitosan-based anion exchange membranes for direct ethanol fuel cells. journal of membrane science & technology, 06(01), 1–9. https://doi.org/10.4172/2155-9589.1000145 fennessey, s. f., & farris, r. j. (2004). fabrication of aligned and molecularly oriented electrospun polyacrylonitrile nanofibers and the mechanical behavior of their twisted yarns. polymer, 45, 4217–4225. https://doi.org/10.1016/j.polymer.2004.04.001 francia, c., ijeri, v. s., specchia, s., & spinelli, p. (2011). estimation of hydrogen crossover through nafion® membranes in pemfcs. journal of power sources, 196(4), 1833–1839. https://doi.org/10.1016/j.jpowsour.2010.09.058 gong, y., liao, x., xu, j., chen, d., & zhang, h. (2016). novel anionconducting interpenetrating polymer network of quaternized polysulfone and poly(vinyl alcohol) for alkaline fuel cells. international journal of hydrogen energy, 41(13), 5816–5823. https://doi.org/10.1016/j.ijhydene.2016.02.037 guo, m., ban, t., wang, y., wang, y., zhang, y., zhang, j., & zhu, x. (2022). exploring highly soluble ether-free polybenzimidazole as anion exchange membranes with long term durability. journal of membrane science, 647(january), 120299. https://doi.org/10.1016/j.memsci.2022.120299 hagesteijn, k. f. l., jiang, s., & ladewig, b. p. (2018). a review of the synthesis and characterization of anion exchange membranes. journal of materials science. 53, 11131–11150; https://doi.org/10.1007/s10853-018-2409-y https://doi.org/10.1002/pen.24855 https://doi.org/10.1016/j.memsci.2022.120778 https://doi.org/10.1016/j.polymer.2019.121931 https://doi.org/10.3390/polym14061197 https://doi.org/10.1007/s11581-021-04338-w https://doi.org/10.1016/j.polymer.2022.124919 https://doi.org/10.1016/j.gee.2020.06.015 https://doi.org/10.1016/j.mset.2018.08.003 https://doi.org/10.4172/2155-9589.1000145 https://doi.org/10.1016/j.polymer.2004.04.001 https://doi.org/10.1016/j.jpowsour.2010.09.058 https://doi.org/10.1016/j.ijhydene.2016.02.037 https://doi.org/10.1016/j.memsci.2022.120299 a.m.samsudin et al int. j. renew. energy dev 2023, 12(2), 375-380 |380 issn: 2252-4940/© 2023. the author(s). published by cbiore huang, j., yu, z., tang, j., wang, p., tan, q., wang, j., & lei, x. (2022). a review on anion exchange membranes for fuel cells: anionexchange polyelectrolytes and synthesis strategies. international journal of hydrogen energy, 47(65), 27800–27820. https://doi.org/10.1016/j.ijhydene.2022.06.140 huang, t., qiu, x., zhang, j., li, x., pei, y., jiang, h., yue, r., yin, y., jiang, z., zhang, x., & guiver, m. d. (2022). hydrogen crossover through microporous anion exchange membranes for fuel cells. journal of power sources, 527, 231143. https://doi.org/10.1016/j.jpowsour.2022.231143 inaba, m., kinumoto, t., kiriake, m., umebayashi, r., tasaka, a., & ogumi, z. (2006). gas crossover and membrane degradation in polymer electrolyte fuel cells. electrochimica acta, 51(26), 5746– 5753. https://doi.org/10.1016/j.electacta.2006.03.008 iravaninia, m., & rowshanzamir, s. (2015). polysulfone-based anion exchange membranes for potential application in solid alkaline fuel cells. journal of renewable energy and environment, 2(2), 59– 65. https://doi.org/10.30501/jree.2015.70071 kang, w., & cannon, j. l. (2015). a membrane-based electro-separation method (mbes) for sample clean-up and norovirus concentration. plos one, 10(10), 1–22. https://doi.org/10.1371/journal.pone.0141484 kumar, p., bharti, r. p., kumar, v., & kundu, p. p. (2018). polymer electrolyte membranes for microbial fuel cells: part a. nafionbased membranes. in progress and recent trends in microbial fuel cells (pp. 47–72). elsevier b.v. https://doi.org/10.1016/b978-0444-64017-8.00004-x li, y., li, m., zhou, s., xue, a., zhang, y., zhao, y., zhong, j., zhang, q., & yang, d. (2020). enhancement of hydroxide conductivity by incorporating nanofiber-like palygorskite into quaternized polysulfone as anion exchange membranes. applied clay science, 195(june), 105702. https://doi.org/10.1016/j.clay.2020.105702 mayadevi, t. s., min, k., choi, o., chae, j. e., kim, h. j., choi, c. h., kang, h., park, c. h., & kim, t. h. (2022). ppos having piperidinium-based conducting head groups with extra molecular interaction sites as new anion exchange membranes. international journal of hydrogen energy, 47(36), 16222–16234. https://doi.org/10.1016/j.ijhydene.2022.03.110 movil, o., frank, l., & staser, j. a. (2015). graphene oxide-polymer nanocomposite anion-exchange membranes. journal of the electrochemical society, 162(4), f419–f426. https://doi.org/10.1149/2.0681504jes o’hayre, r. p. (2017). fuel cells for electrochemical energy conversion. epj web of conferences, 148, 1–16. https://doi.org/10.1051/epjconf/201714800013 patel, a., patra, f., shah, n., & khedkar, c. (2018). application of nanotechnology in the food industry: present status and future prospects. in impact of nanoscience in the food industry (pp. 1– 27). elsevier inc. https://doi.org/10.1016/b978-0-12-8114414.00001-7 ramaswamy, n., & mukerjee, s. (2020). alkaline anion-exchange membrane fuel cells : challenges in electrocatalysis and interfacial charge transfer. chem. rev., 119, 11945–11979. https://doi.org/10.1021/acs.chemrev.9b00157 samsudin, a. m., bodner, m., & hacker, v. (2022). a brief review of poly ( vinyl alcohol ) -based anion exchange membranes for alkaline fuel cells. polymer, 14, 3565. https://doi.org/10.3390/polym14173565 samsudin, a. m., & hacker, v. (2019). preparation and characterization of pva/pdda/nano-zirconia composite anion exchange membranes for fuel cells. polymers, 11, 1399. https://doi.org/10.3390/polym11091399 samsudin, a. m., & hacker, v. (2021). effect of crosslinking on the properties of qpva/pdda anion exchange membranes for fuel cells application. journal of the electrochemical society, 168, 044526. https://doi.org/10.1149/1945-7111/abf781 samsudin, a. m., roschger, m., wolf, s., & hacker, v. (2022). preparation and characterization of qpva/pdda electrospun nanofiber anion exchange membranes for alkaline fuel cells. nanomaterials, 12(22), 3965. https://doi.org/10.3390/nano12223965 samsudin, a. m., wolf, s., roschger, m., & hacker, v. (2021). poly(vinyl alcohol)-based anion exchange membranes for alkaline polymer electrolyte fuel cells. international journal of renewable energy development, 10(3), 435–443. https://doi.org/10.14710/ijred.2021.33168 sood, r., cavaliere, s., jones, d. j., & rozière, j. (2016). electrospun nanofibre composite polymer electrolyte fuel cell and electrolysis membranes. nano energy, 26, 729–745. https://doi.org/10.1016/j.nanoen.2016.06.027 susanto, h., samsudin, a. m., faz, m. w., & rani, m. p. h. (2016). impact of post-treatment on the characteristics of electrospun poly (vinyl alcohol)/chitosan nanofibers. in h. susanto, r. suryana, & k. triyana (eds.), aip conference proceedings (vol. 1725, p. 020087). aip publishing llc. https://doi.org/10.1063/1.4945541 tamura, t., & kawakami, h. (2010). aligned electrospun nanofiber composite membranes for fuel cell electrolytes. nano letters, 10, 1324–1328. https://doi.org/10.1021/nl1007079 tomasino, e., mukherjee, b., ataollahi, n., & scardi, p. (2022). water uptake in an anion exchange membrane based on polyamine: a first-principles study. the journal of physical chemistry b, 126(38), 7418–7428. https://doi.org/10.1021/acs.jpcb.2c04115 vandiver, m. a., caire, b. r., carver, j. r., waldrop, k., hibbs, m. r., varcoe, j. r., herring, a. m., & liberatore, m. w. (2014). mechanical characterization of anion exchange membranes by extensional rheology under controlled hydration. journal of the electrochemical society, 161(10), h677–h683. https://doi.org/10.1149/2.0971410jes wang, y.-j., qiao, j., baker, r., & zhang, j. (2013). alkaline polymer electrolyte membranes for fuel cell applications. chemical society reviews, 42(13), 5768–5787. https://doi.org/10.1039/c3cs60053j wang, z., zhou, s. f., zhuo, y. z., lai, a. n., lu, y. z., & wu, x. bin. (2022). adamantane-based block poly(arylene ether sulfone)s as anion exchange membranes. polymer, 255(july), 125155. https://doi.org/10.1016/j.polymer.2022.125155 yang, j. m., fan, c. s., wang, n. c., & chang, y. h. (2018). evaluation of membrane preparation method on the performance of alkaline polymer electrolyte: comparison between poly(vinyl alcohol)/chitosan blended membrane and poly(vinyl alcohol)/chitosan electrospun nanofiber composite membranes. electrochimica acta, 266, 332–340. https://doi.org/10.1016/j.electacta.2018.02.043 zelovich, t., vogt-maranto, l., hickner, m. a., paddison, s. j., bae, c., dekel, d. r., & tuckerman, m. e. (2019). hydroxide ion diffusion in anion-exchange membranes at low hydration: insights from ab initio molecular dynamics. chemistry of materials, 31(15), 5778–5787. https://doi.org/10.1021/acs.chemmater.9b01824 zhang, g., li, r., wang, x., chen, x., shen, y., & fu, y. (2022). the inhibiting water uptake mechanism of main-chain type nspirocyclic quaternary ammonium ionene blended with polybenzimidazole as anion exchange membrane. separation and purification technology, 291(january), 120950. https://doi.org/10.1016/j.seppur.2022.120950 zhang, j., qiao, j., jiang, g., liu, l., & liu, y. (2013). cross-linked poly(vinyl alcohol)/poly (diallyldimethylammonium chloride) as anion-exchange membrane for fuel cell applications. journal of power sources, 240, 359–367. https://doi.org/10.1016/j.jpowsour.2013.03.162 zheng, y., ash, u., pandey, r. p., ozioko, a. g., ponce-gonzález, j., handl, m., weissbach, t., varcoe, j. r., holdcroft, s., liberatore, m. w., hiesgen, r., & dekel, d. r. (2018). water uptake study of anion exchange membranes. macromolecules, 51(9), 3264– 3278. https://doi.org/10.1021/acs.macromol.8b00034 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1016/j.ijhydene.2022.06.140 https://doi.org/10.1016/j.jpowsour.2022.231143 https://doi.org/10.1016/j.electacta.2006.03.008 https://doi.org/10.30501/jree.2015.70071 https://doi.org/10.1371/journal.pone.0141484 https://doi.org/10.1016/b978-0-444-64017-8.00004-x https://doi.org/10.1016/b978-0-444-64017-8.00004-x https://doi.org/10.1016/j.clay.2020.105702 https://doi.org/10.1016/j.ijhydene.2022.03.110 https://doi.org/10.1149/2.0681504jes https://doi.org/10.1051/epjconf/201714800013 https://doi.org/10.1016/b978-0-12-811441-4.00001-7 https://doi.org/10.1016/b978-0-12-811441-4.00001-7 https://doi.org/10.1021/acs.chemrev.9b00157 https://doi.org/10.3390/polym14173565 https://doi.org/10.3390/polym11091399 https://doi.org/10.1149/1945-7111/abf781 https://doi.org/10.3390/nano12223965 https://doi.org/10.14710/ijred.2021.33168 https://doi.org/10.1016/j.nanoen.2016.06.027 https://doi.org/10.1063/1.4945541 https://doi.org/10.1021/nl1007079 https://doi.org/10.1021/acs.jpcb.2c04115 https://doi.org/10.1149/2.0971410jes https://doi.org/10.1039/c3cs60053j https://doi.org/10.1016/j.polymer.2022.125155 https://doi.org/10.1016/j.electacta.2018.02.043 https://doi.org/10.1021/acs.chemmater.9b01824 https://doi.org/10.1016/j.seppur.2022.120950 https://doi.org/10.1016/j.jpowsour.2013.03.162 https://doi.org/10.1021/acs.macromol.8b00034 qpva-based electrospun anion exchange membrane for fuel cells asep muhamad samsudin a,b0f and viktor hackera 1. introduction microsoft word omondi et al.docx int. journal of renewable energy development 8 (3) 2019: 253-259 p a g e | © ijred – issn: 2252-4940.all rights reserved 253 contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: http://ejournal.undip.ac.id/index.php/ijred anaerobic co-digestion of water hyacinth (e. crassipes) with ruminal slaughterhouse waste for biogas production erick auma omondi1*, peter gikuma-njuru2, and peter kuria ndiba1 1 department of civil and construction engineering, university of nairobi; p.o. box 10344-00100 nairobi, kenya. 2 department of environmental science & land resources management, south eastern kenya university; p.o. box 17090200 kitui, kenya. abstract. the use of biomass as renewable energy source is of interest in reducing dependence on fossil fuels and associated impacts of climate change. water hyacinth (wh), an invasive aquatic plant of environmental concern has large biomass that is available for biogas production. co-digestion of this largely lignocellulose biomass with other substrates may correlate process parameters and improve biogas production. this study evaluated co-digestion of wh biomass with various mix proportions of ruminal slaughterhouse waste (rsw) at 24, 32 and 37°c in order to assess the optimum proportion and temperature. the rate of biomethanation increased with temperature from 0.23 at 24ºc to 0.75 and 0.96 at 32ºc and 37ºc, respectively, and similarly methane yield improved from 14 at 24ºc to 40 and 52 l/kg air dried water hyacinth at 32ºc and 37ºc respectively. a wh: rsw ratio of 30% showed optimum acclimatization and methane yield in a residence time of 60 days. the duration of the initial drop in ph that indicates hydrolysis stage decreased with increase in proportion of rsw, indicating faster hydrolysis and fermentation processes. longer and stable latter alkaline ph zone suggested improved biomethanation and greater biogas production. co-digestion with 30% rsw at 24ºc improved biogas yield by 75% from 8.05 to 14.09l/kg biomass, methane component of biogas by 9% from 59 to 68% and reduced the retention time for substrate by 36%, suggesting synergy in co-digestion with respect to biogas quality. changing the temperature from 24 to 32ºc increased the yield by 186% and reduced retention time by 73%. the results demonstrated synergy in co-digestion of the two substrates and the process dynamics that are useful in a possible process commercialization. ©2019. cbiore-ijred. all rights reserved keywords: co-digestion, biomass, biogas, water hyacinth, c/n ratio, ruminal slaughterhouse waste. article history: received: july 19, 2019; revised: sept 22, 2019; accepted: oct 21, 2019; available online: oct 30, 2019 how to cite this article: omondi, e.a., njuru, p.g. and ndiba, p.k. (2019). anaerobic co-digestion of water hyacinth (e. crassipes) with ruminal slaughterhouse waste for biogas production. international journal of renewable energy development, 8(3), 253-259 https://doi.org/10.14710/ijred.8.3.253-259 1. introduction the use of fossil fuels is increasingly expensive and poses serious health and environmental concerns especially climate change (budiyano et al., 2010). accordingly, biomass is increasingly of interest as a source of renewable energy. short-cycle crops are the most commonly used source of biomass for energy production; however, the use of crops faces the challenge of competing demands for arable land (svetlana and johan, 2010). water hyacinth, an invasive aquatic plant with short doubling times of 7–12 days (reddy and debusk 1985; tag el-din 1992) grows on water and, therefore, does not compete for agricultural land with crops (bett, 2012). o’sullivan et al. (2010) obtained biogas production in the range 200-400 l biogas kg-1 volatile solids (vs). water hyacinth biomass has relatively high carbon to nitrogen ratio, a characteristic desired in substrates for biogas production (subhabrata et al., 2013, omondi et al., 2019). however, the lignocellulose nature of water hyacinth may * corresponding author: omorric@gmail.com (tel: +254 723 631 512) slow down hydrolysis process and conversion to biogas (yadviva et al. 2004). the intricate structure of lignocellulose (bajpai 2016) can limit microbial degradation and result in slow digestion and reduced biogas yield (li, 2015). techniques that are available for improvement of bio digestion include using different pretreatment methods (ofuofule et al, 2009); optimization of dilution on biomethanation of fresh water hyacinth (patil et al, 2011) and effects of particle size, plant nitrogen content and inoculum volume. a simple and inexpensive technology for enhancing microbial degradation of the biomass is correlating process parameters, for example, by co-digestion with other substrates (callaghan et al, 1999; kumar and sharma, 2017). studies have shown that synergies in simultaneous processing of substrates through co-digestion result in better performance than with individual substrates (e.g. li et al. 2011; rao and baral 2011; dias et al. 2014). codigestion has numerous advantages for microbial digestion that include reduced concentration of toxic research article citation: omondi, e.a., njuru, p.g. and ndiba, p.k. (2019). anaerobic co-digestion of water hyacinth (e. crassipes) with ruminal slaughterhouse waste for biogas production. int. journal of renewable energy development, 8(3), 253-259, doi.org/10.14710/ijred.8.3.253-259 p a g e | © ijred – issn: 2252-4940. all rights reserved 254 compounds, increased nutrients concentration, improved substrate loading, supply of buffer capacity and hygienic stabilization of enzymes (tufaner and avsar, 2016). these benefits are important for stability and performance of the anaerobic process (esposito et al. 2012). consequently, codigestion has potential for reduced hydraulic retention times and increased biogas yield. in some previous studies, food waste and cattle manure in the ratio of 2:1 has been found to enhance methane yield by 41.1% and 55.2% corresponding to 388 mlg-1vs and 317 mlg-1vs in batch and semi-continuous reactors respectively (zhang et al., 2012). earnest and singh (2013) observed that codigestion of fruit and vegetable wastes with cow dung in the ratio of 1:1 and 1:2 yielded 245 and 230 ml of biogas respectively and gomez et al., (2006) found that biomethanation potential of primary sludge and vegetable fraction of municipal garbage under mesophilic conditions resulted in biogas yield of 0.60 – 0.80 lg-1vs compared to 0.4 – 0.6 lg-1 vs for co-digestion and primary sludge alone respectively. co-digestion of cattle manure and organic kitchen wastes in the ratios of 1:3, 1:1 and 3:1 was found to enhance biogas yield from 24.12 to 47.13% while improving cumulative biogas yield by 1.01 – 1.84 times (aragaw and andargie, 2013). slaughterhouse waste has significant concentration of nutrients that can complement the digestion of other substrates such as water hyacinth (wei wu, 2010, omondi et al., 2019). however, most of the slaughterhouse waste components with the exception of ruminal waste, have large concentration of proteins, which make them susceptible to ammonia toxicity (callaghan et al., 2002; edstrom et al., 2 0 0 3 ; cuetos et al., 2010, chen et al., 2008). similarly, volatile fatty acids (vfas) tend to accumulate in the reactors causing progressive drops in ph that stress and inhibit the activity of methanogenic archaea (siegert and banks, 2005). rumen contents which have limited protein concentration and occur in the largest proportion in the waste would therefore be the desirable component for co-digestion with water hyacinth. furthermore, rumen waste contains cellulolytic anaerobic bacteria that are suitable as inoculum for degradation of cellulose (aurora, 1983; castillo, 1995). this study evaluated synergy in co-digestion of water hyacinth with ruminal slaughterhouse waste in biogas production. 2. materials and methods 2.1 overview of methods the study investigated biogas production in co-digestion of water hyacinth (wh) from lake victoria with ruminal slaughterhouse waste (rsw). the co-digestion was conducted in batch digesters while biogas output was measured by displacement method. 2.2 sample collection and preparation water hyacinth used was obtained from the shores of winam gulf, lake victoria, in kisumu city at coordinates 0° 5’39.71”s, 34045’2.44”e while ruminal slaughterhouse waste was collected from nairobi’s dagoreti slaughterhouse located at coordinates 1°17'3.71"s, 36°41'1.98"e (figure 1). fresh and healthy mature water hyacinth plants were obtained and packed in sampling bags and transported, within 12 hours, to the university of nairobi’s environmental engineering laboratory, where they were stored in a cold room. approximately 4kg wet samples of fresh ruminal slaughterhouse were placed in sampling bags and transported immediately to the laboratory, where they were stored in a cold room at 4ºc until processing for study. approximately, 5kg of whole water hyacinth plants including leaves, stems and roots, were cut to small sizes of about 2 cm. approximately 50 g was kept for determination of total water content while the rest was dried under the sun for a period of 7 days. the sun-dried water hyacinth was ground to fine particles using a mortar and pestle, placed in plastic bags and stored in a refrigerator. approximately 50 g of the fresh slaughterhouse waste sample was kept for determination of total moisture content while the remaining portion was dried in the sun for a period of 3 days to improve handling and ease storage. the sun-dried samples were kept in plastic bags and stored for biogas production. figure 1. map showing location of (a) wh sampling point in winam gulf, kisumu, kenya (b)slaughter house waste sampling point, dagoreti, nairobi (from: omondi et al. 2019)) 2.3 experimental set-up the experimental setup consisted of eight sets of three round bottom 1,000 ml flasks and a graduated measuring cylinder (figure 2). all the flasks were fitted with tight fitting rubber cocks for airtightness. the first flask was used as the reactor for anaerobic digestion. the reactor was fitted with a thermometer and a ph meter, hi98103 checker ph tester from hanna instruments, for monitoring temperature and ph respectively. a balloon with a needle inserted into reactor headspace was set up to sample gas for characterization. the second flask contained a scrubber solution for co2 and other minor gases, comprising of an alkaline solution prepared using 1 molar sodium hydroxide solution, prepared by dissolving 40 g sodium hydroxide in 1 l of water. three drops of phenolphthalein indicator were added for monitoring ph variation in the solution. the scrubber solution was replaced when the pink/violet colour of the indicator turned colourless. the change in colour is associated with a drop in ph below 8.2. the third flask was for gas int. journal of renewable energy development 8 (3) 2019: 253-259 p a g e | © ijred – issn: 2252-4940.all rights reserved 255 displacement of water for measurement of the volume of gas produced. water in the displacement bottle was charged with a few drops of methyl orange to make it easier to read the volume in the graduated cylinder. the bottle was kept covered with an aluminium foil to minimize loss of water by evaporation. figure 2. biogas production set up (modified from: omondi et al., 2019) 2.4 anaerobic digestion and biogas production substrate for bio-digestion were prepared by mixing 150 g of wh and rsw in different proportion with 500 ml of water in 1000 ml round bottom reactor flasks; a total of eight reactor flasks labelled d1 to d8. the mix proportions used are shown in table 1. the reactors were tightly sealed using rubber cocks and kept airtight to operate under anaerobic digestion mode for a residence time of 60 days. the biogas generated was passed through the scrubber solution. the volume of resultant methane gas was measured through water displacement method into the graduated measuring cylinder (esposito et al., 2012). the cumulative volume of methane generated, ph and temperature were recorded daily at 9 am. room temperature was also recorded throughout the test. table 1. mix proportions of dried substrates digester water hyacinth (g) slaughterhouse waste (g) percent of cosubstrate (%) d1 150 nil 0 d2 142.5 7.5 5 d3 135 15 10 d4 127.5 22.5 15 d5 120 30 20 d6 105 45 30 d7 75 75 50 d8 0 150 100 gas for characterization was sampled in balloons through a needle in the headspace. gas composition was determined, in triplicate for each parameter, using a gas chromatograph fitted with flame indication detector (gcfid) (sugumaran et al., 2014). the reactors were operated at three different temperatures, room temperature of about 24ºc, 32 and 37ºc. 2.5 biogas characterization the quality of biogas depends mainly on the presence of methane in it where a good quality biogas has high percentage of methane and is therefore desirable for maximum energy production. the percentage of methane in biogas is generally determined by the orsat apparatus, gas chromatograph etc. (holman, 1995). the percentage of methane ch4 can be estimated through recognition of co2 percentage from equation 2: ch4 = 100% [co2% + 0.2% h2s] ……… vol. % (konstandt 1976) (1) in this estimation, methane content is measured by absorption of carbon dioxide with10%, 33% and 40% of koh (habel-hadi, 2008) respectively. the assumption by using this method is that biogas is mainly constituted of methane and carbon dioxide gas, where the other gases produced during anaerobic process are neglected. gas chromatography (gc) is an optimal analytical instrument for the analysis of components such as ch4, co2, h2s and siloxanes which are present in the gas (anderson et al., 2010).this study adopted gc method in analyzing produced biogas. 3. results and discussions 3.1 variations of ph with duration of co-digestion the various stages of anaerobic digestion take place at different ph and hence the ph of the digesting substrates can give an indication of the dominant digestion stage at any time and its duration. generally, the first digestion stage, hydrolysis of lipids and protein to volatile fatty acids and amino acids, resulted in a drop in ph while the onset of acidogenesis stage resulted in rise in ph due to production of co2 and nh3 and the associated co3hnh4 (e.g. malakahmad et al., 2012). further rise in ph occurred in the predominantly methanogenesis third stage because of ceased hydrolysis of volatile fatty acid and continued production of co3hnh4. the hydrolysis stage for rsw mix proportion of less than 15% had a ph less than 6.2 (figure 3a) but increased to 6.8 to 7.5 for the rsw proportions of 20 -100%. varying the rsw proportion from 15 to 20% resulted in greatest increase in the hydrolysis ph from about 6.2 to 6.8. moreover, the reduction of duration of the hydrolysis stage with increasing rsw proportion, from 33 days for 5% rsw to 25 days for 50%, maybe an indication of prolonged acidogenesis and methanogenesis stages with codigestion. the observation correlates with previous studies (e.g. feng et al., 2009). for the 30% rsw proportion, digestion at different temperatures (24, 32 and 37ºc) showed varied changes in ph with time (figure 3b). after the seventh day, when ph was similar for all the reactor temperatures, there was a clear pattern of higher increase of ph with temperature, which may be a result of increased biological activity. 3.2 biogas production for various substrates mix proportion in figure 4, variations in cumulative biogas production over 60 day for reactors with different proportions of wh and rsw operated at room temperature are presented. during the first 7 days, all the mix proportions except 50% and 100% showed some increase in cumulative biogas production. this indicates excellent but acclimatization of wh and wh with low portions of rsw. this may be attributed to high volatile solids originally present in wh biomass which leads to volatile organic acids produced citation: omondi, e.a., njuru, p.g. and ndiba, p.k. (2019). anaerobic co-digestion of water hyacinth (e. crassipes) with ruminal slaughterhouse waste for biogas production. int. journal of renewable energy development, 8(3), 253-259, doi.org/10.14710/ijred.8.3.253-259 p a g e | © ijred – issn: 2252-4940. all rights reserved 256 during hydrolysis of the substrate that tend to reduce the ph, an effect that is counteracted by destruction of the volatile acids and reformation of bicarbonate buffer during methane formation. the 50% and 100% showed a lag in the initial days, with no or minimal production of biogas which may also be attributed to low volatile solids originally present in rsw biomass which leads to slow acclimatization but quick hydrolysis of the substrate characterized by minimal ph drop with methane formers quickly outpacing the acid formers in the leading to a stable biogas yield. achieving a balanced condition requires careful co-digestion to overcome the low growth rate of methane bacteria and achieve a stable ad process (kugelman, 1971). in this study, the largest biogas cumulative yield was observed for slaughterhouse waste alone (100% rsw) (17.8 l ch4/kg substrate) followed by 50 and 30% rsw while the smallest yield was for water hyacinth alone (0% rsw) at 8 l ch4/kg substrate. figure 3. ph variations for (a) different rsw substrate proportions (d1=0%, d2=5%, d3=10%, d4=15%, d5=20%, d6=30%, d7=50%, d8=100%) at 24°c; and (b) 30% rsw mix proportion at 24, 32 and 37ºc 3.3 effect of co-digestion on retention time the effect of co-digestion on retention time (rt) was determined by relating the time it takes to produce equivalent volume of methane for wh alone (0%rsw) at 60 days. thus, the time it takes to produce 8l/kg methane was determined for various mix proportions in codigestion (fig 4). from the results, co-digestion reduced the retention time by 9, 15, 18, 20, 22 and 26 days for 5, 15, 10, 20, 30 and 50% rsw respectively. the results indicate that, co-digesting wh biogas plant with 30% rsw for example will reduce rt by 22 days. therefore, codigestion of wh with 5-50%rsw has significant reduction on rt. co-digestion with proportions greater than 50% rsw will have no further impact on rt as the reduction in rt for rsw alone coincides with that of 50% rsw. figure 4. biogas production for various mix proportions at 24°c 3.4 influence of temperature on biomethenation co-digestion of wh and rsw at 30% rsw proportion at various temperatures; namely, 24, 32 and 37ºc showed the influence of temperature on biomethanation. for the higher temperatures, startup time reduced from three days to one day. the rate of biomethanation improved from 0.23 to 0.75 and 0.96 at 32 and 37ºc respectively. increasing the temperature from 24 to 32ºc increased methane yield from 14 to 40 l/kg or 186%, but increasing the operating temperature to 37ºc only increased the yield by a further 30% to 52 l/kg (fig. 5). consequently, increasing the operating temperature from 32 to 37ºc may not be merited unless the cost benefit of the gas production and reduced capital cost surpasses the extra cost of energy. apart from the increase in biogas yield, change in temperature from 24 to 32ºc reduced the retention time by 44 days, from 60 to 16 days (fig. 5). further increase in temperature from 32 to 37ºc only reduced the retention time by 2 more days, a reduction not very significant to justify the temperature increase. 6,0 6,5 7,0 7,5 8,0 8,5 9,0 5 10 15 20 25 30 35 40 45 50 55 60 d1 d2 d3 d4 d5 d6 d7 d8 ph (a) 7,0 7,5 8,0 8,5 9,0 9,5 10,0 0 5 10 15 20 25 30 35 40 45 50 55 60 d6 (24°c) d6 (32°c) d6 (37°c) duration (days) ph (b) 0 2 4 6 8 10 12 14 16 18 20 0 5 10 15 20 25 30 35 40 45 50 55 60 0% 5% 10% 15% 20% 30% duration (days) x1 03 c um ul at iv e c h 4 m l/k g bi om as s int. journal of renewable energy development 8 (3) 2019: 253-259 p a g e | © ijred – issn: 2252-4940.all rights reserved 257 fig 5. cumulative methane gas production for 30%rsw at 24, 32 and 37°c temperature conditions. 3.5 biogas characteristics the percentage compositions of biogas produced at 24ºc for different rsw are presented in figure 6. methane gas proportion increased with increase in rsw mix proportion in the reactor mixture, from 59% for water hyacinth alone (0% rsw) to a maximum value of 68% for 20% and 30% rsw mix proportions and then decreased to a minimum value of 58% for rsw alone (100% rsw). in contrast the composition of co2 in the biogas decreased with increasing rsw mix proportion in the reactor mixture from 39% for water hyacinth alone (0% rsw) to about 30% for 15%, 20% and 30% rsw mix proportions before increasing to 58% for rsw alone reactor. the proportion of trace gases in the biogas was highest for the rsw alone reactor (4%), approximately double the amount in the other reactor mixtures (2%). consequently, the codigestion of wh with rsw improved the quality of biogas compared to digestion of either substrate alone. this demonstrates a synergy in improvement of biogas quality through co-digestion of the two substrates. figure 6. percentage proportions of methane, carbon dioxide and trace gases in the biogas produced for different co-digestion mix proportions operated at 24°c 4. conclusion this study found that co-digestion of water hyacinth with slaughterhouse increased biogas production by between 25 and 66% for 5 to 20% rsw mix proportion and resulted in 8% improvement in biogas quality. a rsw of 20% produced the best weighted return per unit rsw used. codigestion of wh with rsw reduces the ph fluctuations during the hydrolysis and in turn increases acclimatization and biogas yield varying temperature from 24 to 32ºc had a significant impact in the biogas yield for co-digested wh and rsw biomass at 30% rsw. further increase in temperature from 32 to 37ºc demonstrated insignificant increase in biogas yield and impact on retention time. this study therefore recommends co-digestion of wh with 30% rsw at 32ºc. future studies can determine the outcomes between 24 and 32ºc. references abdel-hadi, m. a. (2008). a simple apparatus for biogas quality determination biological engineering misr j. ag. eng., 25(3), 10551066 aliyu s. and zahangir m., (2016). pretreatment methods of organic wastes for biogas production. journal of applied sciences, 16, 124-137. andersen a., seeley j., and aurand j., (2010). the use of gas chromatography for biogas analysis, the american physical society, 2010. available online: http://www.aps.org aragaw t. and andargie m., (2013). co-digestion of cattle manure with organic kitchen waste to increase biogas production using rumen fluid as inoculums, int j phys sci, 8, 443-450 aurora, s.p., (1983). microbial digestion in ruminants, indian council of agricultural research, new delhi. bajpai p., (2016). pretreatment of lignocellulosic biomass for biofuel production. springer briefs in green chemistry for sustainability, doi 10.1007/978-981-10-0687-6_2 bett m., (2012). a review of techniques for management of invasive plant species. university of eldoret. unpublished version. boontian, n. (2014). conditions of the anaerobic digestion of biomass. world academy of science, engineering and technology international journal of environmental and ecological engineering, 8 (9) budiyano, i. n., widiasa, j. and sunarso, s., (2010). increasing biogas production rate from cattle manure using rumen fluid as inoculums. international journal of chemical and basic applied sciences, 10, 68-75. budiyano, i. n., widiasa, j. and sunarso, s., (2010). the kinetic of biogas production rate from cattle manure in batch mode. international journal of chemical and biomolecular engineering, 3(3), 39-44. callaghan j.f, wase a. j. d, thayanithy, k., forster f.c., (1999). co-digestion of waste organic solids: batch studies. bioresource technology, vol. 67, 117-122 do 10.1016/s0960-8524(98)00108-4 callaghan j.f, wase d.a.j, thayanity k., forster c.f., (2002). continuous co-digestion of cattle slurry with fruit and vegetable wastes and chicken manure. biomass bioenergy 22,71–77 castillo, r.t., p.l. luengo, and j.m. alvarez (1995). temperature effect on anaerobic of bedding manure in a one phase system at different inoculums concentration. agriculture, ecosystems and environment, 54:55-66 chen y, cheng jj, creamer ks. (2008). inhibition of anaerobic process, a review. bioresour technol 99,4044–4064 0 10 20 30 40 50 0 5 10 15 20 25 30 35 40 45 50 55 60 (24°c) (32°c) (37°c) duration (days) x1 03 c um ul at iv e c h 4 m l/k g bi om as s 0 10 20 30 40 50 60 70 80 0%rsw 15%rsw 20%rsw 30%rsw 100%rsw % o f t ot al b io ga ss ch4 co2 trace gases citation: omondi, e.a., njuru, p.g. and ndiba, p.k. (2019). anaerobic co-digestion of water hyacinth (e. crassipes) with ruminal slaughterhouse waste for biogas production. int. journal of renewable energy development, 8(3), 253-259, doi.org/10.14710/ijred.8.3.253-259 p a g e | © ijred – issn: 2252-4940. all rights reserved 258 chin k.k. and goh t.n., (1978). bioconversion of solar energy: methaneproduction through water hyacinth. symp energy biomasswaste 78,215–228 cuetos j., gomez x., otero m., moran a., (2010). anaerobic digestion and co-digestion of slaughterhouse waste (shw): influence of heat and pressure pre-treatment in biogas yield waste manage 30,1780–1789 dias t, fragoso r, duarte e., (2014). anaerobic co-digestion of dairy cattle manure and pear waste. bioresour technol 164:420–423. earnest v.p. and singh l.p. (2013). biomethanation of vegetable and fruit waste in co-digestion process. int. j. of emerg. technol. and advanced eng., 3(6), 493-495. edstrom m., a. nordberg, l. thyselius, (2003). anaerobic treatment of animal byproducts from slaughterhouses at laboratory and pilot scale, appl. biochem. biotechnol. 109 127e138. esposito, g., frunzo, l., giordano, a., liotta, f., panico, a., pirozzi, f., (2012). anaeriobic co-digestion of organic wastes. rev environ sci biotechnol doi 10.1007/s11157-012-9277-8. feng, h., hu, l., mahmood, q., fang, c., qiu, c., &shen, d. (2009). effects of temperature and feed strength on a carrier anaerobic baffled reactor treating dilute wastewater. desalination, 239(1), 111-121. gangulya a., blankchard r., wheatley a., kumar p.c., (2015). optimization of process parameters for catalytic conversion of solid bio-waste during thermophilic anaerobic digestion procedia. environmental sciences 35, 763 – 770 gomez x. and cuetos m.j., (2006). anaerobic co-digestion of primary sludge and the fruit and vegetable fraction of the municipal solid wastes: conditions for mixing and evaluation of the organic loading rate. renewable energy, 31(12), 2017-2024. holman, j. p. (1995). experimental methods for engineers, 6th ed. new delhi: tata mcgraw-hill, p. 539-43. juliet b., vasilije m., philip l., (2016). biomass resources and biofuels potential for the production of transportation fuels in nigeria. renewable and sustainable energy reviews 63,172-192 konstandt, h. g. (1976). engineering’s operation and economics of methane gas production. seminar on microbial energy conversion, gottingen, erich goetze verlag, germany kugelman i. j. (1971). toxicity, synergism, and antagonism in anaerobic waste treatment processes. americanstandard, inc., new brunswick, n. j. k. k. chin singapore polytechnic, singapore 2 kumar a. k. and sharma s., (2017). recent updates on different methods of pretreatment of lignocellulosic feedstocks: a review. bioresour bioprocess. 4(1), 7. li j, jha ak, he j, ban q, chang s, wang p., (2011). assessment of the effects of dry anaerobic co-digestion of cow dung with waste water sludge on biogas yield and biodegradability. int j phys sci 6:3679–3688 li sun., (2015). biogas production from lignocellulosic materials. doctoral thesis, swedish university of agricultural sciences, issn 1652-6880. malakahmad, a., zain, s. m., and basri, n. e. a. (2012). biomethanation of kitchen waste and sewage sludge in anaerobic baffled reactor. paper presented at the humanities, science and engineering research (shuser), 2012 ieee symposium on. njogu, p., kinyua, r., muthoni, p. and nemoto, y., (2015). biogas production using water hyacinth (eicchornia crassipes) for electricity generation in kenya. energy and power engineering, 7, 209-216. o’sullivan c., rounsefell b., grinham a., clarke w., udy j., (2010). anaerobic digestion of harvested aquatic weeds: water hyacinth (eichhornia crassipes). cabomba (cabomba caroliniana) and salvinia (salvinia molesta). ecol eng 36:1459–1468 omondi, e.a., ndiba, p.k. and njuru, p.g. (2019). characterization of water hyacinth (e. crassipes) from lake victoria and ruminal slaughterhouse waste as cosubstrates in biogas production sn appl. sci. 1: 848. patil j.h., raj m.a., gavimath c.c., hooli v.r., (2011). a comparative study on anaerobic co-digestion of water hyacinth with poultry litter and cow dung. int j chem sci appl 2(2):148–155 petrell rj, bagnall lo., (1991). hydromechanical properties of water hyacinth mats. aquacult eng 10:133–147 prashant b., suresh s., kumar a. and krishnakumar p. (2016). a review on enhancement of biogas yield by pretreatment and addition of additives matec web of conferences 62 620. rao p.v. and baral s.s., (2011). experimental design of mixture for the anaerobic co-digestion of sewage sludge. chem eng j 172:977–986. reddy k.r. and debusk w.f., (1985). growth characteristics of aquatic macrophytes cultured in nutrient-enriched water. ii. azolla, duckweed, and salvinia. econ bot 39(2):200–208 siegert i., and c. banks., (2005). the effect of volatile fatty acid additions on the anaerobic digestion of cellulose and glucose in batch reactors, process biochem 40:3412– 3418 subhabrata a. amit g., bhattacharya a., dey a., chatterjee p. k., (2013). enzymatic hydrolysis of water hyacinth biomass for the production of ethanol: optimization of driving parameters. indian journal of experimental biology. 51. 556-66. svetlana l. and johan v., (2010). biomass for energy versus food and feed, land use analyses and water supply. swedish university of agricultural sciences department of energy and technology report 022 issn 1654-9406 svetlana l., johan v., (2009). global potential of sustainable biomass for energy. slu, institutionen för energi och teknik, swedish university of agricultural sciences department of energy and technology report 013, issn 1654-9406 tag el-din ar., (1992). utilization of water hyacinth hay in feeding of growing sheep. indian j anim sci 62(10):989– 992 thomas, i. kai, w. juan, c. ixcaraguá, l. vera, b. garabed, a. gerd, b. and irina, s., (2011). comparison of different pretreatment methods for lignocellulosic materials. part i: conversion of rye straw to valuable products. hamburg university of technology, institute of thermal separation processes, eißendorfer straße 38, d-21073 hamburg, germany volume 76, pages 14851496 tufaner f.and avsar y., (2016). effects of co-substrate on biogas production from cattle manure: a review. int. j. environ. sci. technol. (2016) 13:2303–2312 vaidyanathan s., kavadia k. m., shroff k. c., mahajan s. p., (1985). biogas production in batch and semicontinuous digesters using water hyacinth. biotechnol bioeng 27:905–908 wei wu., (2010). anaerobic co-digestion of biomass for methane production: recent research achievements. accessed on. http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1. 1.456.258 yadvika, santosh s., sreekrishnan t. r., kohli s., rana v. (2004) enhancement of biogas production from solid substrates using different techniques – a review. journal of bioresource technology; 95:1–10. yiqing y., (2016). leading pretreatments for enhancing the degradability of lignocellulosic wastes and the final products. department of biological technology engineering, washington state university, pullman wa usa. environmental technology reviews pages 103-111. int. journal of renewable energy development 8 (3) 2019: 253-259 p a g e | © ijred – issn: 2252-4940.all rights reserved 259 zhang y., banks c.j. and heaven s. (2012). co-digestion of source segregated domestic food waste to improve process stability. bioresour. technol., 114, 168–178. © 2019. this article is an open access article distributed under the terms and conditions of the creative commons attribution (cc by) license (http://creativecommons.org/licenses/by/4.0/) microsoft word singirikonda.docx int. journal of renewable energy development 9 (1) 2020: 43-51 p a g e | © ijred – issn: 2252-4940.all rights reserved 43 contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: http://ejournal.undip.ac.id/index.php/ijred a novel approach using adaptive neuro fuzzy based droop control standalone microgrid in presences of multiple sources srinivas singirikonda and y.p.obulesu* school of electrical engineering, vellore institute of technology, vellore, tamilnadu-632014 abstract. in this paper, a novel q/p droop control strategy for regulating the voltage and frequency in standalone micro grid with multiple renewable sources like solar and wind is presented. the frequency and voltage control strategy is applied to a standalone micro grid with high penetration of intermittent renewable generation system. adaptive neuro-fuzzy logic interface system (anfis) controller is used for frequency and voltage control for renewable generation system. battery energy storage system (bess) is used to generate nominal system frequency instead of using the synchronous generator for frequency control strategy. a synchronous generator is used to maintain the state of charge (soc) of the bess, but it has limited capacity. for voltage control strategy, we proposed reactive power/active power (q/p) droop control to the conventional reactive power controller which provides voltage damping effect. the induced voltage fluctuations are reduced to get nominal output power. the proposed model is tested on different cases and results show that the proposed method is capable of compensating voltage and frequency variations occurring in the micro grid with minimal rated synchronous generator. ©2020. cbiore-ijred. all rights reserved keywords: adaptive neuro-fuzzy interface system (anfis), battery energy storage system (bess), state of charge (soc), frequency control, q/p droop control, standalone microgrid, voltage damping effect, voltage control article history: received: oct 20, 2017; revised: june 12, 2019; accepted: december 22, 2019; available online: february 15,2020 how to cite this article: singirikonda, s and obulesu, y.p. (2020). a novel approach using adaptive neuro fuzzy based droop control standalone microgrid in presences of multiple sources. international journal of renewable energy development,9(1),43-51 https://doi.org/10.14710/ijred.9.1.43-51 1. introduction micro-grids operating in standalone mode are more commonly exposed to variation in voltage and frequency. the grid becomes weaker than a conventional power system due to an isolated system. in recent years, dependence on renewable energy sources like solar and wind generation system (park & yu, 2001; velasco de la fuentev et al. 2013) has increased the instability of grid due its varying input nature. in case of isolated networks diesel based synchronous machines are used for voltage and nominal frequency control. the pattern recognition and decision making are done by mapping input points to the output using fuzzy logic interface. mostly in an isolated power system, the diesel generator based on a synchronous generator which is used to generate nominal system frequency and voltage with the help of adaptive neuro-fuzzy interface system (anfis). the mapping point of an input to the output using fuzzy logic interface provides a basis from which decisions can be made and the patterns discerned. the simulink software system can access the fuzzy logic test system in a block diagram (sumina, erceg and idzotic. 2005). it describes all membership functions, logical operators and if-then rules. this control strategy is applied specially to penetrate the intermittent renewable power generation to control the frequency and voltage for stable operation of * corresponding author: yp.obulesu@vit.ac.in the system. several methods are being examined to support frequency control. the strategies are enabling to dispatch wind power to operate in a similar manner of a conventional power plant. wind power is a fluctuating motive source, the effectiveness of active power control of wind turbine generators (wtg) will depend upon wind speed (keung et al. 2009). wtg's are supplemented with doubly fed induction generator (dfig) to expand the flexibility of wind power procurement and enhance the controllability (ekanayake, holdsworth and jenkins, 2003). the control approach of dfig is to set a point of active power at fixed pitch angle (ekanayake and jenkins. 2004). with the proper reference frame, idr will come up with the electromagnetic torque (tem) against turbines mechanical torque (tmech) at some rotor speed. consequently, the torque difference between tmech and tem can make rotor to accelerate / decelerate. emmech r tt dt dw j -= (1) )( 2 1 e r ttm hdt dw -= (2) where, wr rotor speed, j inertia of motion, h -inertia constant (s). the primary frequency support from de-loaded wind turbines using variable droop was developed (diaz et research article citation: singirikonda, s. and obulesu, y.p. (2020). a novel approach using adaptive neuro fuzzy based droop control standalone microgrid in presences of multiple sources. int. journal of renewable energy development,9(1), 43-51, doi.org/10.14710/ijred.9.1.43-51 p a g e | © ijred – issn: 2252-4940. all rights reserved 44 al. 2010). in micro-grids with high penetration of wind energy, the fluctuations in the wind form output due to variations in wind speed cause frequency disturbances. a frequency droop control was applied to pv power generation (lalor. mullane and o’malley. 2005). even though the fuel cost is free but its cost of installation is high. pv's operate in the maximum power point tracking (mppt) model to generate maximum income. however, as penetration of pv's increase the frequency regulation capability (mainly provided by synchronous generators) and inertia from synchronous generators decrease which leads to severe frequency fluctuations under some disturbances. moreover, load changes can lead to some significant frequency deviations if pvs don’t have frequency regulation capability. to avoid this, the pvs are designed with a virtual governor to have frequency droop characteristics similar to that of the synchronous generator. however, frequency control strategies using intermittent renewable generation are not beneficial economically. there are various ways to control voltage drop by installing regulators in substations, using online transformer tap changers, shunt capacitors, increasing the size of conductors etc. some sensors, such as smart meters (sm) measure voltage in reference and current at each branch send this information/recorded data to the control center (carvalho et al, 2008; schlabbach, blume & stephanblome, 2001). decentralized voltage control is another method which uses local data to control voltage issues. but this device operates independently and there is no communication between loads and the substation. the effect and the reliability must be maintained for these methods. the linear quadratic tracking method is one of the voltage control method used to obtain desired results. the voltage is considered at each node then the controller increases / decreases voltage to minimize the error. the monitoring of controller is based on entire system conditions. this process can be categorized as decentralized control, but it increases system complexity and it needs more study. analyzing and modeling of power distribution would become more complex and time taking. most control strategies have applied optimization algorithms to meet specific objectives, such as minimizing loss, improving voltage profile, mitigating voltage fluctuation, maintaining voltage within regulated limits (kim 2010). however, these methods will never be perfectly accurate, since they are based on forecasting load demand, wind speed and solar irradiance. the voltage compensator, shunt capacitors, lqt methods which lead to increased additional cost. q/v droop control is widely used for voltage compensation, but the compensation is triggered by sensing the voltage deviation. section 2 describes about proposed methodologies and the control strategy which includes q/v droop control and anfis controller are explained in section 3. the test system operation and its simulation results are observed in different cases in section 4. 2.proposed methodology in a remote power system, the active power/frequency (p/f) and reactive power/voltage (q/v) droop control are used to generate nominal system frequency, voltage and some voltage compensation devices are used for control strategy. if the generating system units’ droop is increased, it's response to the system frequency deviation diminishes. however, frequency control strategies using intermittent renewable generation are not beneficial, because they cannot make the most of their ability to utilize free energy. bess is used to support the frequency of micro-grid. the system stability and operational security can be improved by using bess (divya and ostergaard 2009). by improving the controllability of res generators, bess provides a resolution to overcome the frequency control issues. q/v droop control is widely used for voltage compensation, but the compensation is triggered by sensing the voltage deviation. the recommended strategies include 1. bess is used to generate small system frequency instead of using diesel generators which does not depend on the mechanical inertia of a synchronous generator. 2. soc (state of charge) of the bess is used by the diesel generator at a convinced value and the reference significance of the soc is adjusted to limit the output power of the diesel generators to within a permissible range. 3. q/p droop power is added to the renewable generation which has damping effect to avoid voltage fluctuations induced by its active power fluctuations. 4. adaptive neuro-fuzzy logic controller reduces the frequency and voltage fluctuations and improves the system performance. 3. control strategies to maintain frequency and voltage control, there are many strategies to a conventional power plant. a frequency control droop was added to pv generation. but these control strategies are not economically beneficial, since they cannot, maximize their usage of free energy. so, by adopting bess (battery energy storage system) the control strategies enabling to support system frequency deviating from its nominal value (shayeghi, shayanfar and jalili 2009). with the aid of active power/frequency (p/f) and reactive power/voltage (q/v) droop control and voltage compensation devices are applied to the isolated power system. the q/v droop control is widely used for mitigating voltage fluctuations, since the voltage fluctuations are triggered by sensing the voltage deviation. a. test system configuration the proposed control strategy with anfis (adaptive neuro-fuzzy interface system) is tested on the below test system as shown in fig 1, location of loads and power generation system are also indicated. the distance. the ratings of the power generation using different sources considered in this analysis are given in table 1 (rana, singh and mishra, 2017). table 1. power ratings of different sources considered for analysis no name of source rating(mw) 1 diesel generator 14 2 wind generator 9.1 3 pv system 1 4 bess 15 int. journal of renewable energy development 9 (1) 2020: 43-51 p a g e | © ijred – issn: 2252-4940.all rights reserved 45 fig.1 power system configuration block diagram. the nominal system frequency and voltage considered in this study are 50hz and 11kv respectively and the change in load during the day and night are shown in the table 2. the inverters are modeled as two level and the gate signals to the inverter switches are generated using conventional sinusoidal pulse width modulation. table 2 system load demand bus number day(mw) night(mw) 2 3 1.5 3 2 1 5 1 0.5 6 1 0.5 8 2 1 9 0.2 0.1 10 0.2 0.1 11 0.1 0.05 12 0.5 0.25 total 10 5 b. frequency control strategy battery system is used to deliver power at nominal frequency instead of using synchronous machine. as the frequency of the system is depending on the generator speed and inertia bess rather than diesel generator is adopted to overcome this weakness to control the system frequency (scott, wilreker and shaltens, 1984). the bess controls the nominal system frequency by adopting relevant switching mechanism based on the control scheme designed. chargeable characteristic of bess is used to the execute frequency control strategy which enables battery system to take twice the amount of change in load value than any other devices with same rate of power. the rapidly changing charging and discharging abilities of bess make it to respond quickly to the fluctuations in output power of renewable generation system (senjyu, miyazato and yona 2008). however, soc and implementation of control scheme using frequency droop in figure 2 is not achieved by the bess alone. bess should work in synchronization with diesel generator where the frequency, voltage and phase are to be matched. to generate nominal system frequency, the diesel generator should be controlled (serban, marinescu 2011). during normal operation of diesel generator in fig.3, the switch is moved to node a to control the power output diesel generator reducing the discharge from bess and hence maintaining soc at the desired value socref. soc load control is same as the conventional load frequency control and socref is chosen as 0.5pu. the charge and discharge cycle of the bess is controlled by the operator depending on the requirement. fig.2 implementation of the rid-side inverter control scheme fig.3 proposed soc based scheme for control of diesel generator the diesel generator output active power of is maintained within specified value ranging from 0 p.u to 1 p.u by adding the anti-windup function and output limiter value at the output of the pi-controller. the frequency of the system is mostly depending upon the effective implementation of bess control strategy. to overcome the reliability problem that may arise due to tripping action of bess, the switch is connected to node b, when the bess is isolated from the network due to reduction in soc value. during the node b connection of switch, the diesel generator is controlled same as conventional one. the pdi,ref is obtained from the output of the pi controller, it adjusts the valve actuator of diesel engine there by increasing the mechanical input to the synchronous generator, where tv is the valve actuator time constant 0.05s and td diesel engine time constant 0.5s. pdi is the output active power of the diesel generator. the diesel generators acting as auxiliary source for soc control which resembles the frequency control and its priority is changed to control soc of the bess rather than frequency control. c. voltage control strategy the excitation of the diesel generator helps in maintaining its nominal voltage with in the specified limits. in contradictory to frequency control which can be done in central control, the voltage control should be done locally. the control algorithm for grid side inverter control of the wind generator is shown in fig.4. to solve these voltage variations caused by the renewable system, a new q/p droop strategy for voltage control is applied to the distributed generators (s, r, jang 1993). base bus rate qv v v p q k d = (3) citation: singirikonda, s. and obulesu, y.p. (2020). a novel approach using adaptive neuro fuzzy based droop control standalone microgrid in presences of multiple sources. int. journal of renewable energy development,9(1), 43-51, doi.org/10.14710/ijred.9.1.43-51 p a g e | © ijred – issn: 2252-4940. all rights reserved 46 where vid is the d-axis component of inverter voltage and viq q-axis component of inverter voltage. iid and iiq are the dand qcomponents of the inverter currents, ws represents the angular frequency of the system voltage, lf represents the inductance of the filter, kqp and kqv are the droop coefficients of q/p and q/v, operating points of active and reactive power are represented by po and qo, vbus is the voltage at generation bus. fig. 4 grid-side inverter control of the wind/solar system when voltage droop is not activated the generating system is made to operate at unity power factor when qo kqv is varied from 0 to 25 depending on the ability of power generation. mppt control is applied by giving wind speed or solar irradiance as inputs to generate reference active power value (pref). pref, p measured value are given to pi controller converting it to iid,ref . id which is parks transformation value. the error obtained by comparing iid,ref and iid is fed to pi controller generating vd*. it can be written as, vid,ref = vsd + vd*𝛚slfiiq (4) q/p droop control is the comparison of active power whereas q/v droop control is the comparison of voltage control. kqp, kqv are the q/p and q/v droop coefficients respectively. kqp and kqv convert the error to reactive power component. qref= qo + qp + qv (5) where qo is the operating point of reactive power and is set to zero. again qref and q are given to pi controller and generates iiq,ref. iiq,ref are compared with iq in the pi controller generating an error vq*. viq,ref = vq* + vsq + 𝛚slfiid (6) therefore, vid,ref and viq,ref are fed as input to the dq-abc transformation and is converted into viabc,ref using inverse parks transformation for sinusoidal pulse width modulation which generates six pulses for the inverter. the speed of the generator is taken as 1.2 p.u, pitch angle is 0 and the speed of the wind is 11m/s. instead of using pi controller in the control schemes, adaptive neuro fuzzy logic controllers are used to improve stability and performance of the system. solar irradiance is about 660w/m2. by adding mppt control scheme and boost converters to maximize the utilization of free energy and to maintain constant output power. d. adaptive neuro-fuzzy interface system (anfis) the effective technique called anfis (adaptive neuro-fuzzy interface system) which was developed by dr. roger jang. apart from various optimizing methodologies in soft computing, the fuzzy logic and neuro computing has visibility, which leads to neurofuzzy systems. the combination of artificial neural network (ann) and fuzzy interface systems (fis) has attracted the interest of researchers in various applications. fuzzy logic interface system is a mapping point to map an input space to output space from starting point to the ending for all. fuzzy logic is an intriguing area of research because it has a premium quality of trading off among significance and precision. neuro adaptive learning methods similar to methods used for training neural networks is used for tuning parameters of fuzzy membership functions. this methodology is called as adaptive neuro-fuzzy inference system (anfis) . the backpropagation (bp) algorithm is used to trine the adaptive neural network and 7*7=49 rule based fuzzy logic command-line functions are used for training sugeno-type fuzzy inference systems using given input/output training data (jang 1993; srinivas singirikonda, sathish goud, & harika reddy, 2014). 4. result and discussion to establish the effectiveness of the proposed control strategies, simulation results are observed during the day time in the standalone micro-grid with high penetration of renewable generation system. voltage waveforms of pv, wind power, bess and diesel generator are clearly presented in matlab simulation. the matlab simulation diagram for adaptive neuro fuzzy control strategy for standalone micro grid system with multiple renewable sources show in fig.5. fig.5 matlab simulation diagram of adaptive neuro fuzzy control strategy for standalone micro grid system with multiple renewable sources case i: day time during the day period, the speed of the wind is considered to be varying from 10.5 to 11.5 m/s and set to an average of 11m/s, the solar irradiance ranges at 660w/m2. the active and reactive power of wind, solar are set at 0.413 and 0.495 respectively. the bess come into action during voltage variations whereas the duty of the diesel generator is to support bess by maintaining soc. int. journal of renewable energy development 9 (1) 2020: 43-51 p a g e | © ijred – issn: 2252-4940.all rights reserved 47 there are some oscillations of soc at diesel generator due to inertia of the machine leading to slow system dynamics and output variations in active power of renewable generation systems. without droop control, filter and inverter switching losses may lead to slight reduction in soc using droop control method soc fluctuations are reduced with the support of diesel generation, even though the fluctuations in soc is maintained at desired value. the frequency also fluctuates from its desired value without droop but the deviations are reduced with proposed droop method. fig.6 frequency control results for case i: (a) active power of wind and pv (b) active power of bess (c) active power of diesel generator (d) soc (e) frequency frequency fig.7 voltage control results for case i: (a) reactive power of wind power (b) reactive power of pv power (c) bus voltage of wind power (d) bus voltage of pv (e) frequency fig.6 shows the output active power of pv and wind, and active power flow characteristics of diesel generator bess. the results show that the diesel generator takes full response in the absence of droop control, for the output fluctuation of the renewable generation system with droop control, bess supports diesel generator to meet the power demand with p/f droop control method. fig.7 shows the variation of reactive power of wind and solar power respectively without implementation of droop control, the renewable generation system has same power factor, but by applying q/v droop control, the reactive power is controlled by compensating voltage deviation. by implementing the proposed method, the reactive powers of solar and wind are controlled, also mitigates the voltage fluctuation. the bus voltage of pv and wind are kept near to nominal value using q/v droop control. even though, the fluctuations are not effectively prevented. there is a considerable reduction in voltage fluctuations when q/p droop control is implemented. fig.8 comparison of active power with adaptive neuro fuzzy logic controller and pi-controller during day time the fig 8 shows simulation results for adaptive neuro fuzzy interface system (anfis) and pi-controller. anfis response rate is faster than pi controller and the simulation time to get output is less and easy to access and also the participation of dg id reduced to some extent giving optimal solution. case ii: night time at night time, the solar irradiance is 0 w/m2 and speed of the wind varies from 7.1 to 10.2 m/s and set to an average of 8.5 m/s. the reactive power value kqp are set as 0.473 for wind and 0.514 for pv power. fig.10 shows the power output from pv and wind, also the output active power of diesel generator as well as battery system. the output power response is like that of day time but the diesel generator output power is reduced and it is observed from the simulation results, accordingly socref is to be increased. the diesel generator is made to generate active power within allowable range. it is also observed that the frequency deviation during the night is more compared to day time, but frequency is maintained at 50 hz during the proposed method. fig.9 frequency control results for case ii: (a) active power of wind and pv (b) active power of bess (c) active power of diesel generator (d) soc (e) frequency. citation: singirikonda, s. and obulesu, y.p. (2020). a novel approach using adaptive neuro fuzzy based droop control standalone microgrid in presences of multiple sources. int. journal of renewable energy development,9(1), 43-51, doi.org/10.14710/ijred.9.1.43-51 p a g e | © ijred – issn: 2252-4940. all rights reserved 48 fig.10 voltage control results for case ii: (a) reactive power of wind power (b) reactive power of pv power (c) bus voltage of wind power (d) bus voltage of pv the fig.10 shows that during voltage control mode the output power fluctuation of wind generator is greater than that in day case. the more compensation of reactive power has led to more wind power output power fluctuation of during the day. since there is no solar irradiance, the voltage fluctuations are prevented. the change in voltage of the pv during the droop and proposed method is less. in fig.11, the violet color indicates the anfis controller output whereas the blue color denotes the pi controller. more over the bus voltages are maintained at a constant rated value even though the solar pv is not able to produce active power. the active participation of dg in compensating load power requirement makes the system to operate at constant frequency. the below figure 11 show that, the system operation with fuzzy control gives fast response in case of network variations. fig.11 comparison of active power with adaptive neuro fuzzy logic controller and pi-controller during night time. case iii: worst case (when there is no solar irradiance and wind speed) to study the robustness of the proposed method the absence of both solar and wind is considered as worst case. in this case, solar irradiance has to be varied, and the load demand is same as day time. hence the kqp is set as 0.413 for wind and 0.495 pv based on day time data. fig.12 shows active power response of the of wind and pv power generation systems. however, frequency remains unchanged. fig.12 frequency control results for case iii: (a) active power of wind and pv (b) active power of bess (c) active power of diesel generator (d) soc (e) frequency the fig.13 represents the voltage control simulation results. as a result, there are some deviations around some points but the proposed method performs better than others. it is observed that the participation of bess and dg is more in this case and the reactive power requirement of the load, wind generator are compensated by the solar inverter setup. the bus voltages are also maintained at constant rated value even in the worst-case operation. the simulation results of fig.14 show the comparison of both pi and adaptive neuro-fuzzy logic controllers even in worst case has better performance. fig.13 voltage control results for case iii: (a) reactive power of wind power (b) reactive power of pv power (c) bus voltage of wind power (d) bus voltage of pv fig.14 comparison of active power with adaptive neuro fuzzy logic controller and pi-controller during worst case int. journal of renewable energy development 9 (1) 2020: 43-51 p a g e | © ijred – issn: 2252-4940.all rights reserved 49 case iv: effect of load change and bess tripping on the system performance. fig.15 load change simulation results for case iv: (a) active power of bess and diesel generator (b) soc (c) frequency here consider two cases (i) load change (ii) tripping of bess for the frequency control strategy. fig.15 shows the load change in day time. there is a load decrement at 0.5mw at 3sec of time. soc and frequency are maintained same as previous cases. the fig.16 shows the load change simulation of adaptive fuzzy controller scheme and pi controller. taking time (sec) on x-axis and active power (mw) on y-axis. fig.17 represents the results for the case of while battery system is tripped. the bess is tripped out of the system due to fault maintenance etc. the diesel generator operates by changing its switch position to ‘b’ when bess trips out of the system. fig.18 shows the tripping of bess simulation results of adaptive fuzzy controller and pi controller scheme. fig.16 comparison of active power during load change with adaptive neuro fuzzy logic controller and pi-controller fig.17 bess tripping simulation results for case iv: (a) active power of bess and diesel generator (b) frequency fig.18 comparison of active power with adaptive neuro fuzzy logic controller and pi-controller during bess tripping case v: considering pv bus only fig.19 simulation results for case v: (a) reactive power of pv power (b) bus voltage of pv power in this case, the output of wind power system is kept constant and active power fluctuation of the pv system is considered and the effect of voltage control strategy on pv power system bus are observed. the pv power system shows results in fig 19 and the reactive power is limited. case vi: adjusting charge/ discharge of bess bess should be controllable for the energy efficiency perspective. by varying the slope of the ramp of soc, bess is controlled to output the desired level of active power. fig.20 shows charging of soc at 1mw and active power of battery storage system and diesel generator. fig.20 simulation results for case vi: (a) active power of bess and diesel generator during charging (b) soc during charging citation: singirikonda, s. and obulesu, y.p. (2020). a novel approach using adaptive neuro fuzzy based droop control standalone microgrid in presences of multiple sources. int. journal of renewable energy development,9(1), 43-51, doi.org/10.14710/ijred.9.1.43-51 p a g e | © ijred – issn: 2252-4940. all rights reserved 50 bess is consequently adjusted such a way to discharge if the power output of the diesel generator varies at t ramp rate and the effect is shown in fig.21. the bess is controlled in this manner, to discharge active power required by adjusting ramp rate of soc which includes the controller of diesel generator. by comparing with different cases, the system performance in case-1 with anfc is better when compared to all other cases because, all sources are active at day time and whereas solar energy is limited in night time. with the use of adaptive neuro fuzzy controller, the output active power wind and pv are 7mw and 50kw.frequency and soc are maintained as 50hz and 0.5p.u. fig.21 simulation results for case vi: (a) active power of bess and diesel generator during discharging (b) soc during discharging. table 3 comparison of results with and without controller cases without controller (pi controller) with adaptive neuro fuzzy logic controller 1. day time active power is 2.7mw at 1.4sec. frequency is 49.8hz with some oscillations. soc is 0.48 p.u active power is 2.8mw at 1. 2sec.the frequency is maintained 50hz constant. soc is 0.5p.u with less damp of oscillations and settles faster than pi 2. night time active power of wind and pv are 3.8mw, diesel generator is 3mw.reactive power of wind is 0.48 and pv is 0.52. bus voltage of wind and pv are 0.97 and 0.98 p.u active power of wind and pv are 4mw, diesel generator is 3.2mw reactive power of wind is 0.473 and pv is 0.514. bus voltage of wind and pv are 0.98 and 0.99 p.u 3. worst case (no solar irradiance and wind speed) active power is 2.6mw at 1.8sec time. reactive power of wind and pv are 0.413 and 0.495mvar active power is 2.8mw at 1.5sec time. reactive power of wind and pv are 0.5and 0.8mvar 4. i) load change active power is 4.9mw at 3.1sec of time active power is 4.9mw at 2.8sec of time ii) tripping of bess active power is 2mw at 3.1sec of time active power is 2mw at 2.8sec of time and the frequency is maintained constant as 50hz. 5. conclusion to mitigate the problems of diminishing voltage and frequency fluctuations, adaptive neuro fuzzy interface system is used, which has quick response rate compared to pi-controller. the implementation of bess leads to stable operation of the system maintaining the frequency at nominal value of 50hz. without any deviation. for this reliable voltage control a novel q/p droop is introduces into the control scheme for controlling the reactive power flow in test system with multiple renewable generators. the active power fluctuations are effectively prevented by damping voltage fluctuations in the renewable generation. the output active power of pi and adaptive neuro fuzzy controllers are compared and simulation results are observed on the graph during different cases. simulation results are observed in matlab software by using these control strategies. the anfis controller improves system stability without any interruptions and produces effective performance. references carvalho, pedro m, s., correia, pedro f. & ferreira, luís a, f, m. (2008) distributed reactive power generation control for voltage rise mitigation in distribution networks. ieee transactions on power systems, 23(2), 766–772. diaz, g., gonzalez-moran, c., gomez-aleixandre, j. &diez, a. (2010) scheduling of droop coefficients for frequency and voltage regulation in isolated micro-grids. ieee transactions on power systems, 25(1), 489– 496. divya, k, c., ostergaard, j. (2009) battery energy storage technology for power systems – an overview. electrical power systems research, 79(4), 511–520. ekanayake, j., & jenkins, n. (2004) comparison of the response of doubly fed and fixed-speed induction generator wind turbines to changes in network frequency. ieee transactions on energy convers,19(4), 800–803. ekanayake, j., holdsworth, l., & jenkins, n. (2003) control of dfig wind turbines. power engineering, 17(1), 28–32. jang, j. (1993) anfis: adaptive-network-based fuzzy inference systems, ieee transactions on systems, man, and cybernetics, 23(3) ,665-685. keung, p., li, p., banakar, h. & ooi, b. (2009) kinetic energy of wind turbine generators for system frequency support. ieee transactions on power systems, 24(1), 279–287. kim, y, j. (2010) optimal control of dg output voltage considering switching operation of ultc and sc in distribution power systems. m.s. thesis, seoul nat university, seoul, korea. lalor, g., a. mullane, a. & o’malley, m. (2005) frequency control and wind turbine technologies. ieee transactions on power systems, 20(4), 1905– 1913. park, m. & yu, i, k. (2001) photovoltaic generation system simulation using real field weather condition. journal of kiee, 5(2),169–174. rana, r., singh, m. & mishra, s. (2017) design of modified droop controller for frequency support in microgrid using fleet of electric vehicles. ieee transactions on power systems , 32(5), 3627 – 3636. int. journal of renewable energy development 9 (1) 2020: 43-51 p a g e | © ijred – issn: 2252-4940.all rights reserved 51 schlabbach, j., blume, d. & stephanblome, t. (2001) voltage quality in electric power system. iet digital library. scott, gw., wilreker, vf., shaltens rk. (1984) wind turbine generator interaction with diesel generators on an isolated power system. ieee transactions on power apparatus and systems,103(5), 28-29. senjyu, t., miyazato, y. & yona, a. (2008) optimal distribution voltage control and coordination with distributed generation. ieee transactions on power delivery, 23(2), 1236–1242. serban, i., marinescu, c. (2011) frequency control issues in microgrids with renewable energy sources. 7th international symposium on advanced topics in electrical engineering (atee), bucharest, romania, 1–6. shayeghi, h., shayanfar, h, a., jalili, a. (2009) load frequency control strategies: a state-of-the-art survey for the researcher. energy convers manage, 50(2), 344–353. srinivas, singirikonda., sathish goud, g & harika reddy, m. (2014) transient stability of a.c generator controlled by using fuzzy logic controller. international journal of engineering research and applications, 4 (3), 389-395. sumina, d., erceg, g. & idzotic, t. (2005) excitation control of a synchronous generator using fuzzy logic stabilizing controller. 2005 european conference on power electronics and applications, dresden, germany. velasco de la fuente, d., trujillo, c., garcera, g., figueres, e & ortega, r. (2013) photovoltaic power system with battery backup with grid-connection and an islanded operation capabilities. ieee transactions on industrial electronics, 60(4),1571 – 1581. © 2020. this article is an open access article distributed under the terms and conditions of the creative commons attribution (cc by) license (http://creativecommons.org/licenses/by/4.0/) microsoft word soulayman_et al.docx int. journal of renewable energy development 8 (1) 2019: 33-39 p a g e | © ijred – issn: 2252-4940. 2019. all rights reserved 33 contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: http://ejournal.undip.ac.id/index.php/ijred synthesis parameters of biodiesel from frying oils wastes soulayman soulayman* and dayoub ola department of applied physics, higher institute for applied sciences and technology (hiast), damascus, syria abstract. this study is devoted to produce biodiesel from recycled wastes frying oils (wfo) using commercial grade chemicals in an attempt to help reducing the cost of biodiesel and pollution coming from wfo. the base – catalyzed transesterification method was applied. the variables affecting the yield and characteristics of the biodiesel produced from wfo were studied. sodium hydroxide is used as catalyst. different reaction times, different methanol/wfo volume ratios and different catalyst/wfo weight ratios were used with purpose of achieving the best conditions for biodiesel production a series of experiments were carried out, using methanol/wfo volume ratios from 10% to 30% and catalyst/wfo weight ratio from 0.2% to 0.8%. it was found that in treating wfo which contains 0.12% of water by weight and having an acid value of 0.52 mg of koh/g of oil, and an iodine value 130.42 gi/100 g of oil, no need in acidic pretreatment. moreover, it was found that, for wfo with an acidic value of 0.52 mg koh/gwfos, results show that a methanol/wfo volume ratio of 13% and a catalyst/wfo weight ratio of 0.4% give the highest yield of methyl esters. a pilot production unit of 400l/day of production capacity was designed and constructed on the basis of laboratory experiments and the process was verified on the pilot scale. ©2019. cbiore-ijred. all rights reserved keywords: recycled frying oils, biodiesel, transesterification method, commercial chemicals, experiments, pilot station. article history: received october 18th 2017; received in revised form may 17th 2018; accepted december 8th 2018; available online how to cite this article: soulayman, s. and ola, d. (2019) synthesis parameters of biodiesel from frying oils wastes. int. journal of renewable energy development, 8(1), 33-39. https://doi.org/10.14710/ijred.8.1.33-39 1. introduction in syria, thousands litres of fatty wastes are discarded each year into sewage systems. this adds to the cost of treating effluent or pollutes waterways, or is integrated into the food chain through animal feeding, thus becoming a potential cause of human health issues. there are several end–uses for this waste, such as the production of soaps or of energy by anaerobic digestion, thermal cracking (zaher, 2003), and more recently the production of biodiesel. from a waste management standpoint, producing biodiesel from used cooking oil is environmentally beneficial, since it provides a cleaner way for disposing these products in comparison with what is typically done. moreover, the fuel thus obtained performs in similar way to fossil fuel, with the advantage of reducing greenhouse emissions because it is a renewable resource. anyway, a global biofuel economy, with a division of labour favouring the most efficient producers, would be a key boon to developing countries. their yearround growing seasons and cheap farm labour offer a valuable competitive advantage over colder and high production cost rich countries. yet, the emerging global market in biofuels is running into political trouble. many reviews of biodiesel production processes by transesterification are available, (ma & hanna 1999, demirbas, 2003, lotero et al.. 2005, ferreira, cardoso & da silva, 2012, gnanaprakasam et al., 2013, talebian * corresponding author: soulayman1952@gmail.com kiakalaieh, amin, mazaheri, 2013, araujo, andrade, silva, dupas, 2013, alemayehu gashaw & abile teshita, 2014). however, the use of several low molecular weight alcohols and homogeneous acid and basic catalysts for transesterification has been studied with success over years (nye, 1983, peterson& scarrah, 1984, schuchardi, sercheli & vargas, 1998, khan, 2002). the production of biodiesel continues to be an axis of research for different research groups all over the world (see for example (dorado et al., 2002 a, dorado et al., 2002b, guo & leung, 2003, zhang, dube, mclean, and kates, 2003, van gerpen, 2005, felizardo et al., 2006, charpe & rathod 2011, puna et al., 2013, medina-valtierra & ramirez-ortiz, 2013, seid yimer, & omprakash sahu, 2014)). basic homogeneous catalysts have surpassed the acid ones, because the reaction is faster and lead to the same yield of transesterification (above 90% for both) (schuchardi, sercheli & vargas, 1998). on the other hand, sodium and potassium hydroxides react with the alcohol to give water. this reaction is undesirable, because water can react with triglycerides, fatty acids, or esters in hydrolysis reactions and subsequent saponification (peterson& scarrah, 1984). in spite of their lower reactivity, sodium and potassium hydroxides can also lead to high reaction yields just by increasing the catalyst quantities. the transesterification reaction reaches its equilibrium state and the transformation accurse essentially by mixing the research article citation: soulayman, s. and ola, d. (2019) synthesis parameters of biodiesel from frying oils wastes. int. journal of renewable energy development, 8(1), 33-39, doi.org/10.14710/ijred.8.1.33-39 p a g e | © ijred – issn: 2252-4940. 2019. all rights reserved 34 reactants. the presence of a catalyst accelerates the adjustment of the equilibrium considerably (ma & hanna 1999). the stoichiometric reaction requires 1 mole of triglyceride and 3 mole of alcohol. this study is intended to consider aspects related to the facility of the production of biodiesel from waste/recycled oils using commercial grade chemicals in an attempt to help reduce the cost of biodiesel and reduce waste and pollution from waste oils. the variables such as methanol to wfo molar ratio, the reaction temperature, the reaction time and the amount of catalyst that affect the yield and characteristics of the wfo based biodiesel were studied. the achieved results were analysed and a set of recommendation was proposed. an educational pilot production unit of 400l/day of production capacity was designed, constructed and exploited on the basis of experimental results. 2. materials and methods 2.1 materials this research was carried out at higher institute for applied sciences and technology (hiast), damascus, syria during the years 2010-2015. the wfo was collected from hiast's restaurant. it consisted only of soybean oil. the cooking temperature of the oil varied from 150 to more then 200 oc. the oil was kept at these temperatures for 6 hours per day and was replaced weekly. because of their low cost, commercial grade chemicals such as methanol and sodium hydroxide (naoh) were chosen as alcohol and alkaline catalyst for transesterification process. pure (merck gr) chemicals were also used but practical differences were not observed. therefore, all provided results are based on grade commercial chemicals. moreover, a pilot production unit has been designed and constructed. this unit operates with wfo feedstock and commercial grade chemicals. fig. 1 describes the processing steps and methods used to produce biodiesel and pure glycerine from wfo with the production capacity of 400 l/day. 2.2 frying oil pre-treatment when reacting the wfo, filtration operation is always done to remove any suspended matter. moreover, in the conventional transesterification of fats/vegetable oils for biodiesel production, free fatty acids and water always produce negative effects, since their presence causes soap formation, consumes catalyst and reduces catalyst effectiveness, all resulting in a low conversion (kusdiana & saka, 2004). the waste frying oil (wfo) pretreatment to remove water was done using primary vacuum pump at 80 oc and by heating treatment at 110 oc. the treated oil contains 0.12 wt % of water, an acid value of 0.52 mg of koh/g of oil, and an iodine value of 130.42 gi/100 g of oil. generally, the synthesis of biodiesel from low quality oils such as wfo containing a low to moderate amount of free fatty acids (ffas) in addition to moisture and other impurities is challenging due to undesirable side reactions. the pretreatment stages, involving an acid catalyzed pretreatment integrated with water separation, are necessitated to reduce ffas concentration and water to below threshold limits prior of being processed by standard biodiesel manufacturing. besides reducing ffas concentrations, acid catalysts are able to catalyze triglycerides (tg) transesterification, opening the door for the use of acid catalysts to perform simultaneous ffa esterification and tg transesterification. fig. 1. flow chart to produce biodiesel and pure glycerin from wfo. 2.3 basic – catalyzed transesterification procedure a sample of 300 (ml) of wfo was transferred to a two – neck woulf flask equipped with a thermometer and reflux condenser. a magnetic bar was used for stirring. naoh pellets were completely dissolved in methanol and added to pretreated restaurant waste oil. the mixture was heated until the desired temperature was reached (60oc), and the transesterification reaction begun. the reactor was kept at around 60 oc for 2 h. the study was carried out using a methanol/wfo volume ratio which varied from 10 % to 30 % and catalyst quantity equivalent to 0.36 – 1.44 wt % of oil and stirring speed of 600 rpm. the scheme of the experimental stand is shown on fig. 2 with primary vacuum pump. the scheme of the experimental stand without using primary vacuum pump is similar to that of fig.2 with exception that thermal treatment is provided under temperature of 125oc. the same primary vacuum pump used in the wfo pretreatment was used also in biodiesel drying. 2.4 purification of methyl esters phase at the end of the reaction period (2 h), the mixture was transferred carefully to a separating funnel to stand there for two hours. the lower layer (glycerol, methanol and most of the catalysts) was drained out. the upper layer (methyl esters, some methanol and traces of the catalyst) was then heated at temperature 100 oc for 0.5 h in order to remove the excess of methanol and subsequently cleaned thoroughly by washing with water to provide a purified biodiesel (to bring ph down to 7). the washed methyl esters were then dried by distillation. alcohol catalyst waste frying oil mixing reactor reaction separation drying distillation control analysis washing alcohol glycerol distillation production storage h2so 4 baco3 filtration pure glycerin pretreatment h2so 4 int. journal of renewable energy development 8 (1) 2019: 33-39 p a g e | © ijred – issn: 2252-4940. 2019. all rights reserved 35 fig. 2. biodiesel production using primary vacuum pump flow chart. 3. results and discussion 3.1 wfo and biodiesel gas chromatography analysis gc (gas chromatography) analysis was performed for identifying the hydrocarbon compounds of produced biodiesel. the separation was carried out by using capillary column rtx5ms 30 m × 0.25 mm id, 0.25 µm with helium at 137.7 ml/min as a carrier gas and 1:100 split ratio. fatty acid composition of the waste frying oil is given in table 1 while the biodiesel composition is given in table 2. the wfo results are in accordance with those of (zambiazi, przybylski, zambiazi & mendonça, 2007), and (giakoumis, 2013). the little bit differences between the provided compositions of different authors are normal. hammond et al., (2005) mentioned that fatty acid composition of soybean oil changes considerably with maturity and with seed oil production. in typical soybean triacylglycerols, the palmitate and linolenate tend to decrease with maturity, whereas linoleate increases. oleate tends to increase to a maximum and then decline slightly. soybeans selected for a typical fatty acid compositions show quite different patterns. tabel 1 fatty acid composition of waste frying oil (wfo). fatty acid name formula weight (%) a b c myristic ch3(ch2)12cooh 1.06 0.06 0.12 palmitic ch3(ch2)14cooh 12.04 9.90 11,44 stearic ch3(ch2)16cooh 3.57 3.94 4.14 oleic ch3(ch2)7ch= ch(ch2)7 cooh 21.79 21.53 23.47 linoleic ch3(ch2)3(ch2ch= ch)2(ch2)7 cooh 54.84 56.02 53.46 linolenic ch3 (ch2ch= ch)3(ch2)7 cooh 6.70 7.15 6.64 a)present work, b) analysis results taken from (zambiazi, przybylski, zambiazi & mendonça, 2007) and c) analysis results taken from ( giakoumis, 2013). the biodiesel composition results obtained in the present work (see table 2) are in accordance with those of (carvalh et al., 2012) and (hammond et al., 2005). tabel 2 methyl ester composition produced from waste frying oil (wfo). methyl ester formula weight (%) a b c d myristic ch3(ch2)12coo ch3 0.06 0.06 0.04±0.5 palmitic ch3(ch2)14coo ch3 11.30 10.7 11,4 10.57±0.43 stearic ch3(ch2)16coo ch3 3.64 3.3 3.2 4.09±0.34 oleic ch3(ch2)7ch= ch(ch2)7 coo ch3 22.37 23.5 21.2 22.98±2.01 linoleic ch3(ch2)3(ch2ch= ch)2(ch2)7 coo ch3 56.15 53.6 52.5 54.51±1.54 linolenic ch3 (ch2ch= ch)3(ch2)7 coo ch3 5.80 6.3 7.5 7.23±0.78 a)present work, b) analysis results taken from (carvalh et al., 2012), c) analysis results taken from (carvalh et al., 2012) and d) analysis results taken from (hammond et al., 2005). 3.2 the methanol/wfo volume ratio influence one of the most important variables affecting the yield of ester is the molar ratio of alcohol to triglyceride (tg) (peterson& scarrah, 1984), and (schuchardi, sercheli & vargas, 1998). the stoichiometric ratio for transesterification requires 3 moles of alcohol and 1 mole of triglyceride to yield 3 moles of fatty acid ester and 1 mole of glycerol: wfo (tg) + 3 ch3oh è 3 fatty acid ester + 1 glycerol the molar ratio is associated with the type of catalyst used. the yield of the reactions is expressed as the mass of methyl esters produced per 100 g of wfo. the excess of methanol is introduced to ensure the accomplishment of the reaction. but, when considering the weight of methyl ester phase, it contains part of the excess methanol, resulting in more than 100% of crude yield (fatty acid ester + excess of methanol). (soulayman, mustafa & hadbah, 2012), used pure (merck gr) chemicals and found that at constant weight ratio of catalyst/wfo, after 2 h of reaction the yield increases directly with the excess of methanol up to methanol/wfo volume ratio of 18.17%. when repeating their experiment without removing the excess of methanol from the yield their statement was verified (see fig. 3). therefore, this kind of yield could be characterized as crude yield. the dependence of crude yield on methanol/wfo volume ratio, presented in fig.3, shows that, when excluding the sample with methanol/wfo volume ratio equals to 0.3 (for the oil sample with an acidic value of 0.52 mg koh/g at 0.004 mass ratio of catalyst/wfo) the crude yield of the reactions increases directly with the methanol content increase. therefore, citation: soulayman, s. and ola, d. (2019) synthesis parameters of biodiesel from frying oils wastes. int. journal of renewable energy development, 8(1), 33-39, doi.org/10.14710/ijred.8.1.33-39 p a g e | © ijred – issn: 2252-4940. 2019. all rights reserved 36 the crude yield was distilled in order to remove the excess of methanol. fig. 3 influence of the methanol/wfo volume ratio on the crude yield of biodiesel production for catalyst/wfo weight ratio of 0.00367(g/ml) without methanol excess distillation. when removing the excess of methanol from the crude yield, it was found that yield, before washing, was constant and equal to 95.73 whatever is the methanol/wfo volume ratio. after washing, the final yield becomes as shown in fig. 4 where it is seen that methanol/wfo volume ratio of 13% is the optimum value which leads to maximum real yield (95.09). the methanol/wfo volume ratio of 16% is also acceptable as it leads to yield of 94.45. fig. 4 influence of the methanol/wfo volume ratio on the yield of biodiesel production for weight catalyst/wfo ratio of 0.00367(g/ml) with methanol excess distillation and after washing. 3.2 the naoh/wfo mass ratio influence the amount of catalyst used in the process is another variable to take into account because it determines the reaction rate, and can cause hydrolysis and saponification. both reactions interfere with the separation of the glycerol rich phase and with the methyl esters purification. alkalicatalyzed transesterification is much faster than acidcatalyzed (freedman, pryde, & mounts, 1984). sodium hydroxide was chosen in the present work to catalyze the transesterifications because it is the cheapest one and is used widely in large-scale processing. soulayman, mustafa & hadbah, (2012), stated that, for a wfo with an acidity of 0.52 mg koh/g wfo and a reaction temperature of 60 oc, the optimal conditions to produce biodiesel by basic – catalyzed transesterification method using pure (merck, gr) chemicals are a methanol/wfo volume ratio of 18.17% and naoh/wfo weight ratio of 0.4 %. when repeating their experiment using commercial grade chemicals and primary vacuum pump additionally, it was found that, for the optimum methanol/wfo volume ratio of 13%, the catalyst quantity is of 0.4% of oil weight. this result is in a full agreement with that of soulayman, mustafa & hadbah, (2012). at value lower than 0.4%, the process became unstable. the yield varied by about 3%. at value higher than 0.4%, the yield decreases considerably (see table 3). tabel 3 the effect of naoh catalyst on wfo transesterification. naoh mass (g) crude yield (ml) yield after excess methanol removal (ml) yield after washing (ml) 1 304 300 290 1 315 300 295 1 300 297 289 1.1 306 300 298 1.1 307 301 294 1.1 308 300 295 1.2 300 297 285 1.2 298 294 282 1.2 299 294 283 used values are: 300ml wfo, 40 ml ch3oh, reaction time 2 h. it should be noted that, soaps formed during base catalyzed transesterification were eliminated by applying sulfuric acid after the transesterification completed. this procedure simplifies also the separation of the product phases and prevents the formation of emulsions if a water wash operation is used for the finished fuel. finally, the optimization data clearly demonstrated the evidence that the transesterification reaction can be accelerated by increasing the reaction temperature, and increasing the amount of base catalyst. for reactions occurring at atmospheric pressure, the temperature is limited by the boiling point of methanol (64.6°c), with the highest temperature used in this study being 60°c. increasing the amount of base catalyst will increase soap formation, but as noted previously, these soaps can be eliminated by using sulfuric acid in a second processing step. 3.3 the effect of reaction time when studying the influence of reaction time on the reaction productivity, it was found (see fig. 5) that the conversion rate increases with reaction time. moreover, the influence of reaction period is less important when period exceeds 75 minutes while it is very important at lower periods. fig. 5 the effect of reaction time on yield. 96 97 98 99 100 101 102 103 104 10 15 20 25 30 c ru de y ie ld r at io (% ) methanol/wfo (100*vol/vol) 92 93 94 95 96 10 20 30 ye ild r at io (% ) methanol/wfo (100*vol/vol) 90 92 94 96 98 100 102 0 30 60 90 120 yi el d ra ti o (% ) time (minutes) int. journal of renewable energy development 8 (1) 2019: 33-39 p a g e | © ijred – issn: 2252-4940. 2019. all rights reserved 37 however, the production scale and the economical feasibility give the main data for determining the suitable time reaction period from production point of view. under the condition of methanol to oil ratio of 6:1, 1% sodium hydroxide catalyst and 60oc, the refined soybean oil was transesterified (freedman, pryde, & mounts, 1984). after 1 h, the conversion rate was 94%. the result of the present work is in accordance with that of (freedman, pryde, & mounts, 1984). 3.5 biodiesel characterization in order to verify the quality of the biodiesel produced under different experimental conditions, the properties of the obtained biodiesel were compared with the specifications of the standard en 14214 (2003). so, several samples of biodiesel were submitted to a series of tests. table 4 gives the obtained results of some samples at more preferable conditions regarding esters concentration (%wet), density at 15 oc (g/cm3), kinematic viscosity at 40 oc (mm2/s) and iodine value (gi/100g) while table 5 gives the obtained results regarding acid value (mg koh/g), water content (% wt), sodium content (mg/kg), methanol content (% wet) and flash point (oc) of the produced biodiesel. the obtained results show that for all the studied conditions, the biodiesel produced in the present study has a density of 0.88 g/cm3. the methanol content in the produced samples was 0.002 (%wt) while the standard en 14214 (2003) allows the max value of 0.2 (%wt). table 4 some physical properties of the produced biodiesel. sample number esters concentration (%wet) density at 15 oc (g/cm3) cinematic viscosity at 40 oc (mm2/s) iodine value (gi/100g) 1 98.73 0.88 4.94 129.6 2 99.15 0.88 4.59 129.7 3 99.17 0.88 4.51 128.7 4 98.79 0.88 4.44 128.3 5 98.46 0.88 4.33 128.8 standards min. 96.5 0.86 – 0.9 3.5-5 less than 120 table 5 some other physical properties of the produced biodiesel. sample number acid value (mg koh/g) water content (%wet) sodium content (mg/kg) flash point (oc) 1 0.20 0.042 1.3 184 2 0.27 0.037 2.0 182 3 0.21 0.033 2.0 182 4 0.30 0.038 3.0 182 5 0.34 0.035 2.8 182 standards max. 0.5 max.0.05 max.5 >120 3.6 acid pre-treatment effectiveness acid pre-treatment is used to reduce free fatty acids ffa. in order to determine the influence of acid pre-treatment process on the crude yield this method was applied on wfo of different acid values. thus an acid-catalyzed pretreatment process before the base-catalyzed transesterification process was applied. in order to compare the influence of applying (1) basic – catalyzed transesterification and (2) acid pre-treatment basic – catalyzed transesterification processes on the biodiesel productivity, the above mentioned two processes were applied using the same experimental variables except adding h2so4 quantity.). table 6 the effect of acid pre-treatment on the crude yield. h2so4 (ml) biodiesel (ml) 0 308 0.3 296 0.6 298 1.0 302 used values are: 300ml wfo, 40 ml ch3oh, reaction time 2 h. acid value 0.52 the results are given in table 6. in this context, when repeating the same series of experiments with changing the wfo acid value, it was found that for wfo of acid value less than 1 the basic – catalyzed transesterification is preferable from industrial and economical point of view even it was recommended that 0.5 1.5% (based on the weight of free fatty acid in the oil) of pure (95-98%) sulphuric acid should be used as a catalyst (abdullah, hasan and yusoff, 2013) 4. pilot production unit most researchers have selected the cheapest and simplest technique of biodiesel production. others have applied the expensive methods once the untreated oil was used as feedstock of biodiesel. however, in commercial processes, highly refined vegetable oils, consisting primarily of triglycerides (tgs) and typically used as feedstocks, are transesterified with low molecular weight alcohols, e.g. methanol and ethanol, using homogeneous alkali catalysts (such as naoh and koh). to be more economically viable, the use of virgin oils accounting for 88% of the total estimated biodiesel production cost, could be replaced with a more economical feedstock, such as waste fats and oils that contain a low to moderate amount of free fatty acids (ffas), moisture and other impurities (ferreira, cardoso & da silva, 2012). even though, the economical impact is highly related to the biodiesel production technique. therefore, the quality of biodiesel should be taken into account when any of the techniques reported by (jain and sharma, 2010) is applied. it is due to the operational process may take place either with catalysts such as alkaline (meher, dharmagadda & naik, 2006), acid, (lotero et al.. 2005), and enzyme (noureddini, gao and philkana, 2005, soumanou and bornscheuer, 2003), or citation: soulayman, s. and ola, d. (2019) synthesis parameters of biodiesel from frying oils wastes. int. journal of renewable energy development, 8(1), 33-39, doi.org/10.14710/ijred.8.1.33-39 p a g e | © ijred – issn: 2252-4940. 2019. all rights reserved 38 without catalysts (madras, kolluru and kumar, 2004). in an industrial scale, the alkaline catalyst is the most preferable way due to low cost, ease of installation and above all, its high reaction rate without requiring a large reactor volume and saving the production fixed capital (berrios, martín, chica and martín, 2010). the recent increase in the world biofuels demand, along with the need to reduce costs while improving the environmental sustainability of the biodiesel production, have led to the search for catalysts that should be economically viable, efficient, and environmentally friendly. in this context, a pilot production unit has been designed and constructed. this unit operates with wfo feedstock and commercial grade chemicals. fig. 1 describes the processing steps and methods used to produce biodiesel and pure glycerine from wfo with the production capacity of 400 l/day. the first main step in the used process is the chemical analysis of the weekly gathered quantity of wfo in order to determine the acid value of the free fatty acid content and to apply the suitable wfo pre-treatment. the acid value of the wfo was determined in order to estimate the free fatty acid content and give an idea of how much acid catalyst and methanol would be needed to push the acid esterification chemical towards methyl ester production. karmee, patria & lin, (2015), reported the economic analyses of the biodiesel production from waste cooking oil in hong kong with different catalysts using aspen plus as simulation tool. as the economic feasibility of the pilot unit is out of the frames of the present work some comments regarding wfo pretreatment and production results will be provided below. the detailed cost analysis of the pilot production unit will be given elsewhere. 4.1 wfo pre-treatment the pre-treatment stage in the pilot unit consist of solid particles removal and water separation always while the application of acid catalyzed pre-esterification depends on the wfo acid value. if acid value is less than 1.5, the acid esterification is not required. so, as the acid value of the used raw materials of wfo was less than 1 over three years of operation, the description of the pre-treatment stage will be restricted to particles removal and water separation. 4.2 production conditions the biodiesel production capacity is 400 l/day. the reaction conditions are kept in optimum conditions, such as the reaction temperature at 60 °c. the volume ratio of methanol to wfo was 13% and mass ratio of naoh to wfo is 0.4%. the reaction time was 2 h. as the acid value of the used wfo is always less than 1 the acid pretreatment is not required. in these optimum conditions the biodiesel conversion is 100% approximately. the daily obtained results with regards to local prices of 2012 demonstrate the economical feasibility of continuous pilot biodiesel production. 4.3 production results in the pretreatment process, the solid particles removal is provided manually where the same container plays the role of decanter to separate the wfo from the water while the rest moisture removal is provided under primary vacuum in heater with a mixing ratio of 200rpm. the suitable quantities of methanol and catalyst naoh are reacted to produce methoxide in a special reactor with a mixing ratio of 200 rpm. the pre-treated wfo as well as the produced methoxide are then transferred into the reactor with a mixing ratio of 800rpm in which the wfo will be transesterified with methoxide. the reaction mixture of this process is decanted in a special decanter. the byproduct glycerol is filtered and transferred to a special decanter, leaving the biodiesel, base catalyst (naoh), and remaining glycerol in the filtrate which will be transferred into a distillation column with a mixing ratio of 800rpm for excess methanol distillation which in its turn transferred to the methanol storage while the rest is transferred to a special decanter with a mixing ratio of 200rpm for washing with water after adding some drops of h2so4. a stream of water and the filtrate are transferred into extraction column. the remaining methanol, koh, and remaining glycerol are extracted by water, while the biodiesel is separated. then the biodiesel is transferred into a distillation column with a mixing ratio of 800rpm to dry it by removing the remaining moisture. the obtained pure biodiesel is then stored in the biodiesel tank. on the other hand, all the remaining substances such as wfo, methanol, glycerol, and naoh in the reaction are transferred into a special reactor for the neutralization process where a suitable quantity of h2so4 is added. after 12 h of neutralization the remaining wfo is decanted; while the remaining methanol, glycerol and water are transferred into two separated distillation columns: one for extracting methanol and transferring it to the methanol storage while the other for water extraction. a suitable quantity of baco3 is added to the water-free and methanol-free glycerol for h2so4 removal. the obtained pure glycerol after filtration is then stored in a tank. 5. conclusion the present work allows achieve the following results: o the production of biodiesel from wfo is feasible by basic catalyzed transesterification. o the determined best working conditions for basic catalyzed transesterification biodiesel production from wfo as follows: primary vacuum pump pretreatment, 2h of reaction, a methanol/wfo volume ratio of 13% and a catalyst/wfo weight ratio of 0.4%. o a good agreement with available data is observed. o the realization of efficient separation/purification of methyl asters phase o the design and construction of biodiesel production educational pilot facility of 400l/day production capacity that meets the specifications of en 14214. 6. acknowledgments the authors wish to thank dr. r. jabra for useful discussions and english reviewing. references barzegarian, r., keshavarz moraveji, m. & aloueyan, a. (2016) experimental investigation on heat transfer characteristics and pressure drop of bphe (brazed plate heat exchanger) using tio2-water nanofluid. experimental thermal and fluid science, 74, 11-18. int. journal of renewable energy development 8 (1) 2019: 33-39 p a g e | © ijred – issn: 2252-4940. 2019. all rights reserved 39 behrangzadeh, a. & heyhat, m.m. (2016) the effect of using nano-silver dispersed water based nanofluid as a passive method for energy efficiency enhancement in a plate heat exchanger. applied thermal engineering, 102, 311-317. han, x.h., cui, l.q., chen, s.j., chen, g.m. & wang, q. (2010) a numerical and experimental study of chevron, corrugatedplate heat exchangers. international communications in heat and mass transfer, 37, 1008-1014. holman, j. p. (2001) experimental methods for engineers (7th edition). new york: mcgraw-hill. huang, d., wu, z. & sunden, b. (2016) effects of hybrid nanofluid mixture in plate heat exchangers. experimental thermal and fluid science, 72, 190-196. kabeel, a e, el maaty, t.a, & el samadony, y. (2013) the effect of using nano-particles on corrugated plate heat exchanger performance. applied thermal engineering, 52, 221-229. kakaç s., liu h. & pramuanjaroenkij a. (2012) heat exchangers: selection, rating, and thermal design. florida, usa: crc press. kan, m., ipek, o. & gurel, b. (2015) plate heat exchangers as a compact design and optimization of different channel angles, acta physica polonica, 12, 49-52. kumar, v., tiwari a. k., & ghosh, s.k. (2016). effect of chevron angle on heat transfer performance in plate heat exchanger using zno/water nanofluid, energy conversion and management, 118, 142-154. pandey, s.d. & nema, v.k. (2012) experimental analysis of heat transfer and friction factor of nanofluid as a coolant in a corrugated plate heat exchanger, experimental thermal and fluid science, 38, 248-256. sarafraz, m. & hormozi, f. (2016) heat transfer, pressure drop and fouling studies of multiwalled carbon nanotube nanofluids inside a plate heat exchanger. experimental thermal and fluid science, 2016, 72, 1-11. serebryakova, m.a. dimov, s.v., bardakhanov, s. p. & novopashin, s. a. (2015) thermal conductivity, viscosity and rheology of a suspension based on al2o3 nanoparticles and mixture of 90% ethylene glycol and 10% water, international journal of heat and mass transfer, 83, 187-191. taghizadeh-tabari, z., zeinali heris, s., moradi, m. & kahani, m. (2016) the study on application of tio2/water nanofluid in plate heat exchanger of milk pasteurization industries. renewable and sustainable energy reviews, 58, 1318-1326. tiwari, a.k., ghosh, p. & sarkar, j. (2013). performance comparison of the plate heat exchanger using different nanofluids. experimental thermal and fluid science, 49, 141-151. tiwari, a.k., ghosh, p., sarkar, j., dahiya, h. & parekh, j. (2014) numerical investigation of heat transfer and fluid flow in plate heat exchanger using nanofluids. international journal of thermal sciences, 85, 93-103. yang, j., jacobi, a. & liu, w. (2017) heat transfer correlations for single-phase flow in plate heat exchangers based on experimental data. applied thermal engineering, 113, 15471557. international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (4), 648-654 | 648 https://doi.org/10.14710/ijred.2023.52532 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id investigating the potential of avocado seeds for bioethanol production: a study on boiled water delignification pretreatment herliati rahman* , ayu nehemia , hadiatun puji astuti department of chemical engineering, faculty of industrial technology, university of jayabaya, indonesia abstract. the increasing need for alternative fuels to replace fossil fuels has made bioethanol a promising option. although numerous sources of sugar generation and agricultural wastes can be converted into ethanol, avocado seeds (as) are particularly attractive as raw materials due to their abundance, high carbohydrate content, and lack of interactions with the food chain. therefore, this study investigated the potential of as for bioethanol production using several steps, including boiled water delignification pretreatment, catalytic hydrolysis, and fermentation with saccharomyces cerevisiae. the delignification pretreatment of as involved soaking in 4% (w/v) sodium hydroxide liquor for 24 hours. then the mixture was heated to 80°c and stirred slowly for 2.5 hours and after that washing with boiled water at 100 oc for 1.5 hours and screening the mixture. subsequently, catalytic hydrolysis and fermentation were carried out using two different concentrations of saccharomyces cerevisiae as yeast, namely 10% (w/v) and 15% (w/v). qualitative sample analysis was conducted using scanning electron microscopy (sem) to observe the effect of delignification pretreatment, while ftir analysis using thermo scientific nicolet is50 was used to test for glucose functional groups. quantitative analysis was performed using gas chromatography 7890b mass spectrophotometry 5977a, agilent dbvrx to determine hydrolysate fermentation. the results revealed that the highest ethanol yield was achieved through fermentation with 15% (w/v) yeast and 40% (v/v) catalyst, resulting in an ethanol yield of 83.755% of the theoretical maximum. keywords: agricultural waste; enzyme; fermentation; hydrolysis @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 16th feb 2023; revised: 14th april 2023; accepted: 20th may 2023; available online: 25th may 2023 1. introduction the global economy is in need of a transition towards more sustainable and environmentally friendly energy sources, as the reliance on fossil fuels remains high (deby et al., 2014; fadhil et al., 2017). therefore, biofuels are being considered as potential replacements for traditional fuels such as petrol, diesel, and aerostatic fuels (ahlgren et al., 2017; rahman herliati et al., 2018). simulation studies of the world energy systems predict that biofuels could contribute significantly, ranging from 10% to 40% of the market, in the long term, indicating a substantial increase in their utilization (acevedo-garcía et al., 2018). in indonesia, the production of biofuels at a competitive price is being explored to support the economic and energy security of the country (raza et al., 2021). several studies have considered biomass as the most promising renewable carbon source for biofuels (yu et al., 2020; zhao et al., 2009). the national research and innovation agency of indonesia predicts that new renewable energy, including biomass, will increase power and heat generation by 2035 (frankowski et al., 2022; rahman et al., 2019; sluiter et al., 2011). bioethanol, derived from fruit waste biomass such as avocado seeds, is a viable biofuel option (dong et al., 2019; frankowski et al., 2022; salehi et al., 2018). this biomass has several advantages, such as low cost, low dependence on the food chain, and colossal availability (risyad et al., 2016; * corresponding author email: herliati@jayabaya.ac.id (h. rahman) muhammad et al., 2020; mueansichai et al., 2022). according to the central bureau of statistics indonesia (bps), 307.3 tons of avocados were produced in indonesia in 2014 (marlina et al., 2018; sukaryo & sri subekti, 2017). the production rate continues to increase yearly, at a growth level of 24.48%, raising the number of avocado seeds (sukaryo & sri subekti, 2017). currently, avocado seeds are indiscriminately discharged into the environment, thereby leading to pollution (risyad et al., 2016; sluiter et al., 2011). however, there is growing interested in utilizing avocado seeds as a crucial biomass resource for bioethanol production, given their significant quantity and high cellulose content (baruah et al., 2018; paredes-sánchez et al., 2021). according to the food and agriculture organization (fao), avocado (persea americana) is a tropical or subtropical fruit native to south america and widely grown in asia (60%), including indonesia (hurtado-fernández et al., 2018; janice et al., 2018). the primary waste is avocado seeds, with a ratio of about 0.33 kg of seeds/kg of avocado (acevedo-garcía et al., 2018; ruiz et al., 2013). due to its availability and high cellulose content (baruah et al., 2018), [19], as is regarded as an expected raw material for bioethanol. the first step in ethanol manufacturing is the generation of glucose or simple sugars from biomass and avocado seeds (acevedo-garcía et al., 2018; ruiz et al., 2013). additionally, in 2004, werby and petersen stated that biomass can yield 12 research article https://doi.org/10.14710/ijred.2023.52532 https://doi.org/10.14710/ijred.2023.52532 mailto:herliati@jayabaya.ac.id https://orcid.org/0000-0002-1319-6458 https://orcid.org/0009-0009-1317-7378 https://orcid.org/0009-0004-7450-9296 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.52532&domain=pdf h. rahman et al int. j. renew. energy dev 2023, 12(4), 648-654 |649 issn: 2252-4940/© 2023. the author(s). published by cbiore building blocks of sugars through biochemical routes, further highlighting the potential of biomass as a renewable resource for ethanol production. the compositions of as are cellulose, hemicellulose, and lignin, forming a complex composition (janice et al., 2018). however, due to their high molecular weight, these components are challenging for microorganisms to efficiently digest into tinny molecular-weight sugars (krajang et al., 2021). lignin can prevent microorganisms from producing ethanol in subsequent processes. therefore, pretreatment of as is necessary to separate lignin and make carbohydrates more accessible for hydrolysis and fermentation (ghazanfar et al., 2022; sultan et al., 2022). this typically involves stages such as boiled water delignification and catalytic hydrolysis using liquor hcl, followed by fermentation, making it a promising area of study for bioethanol production. numerous studies have explored diverse biomass pretreatment methods for obtaining carbohydrate materials from agricultural residues (as). some of the methods examined are steam explosion (se) (acevedo-garcía et al., 2018; ruiz et al., 2013), enzymatic hydrolysis (zakaria et al., 2014), liquid hot water hydrolysis (pérez et al., 2008), and enzymatic hydrolysis combined with ultrasound (subhedar & gogate, 2013). previous researchers using steam explosion during biomass pretreatments to obtain specific sugar raw materials. however, some disadvantages of se include modifying the lignin compounds into chemicals by hemicellulose-derived sugars that can inhibit the following steps; and the prospect possibility of extractives breaking down during the pretreatment (chen et al., 2022). so, this study using boiled water (100oc) during the pretreatment which is more interesting from an ecological and economic perspective. this study chose hydrochloric acid as a catalyst for the hydrolysis stage due to its cost-effectiveness compared to other acids, such as phosphoric and sulfuric acid, which can increase production costs (acevedo-garcía et al., 2018; ruiz et al., 2006). furthermore, hydrochloric acid has been found to exhibit significant effectiveness in digesting the hemicellulose fraction in dilute concentrations (velmurugan & muthukumar, 2011). the use of liquor hydrochloric acid as a catalyst in the hydrolysis of as for obtaining some sugar has never been carried out in previous studies. treating all the pretreated biomass in subsequent steps of catalytic hydrolysis and fermentation is environmentally and economically beneficial. the concentration of lignocellulosic chemicals in the pretreated biomass tends to vary in accordance with factors such as biomass source, pretreatment parameters (duration and temperature), and the presence or absence of base cooking liquor (acevedo-garcía et al., 2018). in some studies, the slurry was filtered and washed, and then the solid cellulose-rich material was separated from the liquid phase. the liquid phase typically contains small amounts of acetic acid, hemicellulose, lignin, degraded carbohydrates, and other substances (subhedar & gogate, 2013; zakaria et al., 2014). preliminary studies stated that after biomass pretreatment, catalytic hydrolysis and fermentation should be carried out simultaneously in a single reactor (liu & dien, 2022). however, the fundamental drawback of the aforementioned step in comparison to a single hydrolysis-fermentation arrangement is that it is typically carried out at low operation temperatures, corresponding to the inefficient activity of cellulolytic enzymes. according to chen et al. (2022), performing saccharification as a separate step prior to in-situ hydrolysis-fermentation is an option, as it can lower the viscosity of the slurry at high substrate concentrations. the glucose in the biomasses can be improved by using a suitable microorganism (kim, 2018). according to ghazanfar et al. (2022), saccharomyces cerevisiae ferments glucose through hydrolysis. previous studies have shown that this yeast can produce high yields of ethanol from various biomass materials, achieving up to 0.24 g ethanol per gram of biomass (acevedo-garcía et al., 2018; ruiz et al., 2013; song et al., 2021). the primary objective of this study is to explore the production of next-generation bioethanol using as pretreated with bd, followed by the subsequent steps of hydrolysis and fermentation using saccharomyces cerevisiae. 2. materials and methods 2.1 raw materials preparation a total of 500 grams of avocado seeds sourced from cafes and restaurants in bogor, west java, indonesia, were thoroughly washed to remove impurities. the seeds were sliced into thin pieces and dried in an oven at 105°c for 24 hours (fülöp & ecker, 2020). additionally, the oven-dried seeds were ground in a laboratory hammer mill using a herzog grinding device, and then sieved to a pass size of 50 mesh before being stored at room temperature. the moisture content and raw material composition of the prepared avocado seeds were determined using the standard technique of the national renewable energy laboratory (nrel) (sluiter et al., 2011). the prepared raw material is shown in figure 1. 2.2 experimental process the experimental process started with weighing 100 grams of asp, which was then mixed with 150 ml of 4% naoh solution and 1350 ml of distilled water and soaking for 24 hours. the mixture was heated to 80°c and stirred slowly for 2.5 hours. afterward, it was filtered and rinsed with boiled water at 100°c to remove the lignin content. the filtered solids were then dried in an oven at 105°c for 24 hours to obtain dried asp, which was characterized using sem with an olympus type quanta 650. another part of the experiment involved the hydrolysis of avocado seeds that were previously treated to remove lignin (chhouk et al., 2017). the hydrolysis reaction was carried out in a 500 ml reaction vessel at 60°c, using 10 g of asp, 100 ml of 30% and 40% hcl solutions. once the hydrolysis reaction was completed in about an hour, the resulting output, which contained high sugar concentration, was fig 1 avocado seed powder (asp) preparation h. rahman et al int. j. renew. energy dev 2023, 12(4), 648-654 |650 issn: 2252-4940/© 2023. the author(s). published by cbiore transferred to a fermenter. saccharomyces cerevisiae (ss) yeast was used for fermentation, with two different concentrations of ss, 10% and 15%, along with additional nutrients such as ammonium sulfate and urea as a nitrogen source (co(nh₂)₂) to support microorganism growth. the overall experimental steps are shown in figure 2. 2.3 testing methods the nrel laboratory analytical process, was employed in this study to quantify the amount of lignin and carbohydrates in biomass (sluiter et al., 2011). ftir was utilized to analyze organic molecules within the ir range of 4000 cm-1400 cm-1, specifically focusing on the structure and functional groups of glucose analytes. the ftir instrument used in this study was thermo scientific nicolet is50, equipped with a multi-coated, conical-shaped germanium tip crystal with a 350-micron spherical finish, a single reflection, throughput > 50%, and a 27° average angle. it also featured an atr (attenuated total reflectance) detector with a liquid nitrogen-cooled mct-a detector (fülöp & ecker, 2020). 2.4 microorganism and growth conditions this study was carried out using ss, a bacterium known for its ability to ferment glucose (mishra et al., 2016). one gram of each sample was diluted in buffered saline and plated on tributyrin agar (tba) plates in 100 μl aliquots. the tba plates were composed of 0.5% peptones, 0.3% yeast extract, 1% agar, and 0.1 ml tributyrin, with the ph adjusted to 5.5. prior to use, tributyrin was sterilized through membrane filtration and the filtrate was added to the basic growth medium. after 24 hours of incubation at 30°c, each colony was selected and streaked to obtain pure cultures (godoy et al.,2018). 2.5 characteristics of the fermentation the experiments were conducted in duplicate using 500-ml fermenters made of glassware equipment from pyrex, germany, with a working volume of 250 ml. the initial substrate concentration of yeast was set at 10% (w/v) and 15% (w/v) for different trials. in addition, the ph of the substrate was adjusted from 4 to 5. temperature, rpm and ph were the only parameters that were set and controlled for each fermentation. periodic monitoring and analysis were carried out by examining 1 ml samples from the fermenters. prior to analysis using gc-ms, the samples were prepared according to the established protocol. 2.6 yields calculation the yield from hydrolysis was calculated using equation (1) and is expressed as g glucose per 100 g carbohydrate present in the original material (acevedo-garcía et al., 2018): % 𝒀𝒊𝒆𝒍𝒅 = (𝑮−𝑮𝟎) 𝒙 𝟎.𝟎𝟑𝒙𝟏𝟎𝟎 𝑪 𝒙 𝟏𝟎𝟎 (1) where [g], [g]0, [c] and [0.03] denote the quantity of glucose produced by the reaction, the amount of glucose in the feed raw material, the amount of carbohydrate presents in 100 g of the original material, and the amount of catalyst at 4% (w/v), respectively. fermentation yields were determined using equation (2) (acevedo-garcía et al., 2018): % 𝒀𝒊𝒆𝒍𝒅 = [𝑬] [𝑮]+[𝑪] 𝒙 𝟏𝟎𝟎 (2) where [g] and [c] are the substrate's starting concentrations of glucose and cellulose (mg/ml), and [e] is the concentration of ethanol in the fermentation broth (mg/ml) (acevedo-garcía et al., 2018). 3. results dan discussion 3.1 samples composition and bd-pretreatment table 1 shows the carbohydrate and lignin content of the as used, which exhibit slight variations from other studies due to fig 2 study flowchart for bioethanol production from as h. rahman et al int. j. renew. energy dev 2023, 12(4), 648-654 |651 issn: 2252-4940/© 2023. the author(s). published by cbiore factors such as diverse growing locations, farming practices, and analytical procedures (ifesan et al., 2015). according to ftir analysis, the dry as showed a total carbohydrate percentage of over 50%. however, some organic chemicals and trace elements, such as starches, resins, and gums, present in smaller concentrations, were not examined (acevedo-garcía et al., 2018). this study is comparable to those described by preliminary studies (ji et al., 2022; liu & dien, 2022; pérez et al., 2008), where the percentages for carbs and lignin ranged from 50 to 75% and 17 to 20%, respectively. this carbohydrate composition indicates that as is a potential material for bioethanol production (ji et al., 2022). however, the lignin fractions in as may hinder converting sugars to bioethanol, which can be alleviated through boiled water lignification pretreatment. figure 3 shows the impact of bd pretreatment on the lignin content, as observed through sem micrograph images. based on the micrograph image, at magnifications of 1000x (figure 3a) and 5000x (figure 3b), as samples without pretreatment showed less cracks or pores in the lignin matrices. however, samples subjected to bd pretreatment figures 3c and 3d at magnificent of 100x and 5000x, exhibited more cracks or pores within the lignin matrices. this result indicates a reduction in the lignin fraction, primarily due to the base solution of sodium hydroxide and thermally degradable lignin (ji et al., 2022). 3.2 hydrolysis after the delignification stage, the filtrate was subjected to hydrolysis by adding hydrochloric acid (hcl) in a ratio of 2:1 to carbohydrates. the mixture was stirred using a magnetic stirrer and heated at 60°c for 2.5 hours. two different concentrations of hcl, 30% and 40% (v/v), were used as catalysts to accelerate the hydrolysis reaction rate. the hydrolysis reaction was carried out in excess water to ensure a pseudo-first order reaction, where changes in carbohydrate concentration determine the reaction rate. at the end of the hydrolysis reaction, the resulting mixture was cooled to room temperature, filtered, and the filtrate, rich in glucose, was used as the feed for the fermentation process. prior to fermentation, the filtrate was pasteurized at 70°c for 15 minutes and adjusted to ph 5 as required. different concentrations of solid substrate (ss), 10% (w/v) and 15% (w/v), were added to each sample, and then the samples were incubated at 30°c for 4 to 6 days in a tightly closed environment to allow anaerobic fermentation. the results of the reaction were analyzed using a refractometer, and table 2 shows that the glucose content in the hydrolyzed samples using 30% and table 1 composition of as (n=3 ± sd) compounds % dry weight carbohydrate 54.85 ± 0.74 lignin 19.32 ± 1.20 fig 3 sem analysis of as without bd (3a, 3b) and with bd pretreatment (3c, 3d) table 2 analysis of hydrolysate from pretreatment of as (n=3 ± sd) catalyst concentration glucose (% brix) hcl 30 % 27.5 ± 0.05 hcl 40 % 29.5 ± 0.05 h. rahman et al int. j. renew. energy dev 2023, 12(4), 648-654 |652 issn: 2252-4940/© 2023. the author(s). published by cbiore 40% hcl were 27.5% and 29.5% brix, respectively. this result indicates that the hydrolysis process to convert carbohydrates into glucose has been successful. in addition, the functional groups of glucose were analyzed using ftir analysis with a thermo scientific nicolet is50 type. the wave number range for glucose functional groups was determined to be between 2500-4000 cm-1, which includes three main functional groups, namely c-h, n-h and o-h with wave numbers of 2800-3000 cm-1, 3300-3600 cm-1, and 3300-3600 cm1, respectively. even though n-h and o-h groups have similar wave numbers, both can be differentiated by their spectral shapes. a widened and halved spectral shape characterized the n-h group, while the o-h group also has a broadened but not halved spectral form. in addition, the wave numbers 2200 cm-1, 2100 cm-1 and 1600-1800 cm-1 are in c≡n, c≡c and three functional groups, namely c=o, c= c, and c=n (r et al., 2017; yu et al., 2020). figure 4 shows the spectrum of glucose, confirming its presence in the hydrolysate with a molecular formula of c6h12o6, indicated by a single broad-spectrum alcohol functional group (o-h) at a wave number of 3340.79 cm1. therefore, the test results shown in figure 4 confirm that the hydrolysate contains the dominant glucose. 3.3 fermentation of hydrolysate the fermentation of glucose in the hydrolysate is the subsequent process after catalytic hydrolysis. in this study, the effect of substrate concentration on the fermentation reaction was investigated using saccharomyces cerevisiae at concentrations of 10% and 15% (w/v). s. cerevisiae is an acidophilic microbe that thrives at a maximum ph of 5 (rahman et al., 2018). therefore, fermentation was carried out under acidic conditions with a ph range of 4-5. analysis of the fermentation results indicated that a slightly higher ethanol content was obtained at a substrate concentration of 15% (w/v) compared to 10% (w/v). the fermented ethanol was subsequently separated by distillation and analyzed to determine the ethanol content obtained. table 3 revealed that the highest bioethanol content of 83.755% was achieved at a concentration of 10% substrate with a fermentation time of 6 days. in contrast, the bioethanol content at a substrate concentration of 10% was lower than that of the 15% substrate. this observation is attributed to the low initial yeast population, which was unable to optimally break down the available glucose into alcohol due to an insufficient number of cells proportional to the medium used, resulting in an extended lag phase (tan et al., 2019). additionally, temperature played a critical role in ethanol production, as it influenced the yeast growth rate. when the temperature is too high, above 37 oc, it can cause the deactivation of the yeast itself. since the fermentation reaction is exothermic, the heat released must be controlled using a cooling system (mishra et al., 2016). quantitative testing of fermented ethanol content using gcms was carried out by injecting the sample using a syringe into the injector. the sample was then evaporated and carried by the carrier gas to the separation column, where an interaction occurs between the mobile and stationary phases. each component in a mixture interacts at a different rate depending on the volatility of that compound. the most volatile component interacts the fastest with the stationary phase, exiting the column first, followed by the other, heavier components. each component is bombarded with electrons through the ionizing chamber resulting in ionization. the ion fragments are then received by the detector and displayed as a mass spectrum, as shown in figures 5 and 6, which depict the ethanol chromatogram from the fermentation. the chromatograms indicate that the ethanol yield was slightly higher at 15% ss concentration compared to 10% ss, as seen from the area of the chromatogram. however, these results are also better than those reported by other researchers who investigated different biomass and fermentation bacteria, such as olive tree pruning fig 4 ftir spectrum of glucose table 3 ethanol yield analysis (n=3±sd) reaction time (day) ss concentration (%) 10 15 4 83.145±2.0275 83.320±0.2748 6 84.345±0.8154 83.755±0.0686 h. rahman et al int. j. renew. energy dev 2023, 12(4), 648-654 |653 issn: 2252-4940/© 2023. the author(s). published by cbiore by e-coli, which achieved 72-82% of the theoretical maximum (ruiz et al., 2006). 4. conclusions in conclusion, this study examined the feasibility of producing bioethanol from avocado seeds through a series of tests. the result showed that avocado seeds are a promising raw material for next-generation ethanol production due to their high sugar content. the study employed boiled water delignification pretreatment of the avocado seeds prior to hydrolysis and fermentation, which allowed for the extraction of lignin from the raw material in subsequent steps. the maximum bioethanol yield from the hydrolysate was found to be 83.755% when compared to the theoretical yield. acknowledgments the authors are grateful to the university of indonesia for providing the testing laboratory facilities needed to carry out this study. funding: thanks to the faculty of industrial technology, jayabaya university, this paper was completed through study funds with contract number: 71.004/kontrak penelitian/fti-uj/xii/2022. conflicts of interest: the authors declare no conflicts of interest to report regarding the present study. references acevedo-garcía, v., padilla-rascón, c., díaz, m. j., moya, m., & castro, e. (2018). fermentable sugars production from acid-catalysed steam exploded barley straw. chemical engineering transactions, 70, 1939–1944. https://doi.org/10.3303/cet1870324 ahlgren, e. o., börjesson hagberg, m., & grahn, m. (2017). transport biofuels in global energy–economy modelling – a review of comprehensive energy systems assessment approaches. in gcb bioenergy (vol. 9, issue 7, pp. 1168–1180). blackwell publishing ltd. https://doi.org/10.1111/gcbb.12431 baruah, j., nath, b. k., sharma, r., kumar, s., deka, r. c., baruah, d. c., & kalita, e. (2018). recent trends in the pretreatment of lignocellulosic biomass for value-added products. in frontiers in energy research, 6(141),1–18). frontiers media s.a. https://doi.org/10.3389/fenrg.2018.00141 chen, d., cen, k., zhuang, x., gan, z., zhou, j., zhang, y., & zhang, h. (2022). insight into biomass pyrolysis mechanism based on cellulose, hemicellulose, and lignin: evolution of volatiles and kinetics, elucidation of reaction pathways, and characterization of gas, biochar and bio‐oil. combustion and flame, 242, 1–17. https://doi.org/10.1016/j.combustflame.2022.112142 chhouk, k., wahyudiono, kanda, h., & goto, m. (2017). comparison of conventional and ultrasound assisted supercritical carbon dioxide extraction of curcumin from turmeric (curcuma longa l.). engineering journal, 21(5), 53–65. https://doi.org/10.4186/ej.2017.21.5.53 deby, a., rahman, h., & widyaningrum, a. (2014). pemanfaatan sampah sayuran sebagai bahan baku bioetanol. jurnal konversi, 3(1), 13–18. https://jurnal.umj.ac.id/index.php/konversi/article/view/1097 dong, c., wang, y., wang, h., lin, c. s. k., hsu, h. y., & leu, s. y. (2019). new generation urban biorefinery toward complete utilization of waste derived lignocellulosic biomass for biofuels and value-added products. energy procedia, 158, 918–925. https://doi.org/10.1016/j.egypro.2019.01.231 fadhil, a. b., ahmed, k. m., & dheyab, m. m. (2017). silybum marianum l. seed oil: a novel feedstock for biodiesel production. arabian journal of chemistry, 10. https://doi.org/10.1016/j.arabjc.2012.11.009 frankowski, j., wawro, a., batog, j., szambelan, k., & łacka, a. (2022). bioethanol production efficiency from sorghum waste biomass. energies, 15(9). https://doi.org/10.3390/en15093132 fülöp, l., & ecker, j. (2020). an overview of biomass conversion: exploring new opportunities. peerj, 8, 1–21. https://doi.org/10.7717/peerj.9586 ghazanfar, m., irfan, m., nadeem, m., shakir, h. a., khan, m., ahmad, i., saeed, s., chen, y., & chen, l. (2022). bioethanol production optimization from koh-pretreated bombax ceiba using saccharomyces cerevisiae through response surface methodology. fermentation, 8(4), 1–19. https://doi.org/10.3390/fermentation8040148 godoy p, mourenza á, hernández-romero s, gonzález-lópez j, manzanera m. (2018). microbial production of ethanol from sludge derived from an urban wastewater treatment plant. front microbial, 9(1). https://doi.org/10.3389/fmicb.2018.02634 halimahtussaddiyah, r., mashuni, & budiarni. (2017). application of local adsorbant from southeast sulawesi clay immobilized saccharomyces cerevisiae bread’s yeast biomass for adsorption of mn(ii) metal ion. journal of physics: conference series, 846. https://doi.org/10.1088/1742-6596/846/1/012010 hurtado-fernández, e., fernández-gutiérrez, a., & carrasco-pancorbo, a. (2018). avocado fruit— persea americana. in exotic fruits. elsevier. https://doi.org/10.1016/b978-0-12-803138-4.00001-0 ifesan, b. o. t., olorunsola, b. o., & ifesan. (2015). nutritional composition and acceptability of candy from avocado seed (persea americana). international journal of agriculture innovations and research,3(6), 1631-1634. https://ijair.org/administrator/components/com_jresearch/files/ publications/ijair_1319_final.pdf figure 5 ethanol chromatogram using ss 10% figure 6 ethanol chromatogram using ss 15% https://doi.org/10.3303/cet1870324 https://doi.org/10.1111/gcbb.12431 https://doi.org/10.3389/fenrg.2018.00141 https://doi.org/10.1016/j.combustflame.2022.112142 https://doi.org/10.4186/ej.2017.21.5.53 https://jurnal.umj.ac.id/index.php/konversi/article/view/1097 https://doi.org/10.1016/j.egypro.2019.01.231 https://doi.org/10.1016/j.arabjc.2012.11.009 https://doi.org/10.3390/en15093132 https://doi.org/10.7717/peerj.9586 https://doi.org/10.3390/fermentation8040148 https://doi.org/10.3389/fmicb.2018.02634 https://doi.org/10.1088/1742-6596/846/1/012010 https://doi.org/10.1016/b978-0-12-803138-4.00001-0 https://ijair.org/administrator/components/com_jresearch/files/publications/ijair_1319_final.pdf https://ijair.org/administrator/components/com_jresearch/files/publications/ijair_1319_final.pdf h. rahman et al int. j. renew. energy dev 2023, 12(4), 648-654 |654 issn: 2252-4940/© 2023. the author(s). published by cbiore jacobus, a. p., gross, j., evans, j. h., ceccato-antonini, s. r., & gombert, a. k. (2021). saccharomyces cerevisiae strains used industrially for bioethanol production. essays in biochemistry, 65(2), 147–161. https://doi.org/10.1042/ebc20200160 janice, d. a., john, a., & jemmy, f. t. (2018). morphological characteristics of avocado (persea americana mill.) in ghana. african journal of plant science, 12(4), 88–97. https://doi.org/10.5897/ajps2017.1625 ji, h., wang, l., tao, f., yao, z., li, x., dong, c., & pang, z. (2022). a hydrotrope pretreatment for stabilized lignin extraction and high titer ethanol production. bioresources and bioprocessing, 9(1), 1-11. https://doi.org/10.1186/s40643-022-00530-6 kim, a. (2018). the influence of tape starter (saccharomyces cerevisiae) concentration toward bioethanol level in the fermentation process of avocado (persea americana mill) seed. j. akademika kim, 7(4), 173–178. http://jurnal.untad.ac.id/jurnal/index.php/jak/article/view/119 40 krajang, m., malairuang, k., sukna, j., rattanapradit, k., & chamsart, s. (2021). single-step ethanol production from raw cassava starch using a combination of raw starch hydrolysis and fermentation, scale-up from 5-l laboratory and 200-l pilot plant to 3000-l industrial fermenters. biotechnology for biofuels, 14(1). https://doi.org/10.1186/s13068-021-01903-3 liu, z. l., & dien, b. s. (2022). cellulosic ethanol production using a dual functional novel yeast. international journal of microbiology 2022, 1–12. hindawi limited. https://doi.org/10.1155/2022/7853935 marlina, l., wahyu pratama, d., kimia, t., & tedc bandung, p. (2018). pengambilan minyak biji alpukat dengan metode ekstraksi. tedc, 12(1), 31–37. https://ejournal.poltektedc.ac.id/index.php/tedc/article/view/12 8 mishra, a., sharma, a. k., sharma, s., mathur, a. s., gupta, r. p., & tuli, d. k. (2016). lignocellulosic bioethanol production employing newly isolated inhibitor and thermotolerant saccharomyces cerevisiae dbtioc s24 strain in ssf and shf. rsc advances, 6(29). https://doi.org/10.1039/c6ra00007j mueansichai, t., rangseesuriyachai, t., thongchul, n., & assabumrungrat, s. (2022). lignocellulosic bioethanol production of napier grass using trichoderma reesei and saccharomyces cerevisiae co-culture fermentation. international journal of renewable energy development, 11(2), 423-433. https://doi.org/10.14710/ijred.2022.43740 muhammad, rina ridara, & masrullita. (2020). sintesis bioplastik dari pati biji alpukat dengan bahan pengisi kitosan. jurnal teknologi kimia unimal, 9(2), 1–11. https://doi.org/10.29103/jtku.v9i2.3340 paredes-sánchez, b. m., paredes-sánchez, j. p., & garcía-nieto, p. j. (2021). evaluation of implementation of biomass and solar resources by energy systems in the coal-mining areas of spain. energies, 15(1), 232. https://doi.org/10.3390/en15010232 pérez, j. a., ballesteros, i., ballesteros, m., sáez, f., negro, m. j., & manzanares, p. (2008). optimizing liquid hot water pretreatment conditions to enhance sugar recovery from wheat straw for fuelethanol production. fuel, 87(17–18), 3640–3647. https://doi.org/10.1016/j.fuel.2008.06.009 rahman, a., prihantini, n. b., & nasruddin. (2019). fatty acid of microalgae as a potential feedstock for biodiesel production in indonesia. aip conference proceedings, 2062. https://doi.org/10.1063/1.5086606 rahman, h., sefaniyah, & indri, a. (2018). pemanfaatan limbah kulit pisang sebagai bahan baku pembuatan bioetanol. jurnal teknologi, 6(1), 1–10. https://doi.org/10.31479/jtek.v6i1.1 raza, m., inayat, a., & abu-jdayil, b. (2021). crude glycerol as a potential feedstock for future energy via thermochemical conversion processes: a review. sustainability (switzerland), 13(22). https://doi.org/10.3390/su132212813 risyad, a., permadani, r. l., & mz, s. (2016). ekstraksi minyak dari biji alpukat (persea americana mill) menggunakan pelatur heptana. jurnal teknik kimia usu,5(1), 34-39. https://doi.org/10.32734/jtk.v5i1.1522 ruiz, e., cara, c., álvarez-díaz, ballesteros, i., negro, m., & castro, e. (2006). enhanced enzymatic hydrolysis of olive tree wood by steam explosion and alkaline peroxide delignification. process biochemistry, 41. https://doi:10.1016/j.procbio.2005.07.007 salehi, r., taghizadeh-alisaraei, a., jahanbakhshi, a., & shahidi, f. (2018). evaluation and measurement of bioethanol extraction from melon waste (qassari cultivar). agricengint, 20(3), 127–131. https://cigrjournal.org/index.php/ejounral/article/view/4821 scott, f., quintero, j., morales, m., conejeros, r., cardona, c., & aroca, g. (2013). process design and sustainability in the production of bioethanol from lignocellulosic materials. electronic journal of biotechnology ,16(3) https://doi.org/10.2225/vol16-issue3-fulltext7 sluiter, a., hames, b., ruiz, r., scarlata, c., sluiter, j., templeton, d., & crocker, d. (2011). determination of structural carbohydrates and lignin in biomass: laboratory analytical procedure (lap); issue date: april 2008; revision date: july 2011 (version 07-08-2011). http://www.nrel.gov/biomass/analytical_procedures.html song, b., lin, r., lam, c. h., wu, h., tsui, t. h., & yu, y. (2021). recent advances and challenges of inter-disciplinary biomass valorization by integrating hydrothermal and biological techniques. renewable and sustainable energy reviews, 135, https://doi.org/10.1016/j.rser.2020.110370 subhedar, p. b., & gogate, p. r. (2013). intensification of enzymatic hydrolysis of lignocellulose using ultrasound for efficient bioethanol production: a review. industrial and engineering chemistry research, 52(34), 11816–11828). https://doi.org/10.1021/ie401286z sukaryo, & sri subekti. (2017). bioetanol dari limbah biji alpokat di kabupaten semarang. jurnal neo teknika, 3(1), 29–34. sultan, i. n., khienpanya, n., tareen, a. k., laemsak, n., sirisansaneeyakul, s., vanichsriratana, w., & parakulsuksatid, p. (2022). kinetic study of ethanol production from different sizes of two-step pretreated oil palm trunk by fed-batch simultaneous saccharification and fermentation. agriculture and natural resources, 56(2), 287–298. https://doi.org/10.34044/j.anres.2022.56.2.07 tan, j. s., phapugrangkul, p., lee, c. k., lai, z.-w., abu bakar, m. h., & murugan, p. (2019). banana frond juice as novel fermentation substrate for bioethanol production by saccharomyces cerevisiae. biocatalysis and agricultural biotechnology, 21. https://doi.org/10.1016/j.bcab.2019.101293 velmurugan, r., & muthukumar, k. (2011). utilization of sugarcane bagasse for bioethanol production: sono-assisted acid hydrolysis approach. bioresource technology, 102(14), 7119–7123. https://doi.org/10.1016/j.biortech.2011.04.045 yu, i. k. m., hanif, a., tsang, d. c. w., shang, j., su, z., song, h., ok, y. s., & poon, c. s. (2020). tuneable functionalities in layered double hydroxide catalysts for thermochemical conversion of biomass-derived glucose to fructose. chemical engineering journal, 383, 122914. https://doi.org/10.1016/j.cej.2019.122914 zakaria, m. r., fujimoto, s., hirata, s., & hassan, m. a. (2014). ball milling pretreatment of oil palm biomass for enhancing enzymatic hydrolysis. applied biochemistry and biotechnology, 173(7), 1778– 1789. https://doi.org/10.1007/s12010-014-0964-5 zhao, x., cheng, k., & liu, d. (2009). organosolv pretreatment of lignocellulosic biomass for enzymatic hydrolysis. applied microbiology and biotechnology, 82(5), 815–827. https://doi.org/10.1007/s00253-009-1883-1. © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1042/ebc20200160 https://doi.org/10.5897/ajps2017.1625 https://doi.org/10.1186/s40643-022-00530-6 http://jurnal.untad.ac.id/jurnal/index.php/jak/article/view/11940 http://jurnal.untad.ac.id/jurnal/index.php/jak/article/view/11940 https://doi.org/10.1186/s13068-021-01903-3 https://doi.org/10.1155/2022/7853935 https://ejournal.poltektedc.ac.id/index.php/tedc/article/view/128 https://ejournal.poltektedc.ac.id/index.php/tedc/article/view/128 https://doi.org/10.1039/c6ra00007j https://doi.org/10.14710/ijred.2022.43740 https://doi.org/10.3390/en15010232 https://doi.org/10.1016/j.fuel.2008.06.009 https://doi.org/10.1063/1.5086606 https://doi.org/10.31479/jtek.v6i1.1 https://doi.org/10.3390/su132212813 https://doi.org/10.32734/jtk.v5i1.1522 https://doi:10.1016/j.procbio.2005.07.007 https://cigrjournal.org/index.php/ejounral/article/view/4821 https://doi.org/10.2225/vol16-issue3-fulltext-7 https://doi.org/10.2225/vol16-issue3-fulltext-7 http://www.nrel.gov/biomass/analytical_procedures.html https://doi.org/10.1016/j.rser.2020.110370 https://doi.org/10.1021/ie401286z https://doi.org/10.34044/j.anres.2022.56.2.07 https://doi.org/10.1016/j.bcab.2019.101293 https://doi.org/10.1016/j.biortech.2011.04.045 https://doi.org/10.1016/j.cej.2019.122914 https://doi.org/10.1007/s12010-014-0964-5 https://doi.org/10.1007/s00253-009-1883-1 international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (3), 627-634 | 627 https://doi.org/10.14710/ijred.2023.52775 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id modeling anaerobic co-digestion of water hyacinth with ruminal slaughterhouse waste for first order, modified gompertz and logistic kinetic models erick auma omondi1* , peter kuria ndiba1, gloria koech chepkoech2, arnold aluda kegode3 1 department of civil and construction engineering, university of nairobi; p.o. box 10344-00100 nairobi, kenya. 2 department of civil and construction engineering, jomo kenyatta university of agriculture and technology; p.o. box 62000 (00200), nairobi, kenya. 3 department of civil & structural engineering, moi university; p.o box 3900-30100, eldoret, kenya. abstract. water hyacinth (eichhornia crassipes), an invasive aquatic weed with large biomass production is of socio-economic and environmental concern in fresh water bodies such as the lake victoria in east africa. efforts towards its control and removal can be complemented by biogas production for use as energy source. the co-digestion of water hyacinth (wh) with ruminal slaughterhouse waste (rsw) has the potential to improve biogas production from wh through collation of processes parameters such as the c/n and c/p ratios, potassium concentration and buffering capacity. knowledge of optimum proportion of the rsw as the minor substrate is of both process and operational importance. moreover, efficient operation of the process requires an understanding of the relationship between the biogas production and the process parameters. kinetic models can be useful tools for describing the biogas production process in batch reactors. while the first order kinetics models assume that the rate of biogas production is proportional to the concentration of the remaining substrates, other models such as the modified gompertz and the logistic models incorporate the lag phase, a key feature of the anaerobic digestion process. this study aimed to establish the optimum proportion of rsw in codigestion with wh under mesophilic conditions, and apply kinetics models to describe the biogas production. the study conducted batch co-digestion of wh with 0, 10, 20 and 30% rsw proportions at mesophilic temperature of 32ºc. co-digestion of wh with 30% rsw proportion improved biogas yield by 113% from 19.15 to 40.85 ch4 ml/(gvs) at 50 days of co-digestion. it also exhibited the most stable daily biogas production and the largest biogas yield. the biomethanation data were fitted with the first order kinetics, modified gompertz and the logistic models. biogas production for codigestion of wh with 30% rsw proportion was best described by the modified gompertz model with a biogas yield potential, mo, of 43.2 ml (gvs)1d-1; maximum biogas production rate, rm, of 1.50 ml (gvs)-1d-1; and duration of lag, λ, of 3.89 d. keywords: kinetics, modified gompertz model, logistic model, first order kinetic model, anaerobic digestion @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 27th feb 2023; revised: 18th april 2023; accepted: 4th may 2023; available online: 15th may 2023 1. introduction global concerns over depletion of fossil fuel sources coupled with the need to reduce greenhouse effects necessitates the search for unconventional energy sources such as waste biomass. production of biofuel from waste biomass would also double up as a sustainable waste management strategy for the biomas (mmusi et al., 2021). accordingly, there is increased interest in production of biofuels such as biogas, biodiesel and bioethanol that are relatively cheap, renewable and eco-friendly (ehiri et al., 2014). on the other hand water hyacinth (wh), a highly reproductive aquatic weed with a doubling period of 7 to 12 days (degaga, 2018), is of environmental concern because its high density hinders penetration of light into water bodies adversely affecting the aquatic life (mironga et al., 2012). it also interferes with the use of water bodies for transportation (honlah et al., 2019). harvesting of the plant for use as a feedstock in biogas production can reduce the associated environmental challenges and generate relatively cheap and * corresponding author email: omorric@gmail.com (a. samour) renewable energy. moreover, because the water hycinth grows on water, it does not compete with crops for agricultural land (bett, 2012). anaerobic digestion of wh alone suffers from process instability and limiting substrate composition and nutrient imbalance. for example, the wh large concentrations of cellulose, hemicellulose and carbohydrates and lesser concentration of lignin (omondi et al., 2019a) does not sufficiently buffer the ph during the acidogenic stage (omondi et al., 2019b) leading to acidic ph that cause prolonged lag phase. moreover, the wh carbon and nitrogen concentrations of 15,480 ± 350 and 1,650 ± 60 mg/kg, respectively result in a c/n ratio of 9.38 (omondi et al., 2019a), which is at the lower limit of the 8-20 optimal range for biogas production., which makes mono-digestion of wh susceptible to ammonia toxicity (kossman et al., 2007). previous studies showed co-digestion of wh with other substrates such as slaughterhouse waste stabilized the digestion research article https://doi.org/10.14710/ijred.2023.52775 https://doi.org/10.14710/ijred.2023.52775 https://doi.org/10.14710/ijred.2023.52775 mailto:omorric@gmail.com https://orcid.org/0009-0007-5716-2341 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.52775&domain=pdf omondi et al. int. j. renew. energy dev 2023, 12(3), 627-634 |628 issn: 2252-4940/©2023. the author(s). published by cbiore process and increased biogas production (omondi et al., 2019b). however, most of the slaughterhouse waste components with the exception of ruminal slaughterhouse waste (rsw), have large concentrations of proteins that make the digestion process susceptible to ammonia toxicity (chen et al., 2008). the rsw is characterized by c/n ratio of 18.75 (omondi et al., 2009a), that is the higher limit that can affect anaerobic digestion (ad) process through nitrogen deficiency. co-digestion of wh with rsw as a minor substrate can potentially balance the c/n ratio for wh, to an optimal level and improve biogas production. biogas production in batch systems is affected by the type and characteristics of biomass, nutrient availability, and the biodigester conditions such as ph and temperature (nguyen et al., 2021). the anaerobic digestion process is commonly characterized by instabilities from feedstock overload, presence of inhibitors and temperature fluctuations (gavala et al., 2003; rabii et al., 2019). consequently, the design of efficient biogas generation system requires an understanding of the relationships between biogas production process, substrate characteristics and biodigester conditions. kinetic models can be useful tools for describing these relationships. kinetic models are developed for specific objectives that may include the establishment of process parameters, process simulation, optimization, and control (kim et al., 2018; oyaro et al., 2021). consequently, the models can be applied to observe, predict, simulate, and optimize the system kinetics or mechanisms at different operation conditions (de oliveira, 2016; pramanik et al., 2019). historically, development of kinetic models was substrate specific with the aim of simulating the kinetics of substrate degradation and biogas production (momodu et al., 2021; wang et al., 2022). presently, the development and study of kinetics of biodegradation primarily considers digestion parameters such as microbial growth rate, substrate utilization rate, bio-kinetic coefficients, and growth constants (borja et al., 2003; nguyen et al., 2021; hadiyanto et al 2023). fitting experimental data with kinetic models can assist estimate process parameters such as the initial conditions, stoichiometry, and kinetic parameters. the first order kinetic model has previously been adopted in batch tests for modelling the rate of hydrolysis (feng et al., 2017; pramanik et al., 2019). however, the model is not well suited for describing the acclimatization processes that exhibit a lag in biogas production (hassan et al., 2022). the lag phase is associated with the rapid acidogenic and acetogenic stages (momodu & adepoju, 2011; lafratta et al., 2021) that depress the ph before development of sufficient methane formers to consume the acids (omondi et al., 2020). to overcome this challenge and to describe substrate consumption under the ad process, other kinetic models such as the modified gompertz and the logistic models include the duration of the lag phase. the logistics model assumes that the rate of biomethanation is proportional to the size of the microbial population as indicated by biogas production rate, and the concentration of digestible substrate that is indicated by the maximum biogas yield potential (rabii et al., 2019). the modified gompertz model, on the other hand, assumes that the rate of biomethanation is proportional to the microbial activity; however, the proportionality decreases with the solids retention time, which can be interpreted as loss of the efficiency of substrate conversion with time (donoso-bravo et al., 2010). the modified gompertz model is one of the most utilized models for the anaerobic digestion process. donoso-bravo et al. (2010) and nguyen et al. (2016) found that the model closely correlated biogas production with four biochemical reaction parameters; namely, biogas yield potential mo, maximum biogas production rate rm, methane production rate constant k, and the duration of lag λ. other kinetic studies under mesophilic conditions have recommended the use of the modified gompertz equation for design of continuous stirred tank reactor (cstr) in digestion of organic wastes originating from pulp and paper industries, food processing industries and wastewater treatment plants (linke, 2006; bakraoui et al., 2019). however, the model is associated with several drawbacks that affect the prediction of methane production (zhu et al., 2019). for example, donoso-bravo et al. (2010) found that the model tends to give higher values for negative lag phase without an objective biological explanation. similarly, li et al. (2019) observed that the model was prone to errors where substrate to inoculum ratio (s/i) exceeded 0.7. this study evaluated the biogas production for various proportions of rsw in co-digestion with wh in a single-stage batch reactor. it then compared the application of three kinetic models; namely, first order kinetic, modified gompertz, and the logistic models in describing the experimental data. 2. materials and methods 2.1 overview of methods co-digestion of wh with various proportions of rsw was conducted in a single stage batch reactor. the biogas output was measured by displacement method. biomethanation was carried out for a retention time of 60 days in a controlled mesophilic condition of 32ºc. cumulative biogas production, slurry temperatures and ph were monitored throughout the study. fig. 1 shows the schematic of the substrate preparation and biogas production and collection 2.2 sample collection and preparation of substrates water hyacinth samples were obtained from the shores of winam gulf in lake victoria, kisumu city, in kenya at coordinates 0° 5’39.71” s, 34045’2.44” e. the ruminal slaughterhouse waste was obtained from the nairobi dagoretti slaughterhouse at coordinates 1°17'3.71" s, 36°41'1.98" e. sampling for wh selected fresh, healthy and mature plants that were then transported in sampling bags to the laboratory awaiting substrate preparation. similarly, rsw was transported in plastic sample buckets. whole wh plants including leaves, fig. 1. schematic of the experimental procedure omondi et al. int. j. renew. energy dev 2023, 12(3), 627-634 |629 issn: 2252-4940/©2023. the author(s). published by cbiore stems and roots, were cut into small sizes of about 2 cm and dried under the sun for a period of 7 days for ease of storage and handling. the sun-dried wh was ground to fine particles by the use of mortar and pestle to form feedstock that was placed in plastic bags and stored in a refrigerator at 4°c. similarly, fresh rsw was dried in the sun for a period of 3 days. the dried samples were kept in plastic bags and stored for biogas production as a co-substrate. 2.3 biomethanation experimental set-up the biomethanation set-up comprised four sets of batch test apparatus. each set consisted of a constant temperature water bath, a digester, an alkaline scrubber solution chamber, a water displacement chamber and a graduated cylinder for collection of the water displaced by the gas (fig. 2). the digester comprised a two neck round bottom flask. one neck was used as an inlet for the substrate and the other one as the gas outlet. after feeding the substrate, the inlet was sealed for the duration of the test. the second chamber comprised a 1,000 ml cylindrical vessel with a gas inlet pipe immersed in an alkaline scrubber solution for co2 and other minor gases. the solution was prepared from 1 molar sodium hydroxide solution containing 40 g sodium hydroxide per 1 l of water. three drops of phenolphthalein indicator were added for monitoring ph variation; the scrubber solution was replaced when the pink/violet color of the indicator turned colorless. the scrubber solution chamber was fitted with a methane gas outlet pipe leading to a 1,000 ml water displacement chamber that was covered with an aluminum foil to prevent loss of water by evaporation. the final unit consisted of a 1,000 ml graduated cylinder for measurement of the volume of water displaced by the gas. 2.4 anaerobic digestion tests batch anaerobic digestion tests were conducted for codigestion of wh with 0, 10, 20 and 30% rsw proportions at a mesophilic temperature of 32ºc. biomass, 150 g, was fed into each reactor and the biomethanation monitored for 60 days. the mix proportions were prepared on the basis of weight/weight of the substrates as illustrated in table 1. daily biogas production was recorded as the volume of water displaced by the scrubbed gas and converted to biogas yield per gram of volatile solids. 2.5. biomethanation kinetics and data analysis the growth functions for anaerobic co-digestion of wh with rsw were fitted with models by the use of non-linear regression analysis curve-fitting tool in ibm spss software. the tool computed the correlation coefficient r2. it also computed the root mean square error (rmse) of the differences between the predicted values and the experiment data as expressed in equation 1. rmse = √ ∑ (𝑃𝑖−𝑂𝑖) 2𝑛 𝑖=1 𝑛 (1) where; pi is the model result, oi is the experimental result and n is the number of data. the studied kinetic models; namely, first order kinetic, modified gompertz and logistic models have previously been used to describe the kinetic methane production for codigestion of sewage sludge with food waste (sulaiman & seswoya, 2012); water hyacinth with poultry liter (patil et al., fig. 2. biomethanation experimental set-up table 1 mix proportions of dried substrates digester water hyacinth (g) slaughter-house waste (g) percent of co-substrate (%) rsw-0 150 nil 0 rsw-10 135 15 10 rsw-20 120 30 20 rsw-30 105 45 30 http://en.wikipedia.org/wiki/phenolphthalein omondi et al. int. j. renew. energy dev 2023, 12(3), 627-634 |630 issn: 2252-4940/©2023. the author(s). published by cbiore 2012); water hyacinth with poultry liter, cow manure and primary sludge (adiga et al., 2012); banana peels with poultry manure (nwosu-obieogu et al., 2020); and different agricultural wastes (zhang et al., 2021). the models are expressed in equations 2, 3 and 4. first order : 𝑀 = 𝑀𝑜 (1 − 𝑒𝑥𝑝(−𝑘𝑡)) (2) modified gompertz: 𝑀 = 𝑀𝑜. exp { −𝑒𝑥𝑝 [ 𝑅𝑚.𝑒 𝑀𝑜 (𝜆 − 𝑡) + 1]}(3) modified logistic: 𝑀 = 𝑀𝑜 { 1+𝑒𝑥𝑝[ 4𝑅𝑚 𝑀𝑜 +(𝑡−𝜆)+2]} (4) where: m is the cumulative biogas production, in ml (gvs) -1 at any time t in days, mo is the biogas yield potential in ml (gvs) -1, rm is the maximum biogas production rate in ml (gvs) -1, k is the first order model constant in l/day, λ is the duration of lag phase in days, e is euler’s constant (2.7183), and k is the methane production rate constant in day-1. 3. results and discussions the following sub-sections present the daily and cumulative biogas production for co-digestion of wh with 0, 10, 20 and 30% rsw proportions, compare the first order kinetics, modified gompertz and logistic models fitting of the experimental data and establish the kinetic parameters. 3.1 experimental methane yield the four studied combinations of wh and rsw substrates; namely, wh with 0, 10, 20 and 30% rsw proportions exhibited biogas production with a lag between day 3 and day 10. it was followed by rapid gas production up to day 20 to day 30 and after which there was a gradual decline to almost zero production after day 49 to day 53 (fig. 3). biogas production for wh alone (0% rsw) was characterized by large fluctuation with nil productions on some days including after the lag period. the instability was attributed to large concentration of carbohydrates, cellulose and hemicellulose in wh biomass without corresponding concentration of lignin (omondi et al., 2019a) that can cause a mismatch of hydrolysis and acidogenesis on one hand and biomethanation on the other. introduction of the lignin rich rsw (omondi et al., 2019a) in proportions of 10, 20 and 30% as co-substrates progressively stabilized the biogas production. the 30% rsw proportion substrate demonstrated the most stable production throughout the digestion period. previous work by omondi et al. (2019b), indicated no significant improvement of the cumulative production beyond the 30% rsw proportion. the cumulative biogas production of the studied substrates combinations (fig. 4) exhibited an initial fast biogas production up to day 3 followed by a lag period of up to about day 10. the lag period is associated with acidification of the substrate during the acidogenic process, which affects the subsequent methanogenesis process. co-digestion of the wh with rsw is considered to reduce the acidification by buffering the ph. the 30% rsw proportion substrate exhibited the shortest lag period and also the largest overall biogas yield. 3.2 analysis of kinetics data the fitting of the experimental biogas gas production data with kinetics models was by the ibm spss software, which searched for biogas yield potential (mo), and the first order reactions constant (k) for the first order kinetic model at the minimum residual sum of squares (rmse) and their 95% confidence fig. 3. daily methane gas production for a) wh-0% rsw, and b) 10% rsw, c) 20% rsw and d) 30% rsw at 32ºc 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 a) duration (days) d a il y c h 4 m l/ (g v s ) 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 b) duration (days) d a il y c h 4 m l/ (g v s ) 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 c) duration (days) d a il y c h 4 m l/ (g v s ) 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 d) duration (days) d a il y c h 4 m l/ (g v s ) omondi et al. int. j. renew. energy dev 2023, 12(3), 627-634 |631 issn: 2252-4940/©2023. the author(s). published by cbiore (table 2). for the modified gompertz and logistic models, the software searched for biogas yield potential (mo), maximum biogas production rate (rm), and duration of lag phase (λ). fitting of the models with the experimental results generated correlation coefficient, r2, values for cumulative methane yields. the fitted curves for first order kinetics, modified gompertz and logistic models for different substrate mixes are illustrated in fig. 5. the curves obtained by the modified gompertz and logistic models closely related with experimental curves with a typical s-shape signifying a relatively slow upward trend (lag phase) followed by a steady biomethanation. however, the curve for first order kinetic model did not depict the s-curve shape. additionally, it had a large departure from the experimental data as also confirmed by rmse values that were 2 orders of magnitude greater than those for the other two models (table 2). 3.3. discussions the largest biogas yield potential for the tested substrates was achieved for the 30% rsw substrate at 43.2 ml/(gvs) while the least was obtained for the 0% rsw at 19.7 ml/(gvs). the results show a 136% increase in biogas production for wh following co-digestion with ruminal slaughterhouse waste. the increase was attributed to synergies in the co-digestion of wh with rsw (omondi et al., 2019b). all the three studied kinetic models showed that the 20% rsw substrate achieved the highest maximum biogas production rate and the 30% rsw substrate the largest maximum biogas yield potential. both the gompertz and the logistic models described the lag phase whereby the 20% rsw had longer lag duration of 6.88 d compared to 5.04 d for 30% rsw. the longer lag duration for the 20% rsw substrate compared to the 30% rsw suggested a limited capacity of the 20% rsw substrate to buffer the ph coupled with slower development of methane formers (omondi et al., 2019b). the highest biogas production rate (rm) of 1.606 ml/(gvsd) was observed for the 20% rsw substrate compared with the lowest rate of 0.738 ml/(gvsd) for wh alone and 1.496 ml/(gvsd) for 30% rsw. the result suggested that the 20% substrate offered the optimum combination of process parameters for biomethanation after the lag phase. however, the maximum biomethanation rate could not be sustained with table 2 methane production kinetic parameters for first order kinetic, modified gompertz and logistic models substrate maximum biogas yield (m), l (gvs)-1 model parameters r2 rmse mo, ml (gvs)-1 k, rm (d-1), ml (gvs)-1d-1 λ d (a) first order kinetic model (m, mo, k) wh 18.89 20.52 0.988 n/a 0.977 2.087 10%rsw 25.59 27.41 0.970 n/a 0.971 2.283 20%rsw 35.35 38.75 0.988 n/a 0.963 5.942 30%rsw 40.66 44.37 0.992 n/a 0.983 3.363 (b) modified gompertz model (m, mo, rm, λ) wh 18.89 19.67 0.738 3.912 0.994 0.017 10%rsw 25.59 25.69 1.218 4.100 0.997 0.020 20%rsw 35.35 36.05 1.606 5.969 0.999 0.020 30%rsw 40.66 43.21 1.496 3.892 0.998 0.021 (c) logistic model (m, mo, rm, λ) wh 18.89 18.842 0.594 4.730 0.988 0.021 10%rsw 25.59 24.97 0.984 4.922 0.991 0.031 20%rsw 35.35 34.81 1.306 6.880 0.994 0.034 30%rsw 40.66 41.04 1.222 5.014 0.992 0.042 fig. 4. cumulative methane gas production for wh, 10, 20 and 30% rsw at 32ºc 0 5 10 15 20 25 30 35 40 45 0 10 20 30 40 50 60 0% 10% 20% 30% duration (days) c u m m u la ti v e c h 4 m l/ (g v s ) omondi et al. int. j. renew. energy dev 2023, 12(3), 627-634 |632 issn: 2252-4940/©2023. the author(s). published by cbiore the lesser rsw proportion, which could be caused by exhaustion of some key balancing ingredients such as the nutrients or lignin contributed by rsw. similar results were obtained for co-digestion of wh with cattle dung, which failed to sustain maximum rate of biomethanationbecause of exhaustion of complementary nutrients derived from the cattle dung. (ali et al., 2022). consequently, the substrate achieved a smaller biogas yield of 35.35 ml (gvs)-1 compared to 40.66 ml (gvs)-1 for the 30% rsw. theoretically, a continuous flow reactor may be designed to operate at the maximum biogas production rate (balmat et al., 2014; sarker et al., 2019). however, the operation would only occur over less than 13 days from day 10 to day 23 and would not consume all the prepared substrate (camacho et al., 2019). consequently, it would fail to achieve the maximum biogas yield as well as complete the waste management of wh by digestion. for the studied substrates, the maximum biogas yield increased by 115% from 18.89 to 40.66 ml (gvs)-1d-1 for 0 and 30% rsw proportions, respectively. consequently, co-digestion of wh with 30% rsw proportion provided the highest biogas production as well as the most effective digestion of wh as waste biomass. the modified gompertz model closely described the experimental data for the studied experimental substrates with correlation vectors (r2) of 0.994 0.999 compared to 0.988 0.994, and 0.963 0.983 for the logistic and first order kinetic models, respectively. these vectors demostrated that kinetic models closely fitted the experimental data for the anaerobic digestion (bakraoui et al., 2020; tobo et al., 2020; hadiyanto et al, 2023). the rmse parameter provided a more pronounced distinction between the kinetic models; the modified gompertz model exhibited the least rmse of 0.017 – 0.021, closely followed by the logistic model with 0.021 – 0.034. comparatively, the first order kinetic model had two orders of magnitude greater rsme of 2.087 5.942, which indicated its lesser suitability for describing the data. 5. conclusion the wh substrate with 30% rsw proportion exhibited the most stable daily biogas production and largest yield in co-digestion of wh with rsw. introduction of rsw to wh substrate progressively improved biomethanation rate with the 20% rsw exhibiting the maximum biogas production rate. however, the 30% rsw presented the largest cumulative biogas production over the 60 days retention period. fig. 5. experimental, first order, modified gompertz and logistic models biogas production potential for a) 0% rsw, b) 10% rsw, c) 20% rsw and d) 30% rsw proportions substrate at 32ºc 0 5 10 15 20 25 30 35 40 45 0 10 20 30 40 50 60 a) experimental gomprtz first order logistic c u m m u la ti v e c h 4 m l/ (g v s ) duration (days) 0 5 10 15 20 25 30 35 40 45 0 10 20 30 40 50 60 b) experimental gomprtz first order logistic c u m m u la ti v e c h 4 m l/ (g v s ) duration (days) 0 5 10 15 20 25 30 35 40 45 0 10 20 30 40 50 60 c) experimental gomp first order logistic c u m m u la ti v e c h 4 m l/ (g v s ) duration (days) 0 5 10 15 20 25 30 35 40 45 0 10 20 30 40 50 60 d) experimental gompertz logistic first order c u m m u la ti v e c h 4 m l/ (g v s ) duration (days) omondi et al. int. j. renew. energy dev 2023, 12(3), 627-634 |633 issn: 2252-4940/©2023. the author(s). published by cbiore the trend in the rate of biomethanation was similar for both modified gompertz and logistic models. similarly, the duration of lag for the two models followed a similar trend characterized by initial increase in biogas production followed by a lag phase that was attributed to lowering the ph by formation of acids, which suppressed methane formers. at 30% rsw proportion, the reactor ph was sufficiently buffered, which allowed uninterrupted growth of methane formers. the co-digestion was best described by the modified gompertz model with an rmse of 0.020 compared to 0.042 and 3.363 for the logistic and the first order kinetics models, respectively. the process kinetics parameters for the modified gompertz model were: (1) biogas yield potential, mo, 43.2 ml (gvs)-1; (2) maximum biogas production rate, rm, 1.50 ml (gvs)-1 d-1; (3) lag phase function, λ, 3.89 d. funding no funding was received for this work from any organization. declaration of competing interest the authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. references adiga, s., ramya, r., shankar, b.b,, patil, j.h & geetha, c.r (2012). kinetics of anaerobic digestion of water hyacinth, poultry litter, cow manure and primary sludge: a comparative study. 2nd international conference on biotechnology and environment management ipcbee vol.42 (2012) iacsit press, singapore. 73-78. https://doi.org/10.7763/ipcbee. 2012.v42.15 ali, s.s., elsamahy, t., abdelfattah, a., mustafa, a., khalil, m.a., mastropetros, s.g., kornaros, m., sun j. & azab m. (2022). exploring the potential of anaerobic co-digestion of water hyacinth and cattle dung for enhanced biomethanation and techno-economic feasibility. fuel, 329(1):125397, https://doi.org/10.1016/j.fuel.2022.125397 bakraoui, m., karouach, f., ouhammou, b., aggour, m., essamri, a. & el bari, h. (2019). kinetics study of the methane production from experimental recycled pulp and paper sludge by cstr technology. journal of material cycles and waste management, 1426-1436. https://doi.org/10.1007/s10163-019-00894-6. bakraoui, m., karouach, f., ouhammou, b., lahboubi, n., el gnaoui, y., kerrou, o., aggoura, m., el bari, h. (2020).kinetics study of methane production from anaerobic digestion of sludge and wastewater recycled pulp and paperby different models simulation. international journal of smart grid and clean energy, 9(1), 170-179. https://doi.org/10.1088/1757899x/946/1/012009 balmant, w., oliveira, b.h., mitchell, d.a., vargas, j.v.c., ordonez, j.c. (2014). optimal operating conditions for maximum biogas production in anaerobic bioreactors. applied thermal engineering, 62(1), 197-206. https://doi.org/10.1016/j.applthermaleng.2013.09.033 bett, m (2012). a review of techniques for management of invasive plant species. university of eldoret. unpublished version. borja, r., martín, a., rincon, b. & raposo, f. (2003). kinetics for substrate utilization and methane production during the mesophilic anaerobic digestion of two phases olive pomace (tpop). journal of agricultural and food chemistry 51(11), 33903395. https://doi.org/10.1021/jf021059n . camacho, g., ruggeri, c.e., mangialardi, l., perisco, m. & malave a.c.l. (2019). continuous two-step anaerobic digestion (tsad) of organic market waste: rationalising process parameters. int j energy environ eng 10, 413–427. https://doi.org/10.1007/s40095-019-0312-1 chen, y., cheng, j.j, creamer, k.s. (2008). inhibition of anaerobic process, a review. bioresour technol. 99, 4044–4064. https://doi.org/10.1016/j.biortech.2007.01.057 de oliveira, p.l, hudebine, d.p, guillaume, d.p & verstraete, j.j (2016). a review of kinetic modeling methodologies for complex processes. oil & gas science & technology revue d’ifp energies nouvelles, institut français du pétrole (ifp), 2016, 71 (3), 45. https://doi.org/10.2516/ogst/2016011hal-01395195 . degaga, a.h. (2018). water hyacinth (eichhornia crassipes) biology and its impacts on ecosystem, biodiversity, economy and human well-being. j. life sci. biomed. 8(6), 94-100; www.jlsb.science-line.com donoso-bravo, a., pérez-elvira, s.i. & fdz-polanco, f. (2010). application of simplified models for anaerobic biodegradability tests. evaluation of pre-treatment processes. chemical engineering journal, 160(2), 607-614. https://doi.org/10.1016/j.cej.2010.03.082 ehiri, r.c., ikelle, i.i., mgbabor,c. & ogbu, c.c. (2014). kinetics of biogas production from a mixture of water hyacinth (eichornia crassipes) and fresh rumen residue. iosr journal of applied chemistry (iosr-jac) 7, (7),36-39. www.iosrjournals.org feng, l., gao, y., kou, w., lang, x., liu, y., li, r., yu, m., shao, l & wang, x. (2017). application of the initial rate method in anaerobic digestion of kitchen waste, biomed research international, 2017, 3808521. https://doi.org/10.1155/2017/3808521 gavala, h.n., angelidaki, i., ahring, b.k. (2003). kinetics and modeling of anaerobic digestion process. adv biochem eng biotechnol., 81, 57-93. https://doi.org/10.1007/3-540-45839-5_3 . hadiyanto, h., octafalahanda, f. m., nabila, j., jati, a. k., christwardana, m., kusmiyati, k., & khoironi, a. (2023). preliminary observation of biogas production from a mixture of cattle manure and bagasse residue in different composition variations. international journal of renewable energy development, 12(2), 390-395. https://doi.org/10.14710/ijred.2023.52446 hassan, s.r, hung, y.-t, dahlan, i & abdul aziz, h (2022). kinetic study of the anaerobic digestion of recycled paper mill effluent (rpme) by using a novel modified anaerobic hybrid baffled (mahb) reactor. water, 14(3),390. https://doi.org/10.3390/w14030390 honlah, e., yao segbefia, a., appiah, d.o., mensah, m. & atakora, p.o. (2012). effects of water hyacinth invasion on the health of the communities, and the education of children along river tano and abby-tano lagoon in ghana. cogent social sciences, 5(1): 1-18. https://doi.org/10.1080/23311886.2019.1619652 . kim, o. d., rocha, m. & maia, p. (2018). a review of dynamic modeling approaches and their application in computational strain optimization for metabolic engineering. frontiers in microbiology, 9, 1690. https://doi.org/10.3389/fmicb.2018.01690 . kossmann, w., ponitz, u., habermehl, s., hoerz, t., kramer, p., klingler, b., kellner, c., wittur, t., lopotek, f., krieg, a., euler, h. (2007) biogas digest. inf advis serv appropr technol (isat) 1,1– 46 lafratta, m., thorpe, r.b., ouki, s.k., shana, a., germain, e., willcocks, m., lee, j. (2021). development and validation of a dynamic first order kinetics model of a periodically operated well-mixed vessel for anaerobic digestion. chemical engineering journal, 426, 131732, https://doi.org/10.1016/j.cej.2021.131732 . li, p., li, w., sun, m., xu, x., zhang, b. & sun, y. (2019). evaluation of biochemical methane potential and kinetics on the anaerobic digestion of vegetable crop residues. energies, 12(1), 26; https://doi.org/10.3390/en12010026 . linke, b. (2006). kinetic study of thermophilic anaerobic digestion of solid wastes from potato processing. biomass and bioenergy 30(10), 892-896. https://doi.org/10.1016/j.biombioe.2006.02.001 marineau, e.d., perryman, m., lawler, s. & pratt, p.d. (2019). management of invasive water hyacinth as both a nuisance weed and invertebrate habitat. san francisco estuary and watershed science, 17(2), 4, 1-19. https://doi.org/10.15447/sfews.2019v17iss5 mironga, j.m., mathooko, j.m. & onywere, s. (2012). effect of water hyacinth infestation on the physicochemical characteristics of lake naivasha. international journal of humanities and social science, 2(7); 103-113. http://environmental.ku.ac.ke/images/stories/research/hyacit h_lknaivasha.pdf https://doi.org/10.7763/ipcbee.%202012.v42.15 https://www.researchgate.net/profile/sameh-ali-10?_sg%5b0%5d=eamjg3jk-xlrorplwj8ujkwav8bzgt0tb5bk9hhuwh7ss6to5dd33x8pi8ou7iruk637_ho.y2vpxxbrjnhiwo5hv5qdb-npia7ogjyybxs7byjr0jhht0ob7c8dmsukpgiluvaejkhk6vjlt1pieo-hqdqixg&_sg%5b1%5d=pkr3s15klxiwudu2pjx4ykh51azlgfxe8lvedurmcmhjidt7bi9afucsqqci_vzbfgwsjpm.e8zkoj4ex8hvudolpzo84ekqwdv0nd3zyqnrl-wy_d9u4xziqq1j-uqcawtkxgf4md4khh9xqbuywqvyaxps1q https://www.researchgate.net/profile/tamer-elsamahy?_sg%5b0%5d=eamjg3jk-xlrorplwj8ujkwav8bzgt0tb5bk9hhuwh7ss6to5dd33x8pi8ou7iruk637_ho.y2vpxxbrjnhiwo5hv5qdb-npia7ogjyybxs7byjr0jhht0ob7c8dmsukpgiluvaejkhk6vjlt1pieo-hqdqixg&_sg%5b1%5d=pkr3s15klxiwudu2pjx4ykh51azlgfxe8lvedurmcmhjidt7bi9afucsqqci_vzbfgwsjpm.e8zkoj4ex8hvudolpzo84ekqwdv0nd3zyqnrl-wy_d9u4xziqq1j-uqcawtkxgf4md4khh9xqbuywqvyaxps1q https://www.researchgate.net/profile/abdallah-abdelfattah?_sg%5b0%5d=eamjg3jk-xlrorplwj8ujkwav8bzgt0tb5bk9hhuwh7ss6to5dd33x8pi8ou7iruk637_ho.y2vpxxbrjnhiwo5hv5qdb-npia7ogjyybxs7byjr0jhht0ob7c8dmsukpgiluvaejkhk6vjlt1pieo-hqdqixg&_sg%5b1%5d=pkr3s15klxiwudu2pjx4ykh51azlgfxe8lvedurmcmhjidt7bi9afucsqqci_vzbfgwsjpm.e8zkoj4ex8hvudolpzo84ekqwdv0nd3zyqnrl-wy_d9u4xziqq1j-uqcawtkxgf4md4khh9xqbuywqvyaxps1q https://www.researchgate.net/profile/maha-azab?_sg%5b0%5d=eamjg3jk-xlrorplwj8ujkwav8bzgt0tb5bk9hhuwh7ss6to5dd33x8pi8ou7iruk637_ho.y2vpxxbrjnhiwo5hv5qdb-npia7ogjyybxs7byjr0jhht0ob7c8dmsukpgiluvaejkhk6vjlt1pieo-hqdqixg&_sg%5b1%5d=pkr3s15klxiwudu2pjx4ykh51azlgfxe8lvedurmcmhjidt7bi9afucsqqci_vzbfgwsjpm.e8zkoj4ex8hvudolpzo84ekqwdv0nd3zyqnrl-wy_d9u4xziqq1j-uqcawtkxgf4md4khh9xqbuywqvyaxps1q https://www.researchgate.net/journal/fuel-0016-2361 http://dx.doi.org/10.1016/j.fuel.2022.125397 https://doi.org/10.1007/s10163-019-00894-6 https://doi.org/10.1088/1757-899x/946/1/012009 https://doi.org/10.1088/1757-899x/946/1/012009 https://doi.org/10.1016/j.applthermaleng.2013.09.033 https://www.researchgate.net/profile/r-borja?_sg%5b0%5d=ta3va8x63d6w4pd0r3engsenkknnj3fln0dodumqxdy3vdb7lnabuyleeiec9rlxsumiouo.5jqfufmuw7qrsoinnms2gzo3qnrmxzuswsekoklwhehpnmac8r1byyzorvn6yzsl9mx_wrhaj2jxyxtr4sjxqa&_sg%5b1%5d=sylnj4mn4n6w7eedcvt1sii0m8t8w5b_k2wdcqf-8b-gkhie22zetujemsayfgsrlxrqts8.cn6bsmgdwepkefmqcbpu88-gmny315yafjk99af8dosb2z73jflww9_p42dvf1wjlz36q3x4eg3o1y0wijehxa https://www.researchgate.net/profile/antonio-martin-7?_sg%5b0%5d=ta3va8x63d6w4pd0r3engsenkknnj3fln0dodumqxdy3vdb7lnabuyleeiec9rlxsumiouo.5jqfufmuw7qrsoinnms2gzo3qnrmxzuswsekoklwhehpnmac8r1byyzorvn6yzsl9mx_wrhaj2jxyxtr4sjxqa&_sg%5b1%5d=sylnj4mn4n6w7eedcvt1sii0m8t8w5b_k2wdcqf-8b-gkhie22zetujemsayfgsrlxrqts8.cn6bsmgdwepkefmqcbpu88-gmny315yafjk99af8dosb2z73jflww9_p42dvf1wjlz36q3x4eg3o1y0wijehxa https://www.researchgate.net/profile/barbara-rincon?_sg%5b0%5d=ta3va8x63d6w4pd0r3engsenkknnj3fln0dodumqxdy3vdb7lnabuyleeiec9rlxsumiouo.5jqfufmuw7qrsoinnms2gzo3qnrmxzuswsekoklwhehpnmac8r1byyzorvn6yzsl9mx_wrhaj2jxyxtr4sjxqa&_sg%5b1%5d=sylnj4mn4n6w7eedcvt1sii0m8t8w5b_k2wdcqf-8b-gkhie22zetujemsayfgsrlxrqts8.cn6bsmgdwepkefmqcbpu88-gmny315yafjk99af8dosb2z73jflww9_p42dvf1wjlz36q3x4eg3o1y0wijehxa https://www.researchgate.net/profile/francisco-raposo?_sg%5b0%5d=ta3va8x63d6w4pd0r3engsenkknnj3fln0dodumqxdy3vdb7lnabuyleeiec9rlxsumiouo.5jqfufmuw7qrsoinnms2gzo3qnrmxzuswsekoklwhehpnmac8r1byyzorvn6yzsl9mx_wrhaj2jxyxtr4sjxqa&_sg%5b1%5d=sylnj4mn4n6w7eedcvt1sii0m8t8w5b_k2wdcqf-8b-gkhie22zetujemsayfgsrlxrqts8.cn6bsmgdwepkefmqcbpu88-gmny315yafjk99af8dosb2z73jflww9_p42dvf1wjlz36q3x4eg3o1y0wijehxa https://www.researchgate.net/journal/journal-of-agricultural-and-food-chemistry-1520-5118 http://dx.doi.org/10.1021/jf021059n https://doi.org/10.1007/s40095-019-0312-1 https://doi.org/10.1016/j.biortech.2007.01.057 https://doi.org/10.2516/ogst/2016011hal-01395195 http://www.jlsb.science-line.com/ https://www.researchgate.net/profile/andres-donoso-bravo?_sg%5b0%5d=fh-ywba_-sbmqttxhmu60orxqfdvhe6qalooxi81fuiejsqg1onai3jyy3nhwq785dsuj_u.akkmuzvkslg3sephtxefi3a6heblfh5zlchymmoxsd8-0-cov5dnv-n41tdzesh2965dsipskvtzcfyh4mcrcw&_sg%5b1%5d=jqohp3ocijx1rrc2cmlssm5gk30zy2wrs91w-y54tjweykwinclwkoxkcq4mo37isd4rfog.m55tpipyn4qgbegeep0aqmdvbfvxeihwwqnsecqto3vm6mk6dkasvdrjmqntqc_vaz0cxrykyarzvhzqpgdipq https://www.researchgate.net/scientific-contributions/si-perez-elvira-32403702?_sg%5b0%5d=fh-ywba_-sbmqttxhmu60orxqfdvhe6qalooxi81fuiejsqg1onai3jyy3nhwq785dsuj_u.akkmuzvkslg3sephtxefi3a6heblfh5zlchymmoxsd8-0-cov5dnv-n41tdzesh2965dsipskvtzcfyh4mcrcw&_sg%5b1%5d=jqohp3ocijx1rrc2cmlssm5gk30zy2wrs91w-y54tjweykwinclwkoxkcq4mo37isd4rfog.m55tpipyn4qgbegeep0aqmdvbfvxeihwwqnsecqto3vm6mk6dkasvdrjmqntqc_vaz0cxrykyarzvhzqpgdipq https://www.researchgate.net/profile/fernando-fdz-polanco?_sg%5b0%5d=fh-ywba_-sbmqttxhmu60orxqfdvhe6qalooxi81fuiejsqg1onai3jyy3nhwq785dsuj_u.akkmuzvkslg3sephtxefi3a6heblfh5zlchymmoxsd8-0-cov5dnv-n41tdzesh2965dsipskvtzcfyh4mcrcw&_sg%5b1%5d=jqohp3ocijx1rrc2cmlssm5gk30zy2wrs91w-y54tjweykwinclwkoxkcq4mo37isd4rfog.m55tpipyn4qgbegeep0aqmdvbfvxeihwwqnsecqto3vm6mk6dkasvdrjmqntqc_vaz0cxrykyarzvhzqpgdipq https://www.researchgate.net/journal/chemical-engineering-journal-1385-8947 https://www.researchgate.net/journal/chemical-engineering-journal-1385-8947 http://dx.doi.org/10.1016/j.cej.2010.03.082 http://www.iosrjournals.org/ https://doi.org/10.1155/2017/3808521 https://doi.org/10.1007/3-540-45839-5_3 https://doi.org/10.14710/ijred.2023.52446 https://doi.org/10.3390/w14030390 https://www.tandfonline.com/author/honlah%2c+emmanuel https://www.tandfonline.com/author/yao+segbefia%2c+alexander https://www.tandfonline.com/author/odame+appiah%2c+divine https://www.tandfonline.com/author/mensah%2c+moses https://www.tandfonline.com/author/atakora%2c+peter+ofori https://doi.org/10.1080/23311886.2019.1619652 https://doi.org/10.3389/fmicb.2018.01690 https://doi.org/10.1016/j.cej.2021.131732 https://doi.org/10.3390/en12010026 https://www.researchgate.net/scientific-contributions/bernd-linke-38168618?_sg%5b0%5d=0etsmxbb84b6gisxvfrxvuxknm3berntqomgbfbhz8cdtmofocxplvuwbfnlindljfbk17w.wsan3inedf3hfyge2jnebrfl3ffc55y-0zk6oniqlcfa6lcygdsqg5sroc53iv8bj-0mk-vjd-7iblewnoqnag&_sg%5b1%5d=ohobycfrw20aaiy5w2sjutgcg1mibfvt34kavo9ygrc_kocjzz9mabxgkxegkjqjehw7sju.apr6rm_oxqvsrzutf3nvgtoxmnik9dpoy2jkpa6suskwnwzu7xg7hrdm-gvovz1t2iqygg8rphzepcxzr7bkww https://www.researchgate.net/journal/biomass-and-bioenergy-0961-9534 http://dx.doi.org/10.1016/j.biombioe.2006.02.001 https://www.researchgate.net/profile/erin-donley-marineau-2?_sg%5b0%5d=njawfarfmfrwk1dpzaigtucek7lwekl0zlgtktiumnug42i_sdpy1lfyokxinv-o9d8icvi.npjd2zok6twvtikgmi1fqiorhv1l48-muvhtr05ftlhyxrt6bakgisswh7bslsrjvc96a0ujq7zq-bp2pjqn3q&_sg%5b1%5d=4pg6a4clz35uzwdj47vahdqol26-pwnvnk6sac_fsfdegcdgk2w4nuzbw9p5wryfzksiugu.itp2fpevdgzkx3gv5fqdscgra6kd5fzzipenysk1ghvuegtzfny2aftjpqllakykibcfzjoopf-feac71i3z0w https://www.researchgate.net/profile/matthew-perryman-2?_sg%5b0%5d=njawfarfmfrwk1dpzaigtucek7lwekl0zlgtktiumnug42i_sdpy1lfyokxinv-o9d8icvi.npjd2zok6twvtikgmi1fqiorhv1l48-muvhtr05ftlhyxrt6bakgisswh7bslsrjvc96a0ujq7zq-bp2pjqn3q&_sg%5b1%5d=4pg6a4clz35uzwdj47vahdqol26-pwnvnk6sac_fsfdegcdgk2w4nuzbw9p5wryfzksiugu.itp2fpevdgzkx3gv5fqdscgra6kd5fzzipenysk1ghvuegtzfny2aftjpqllakykibcfzjoopf-feac71i3z0w https://www.researchgate.net/scientific-contributions/sharon-p-lawler-11007358?_sg%5b0%5d=njawfarfmfrwk1dpzaigtucek7lwekl0zlgtktiumnug42i_sdpy1lfyokxinv-o9d8icvi.npjd2zok6twvtikgmi1fqiorhv1l48-muvhtr05ftlhyxrt6bakgisswh7bslsrjvc96a0ujq7zq-bp2pjqn3q&_sg%5b1%5d=4pg6a4clz35uzwdj47vahdqol26-pwnvnk6sac_fsfdegcdgk2w4nuzbw9p5wryfzksiugu.itp2fpevdgzkx3gv5fqdscgra6kd5fzzipenysk1ghvuegtzfny2aftjpqllakykibcfzjoopf-feac71i3z0w https://www.researchgate.net/profile/paul-pratt?_sg%5b0%5d=njawfarfmfrwk1dpzaigtucek7lwekl0zlgtktiumnug42i_sdpy1lfyokxinv-o9d8icvi.npjd2zok6twvtikgmi1fqiorhv1l48-muvhtr05ftlhyxrt6bakgisswh7bslsrjvc96a0ujq7zq-bp2pjqn3q&_sg%5b1%5d=4pg6a4clz35uzwdj47vahdqol26-pwnvnk6sac_fsfdegcdgk2w4nuzbw9p5wryfzksiugu.itp2fpevdgzkx3gv5fqdscgra6kd5fzzipenysk1ghvuegtzfny2aftjpqllakykibcfzjoopf-feac71i3z0w https://www.researchgate.net/journal/san-francisco-estuary-and-watershed-science-1546-2366 https://www.researchgate.net/journal/san-francisco-estuary-and-watershed-science-1546-2366 http://dx.doi.org/10.15447/sfews.2019v17iss5 https://www.researchgate.net/profile/john-mironga?_sg%5b0%5d=bwgvsbbfzktyjtkwczbvu6lk3_kuztclaxus9rkepk8vouo5rojvufzzvnrw0xnpmh4efje.raenz8j5cgybtaoyegep8qxvsn3z_8ddxsnwvqibjq2--jmdb1oiympola5ovglypihtysihhqdi8nllvsqltq&_sg%5b1%5d=w66rthmohsm7gjjrliz1fxt5ohrc9amzanq9g8iiv7m1kcqqb6no-pqis2ywxje3yk0-ul4.jbcism_cfwvfbjtrtbw7_pjce1yfgra4lh4xhkaghrzkoeqleaplkthogu-k6airhp9oahylcqphgtdmbba_ew https://www.researchgate.net/scientific-contributions/jude-m-mathooko-2057825712?_sg%5b0%5d=bwgvsbbfzktyjtkwczbvu6lk3_kuztclaxus9rkepk8vouo5rojvufzzvnrw0xnpmh4efje.raenz8j5cgybtaoyegep8qxvsn3z_8ddxsnwvqibjq2--jmdb1oiympola5ovglypihtysihhqdi8nllvsqltq&_sg%5b1%5d=w66rthmohsm7gjjrliz1fxt5ohrc9amzanq9g8iiv7m1kcqqb6no-pqis2ywxje3yk0-ul4.jbcism_cfwvfbjtrtbw7_pjce1yfgra4lh4xhkaghrzkoeqleaplkthogu-k6airhp9oahylcqphgtdmbba_ew http://environmental.ku.ac.ke/images/stories/research/hyacith_lknaivasha.pdf http://environmental.ku.ac.ke/images/stories/research/hyacith_lknaivasha.pdf omondi et al. int. j. renew. energy dev 2023, 12(3), 627-634 |634 issn: 2252-4940/©2023. the author(s). published by cbiore mmusi, k., mudiwa, j., rakgati, e. & vishwanathan, v. (2021). biogas a sustainable source of clean energy in sub saharan africa: challenges and opportunities. j app mat sci & engg res, 5(1), 712. momodu, a.s., adepoju, t.d. (2021). system dynamics kinetic model for predicting biogas production in anaerobic condition: preliminary assessment. science progress. 104(4), 1-25. https://doi.org/10.1177/00368504211042479. nguyen, d.d., chang, s.w., jeong, s.y., jeung, j., kim, s.s., guo, w. & ngo, h.h. (2016). dry thermophilic semi-continuous anaerobic digestion of food waste: performance evaluation, modified gompertz model analysis, and energy balance. energy conversion and management, 128:203-210. https://doi.org/10.1016/j.enconman.2016.09.066 . nguyen, t.h., nguyen, m.k., le, t., bui, t.t., nguyen, t.h., nguyen, t.q. & van ngo, a. (2021). kinetics of organic biodegradation and biogas production in the pilot-scale moving bed biofilm reactor (mbbr) for piggery wastewater treatment. journal of analytical methods in chemistry, vol. 2021(6):1-9. https://doi.org/10.1155/2021/6641796 nwosu-obieogu, k., aguele, f.o., onyenwoke, a. & adekunle, k. (2020). kinetic model comparison for biogas production from poultry manure and banana peels. european journal of sustainable development research, 4(2), 1-5, https://doi.org/10.29333/ejosdr/7595 . omondi, e.a., ndiba, p.k. & njuru, p.g. (2019a). characterization of water hyacinth (e. crassipes) from lake victoria and ruminal slaughterhouse waste as co‑substrates in biogas production. sn applied sciences, 1, 848-858. https://doi.org/10.1007/s42452019-0871-z . omondi, e.a., njuru, p.g. & ndiba, p.k. (2019b). anaerobic codigestion of water hyacinth (e. crassipes) with ruminal slaughterhouse waste for biogas production. international journal of renewable energy development, 8(3), 253-259. https://doi.org/10.14710/ijred.8.3.253-259 . omondi, e.a., ndiba, p.k., njuru, p.g. & abuga, d. (2020). dynamics of microbial communities in co-digestion of water hyacinth (eichhornia crassipes) with ruminal slaughterhouse waste under mesophilic conditions. international journal of water resources and environmental engineering, 12(4), 81-89. https://doi.org/10.5897/ijwree2020.0946 oyaro, d.k., oonge, z.i. & odira, p.m. (2021). kinetic modelling of methane production from anaerobic digestion of banana wastes. international journal of engineering research & technology (ijert), 10(3), 104-109. patil, j.h., raj, m.a., muralidhara, p.l., desai, s.m. & mahadeva raju g.k. (2012). kinetics of anaerobic digestion of water hyacinth using poultry litter as inoculum. international journal of environmental science and development, 3(2), 94-98. https://doi.org/10.7763/ijesd.2012.v3.195 pramanik, s.k., suja, f.b., porhemmat, m. & pramanik, b.k. (2019). performance and kinetic model of a single-stage anaerobic digestion system operated at different successive operating stages for the treatment of food waste. processes, 7(9), 600; https://doi.org/10.3390/pr7090600 . sarker, s., lamb, j.j., hjelme, d.r., lien, k.m. (2019). a review of the role of critical parameters in the design and operation of biogas production plants. appl. sci., 9(9), 1915. https://doi.org/10.3390/app9091915 rabii, a., aldin, s., dahman, y. & elbeshbishy, e. (2019). a review on anaerobic co-digestion with a focus on the microbial populations and the effect of multi-stage digester configuration. energies, 12(6), 1106; https://doi.org/10.3390/en12061106 . sulaiman, s.m. & seswoya, r. (2019). kinetics modelling of batch anaerobic co-digestion of domestic primary sewage sludge and food waste in a stirred reactor, iop conf. ser.: mater. sci. eng. 601, 012012. https://doi.org/10.1088/1757899x/601/1/012012 tobo, y.m., bartacek, j., nopens, i. (2020). linking cfd and kinetic models in anaerobic digestion using a compartmental model approach. processes, 8, 703. https://doi.org/10.3390/pr8060703 wang, j., liu, b., sun, m., chen, f., terashima, m. & yasui, h. (2022). a kinetic model for anaerobic digestion and biogas production of plant biomass under high salinity. int. j. environ. res. public health, 19(11), 6943. https://doi.org/10.3390/ijerph19116943 zhang, h., ann, d., cao, y., tian, y., he, j. (2021). modeling the methane production kinetics of anaerobic co-digestion of agricultural wastes using sigmoidal functions. energies 2021, 14, 258. https://doi.org/10.3390/en14020258 zhu, h., yang, j. & xiaowei, c. (2019). application of modified gompertz model to study on biogas production from middle temperature co-digestion of pig manure and dead pigs. e3s web of conferences 118:03022. https://doi.org/10.1051/e3sconf/201911803022 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1177/00368504211042479 https://www.researchgate.net/profile/dinh-duc-nguyen?_sg%5b0%5d=t2aoqgh8iwt5m5tspfy13cmvznfggmthf3cj3lshkygxaypkubuakur7qcwkxzgiqsgipyk.eca5-4e4nrfgkjotnzaguvdcyxpzcnvnvcnov338thib4z20wxtjk6087abjvj95d5wxpehqurkrbbzp3r9dpw&_sg%5b1%5d=nwhtpxpewoxepjhviyudybznbugpc7ugmgbsrix-3zneuceubymvldaztmmvdrzqz_0idjs._ffwb2dyvandqc9jdx6pb_nchadb2ozztqjzlapuq_tc4-duid1qj5mtocit46_nmrhkb2exnneo2nbzkrlscg https://www.researchgate.net/profile/soon-woong-chang?_sg%5b0%5d=t2aoqgh8iwt5m5tspfy13cmvznfggmthf3cj3lshkygxaypkubuakur7qcwkxzgiqsgipyk.eca5-4e4nrfgkjotnzaguvdcyxpzcnvnvcnov338thib4z20wxtjk6087abjvj95d5wxpehqurkrbbzp3r9dpw&_sg%5b1%5d=nwhtpxpewoxepjhviyudybznbugpc7ugmgbsrix-3zneuceubymvldaztmmvdrzqz_0idjs._ffwb2dyvandqc9jdx6pb_nchadb2ozztqjzlapuq_tc4-duid1qj5mtocit46_nmrhkb2exnneo2nbzkrlscg https://www.researchgate.net/scientific-contributions/seong-yeob-jeong-2138609715?_sg%5b0%5d=t2aoqgh8iwt5m5tspfy13cmvznfggmthf3cj3lshkygxaypkubuakur7qcwkxzgiqsgipyk.eca5-4e4nrfgkjotnzaguvdcyxpzcnvnvcnov338thib4z20wxtjk6087abjvj95d5wxpehqurkrbbzp3r9dpw&_sg%5b1%5d=nwhtpxpewoxepjhviyudybznbugpc7ugmgbsrix-3zneuceubymvldaztmmvdrzqz_0idjs._ffwb2dyvandqc9jdx6pb_nchadb2ozztqjzlapuq_tc4-duid1qj5mtocit46_nmrhkb2exnneo2nbzkrlscg https://www.researchgate.net/scientific-contributions/jaehoon-jeung-2115956199 https://www.researchgate.net/scientific-contributions/huu-hao-ngo-38142076?_sg%5b0%5d=t2aoqgh8iwt5m5tspfy13cmvznfggmthf3cj3lshkygxaypkubuakur7qcwkxzgiqsgipyk.eca5-4e4nrfgkjotnzaguvdcyxpzcnvnvcnov338thib4z20wxtjk6087abjvj95d5wxpehqurkrbbzp3r9dpw&_sg%5b1%5d=nwhtpxpewoxepjhviyudybznbugpc7ugmgbsrix-3zneuceubymvldaztmmvdrzqz_0idjs._ffwb2dyvandqc9jdx6pb_nchadb2ozztqjzlapuq_tc4-duid1qj5mtocit46_nmrhkb2exnneo2nbzkrlscg https://www.researchgate.net/scientific-contributions/huu-hao-ngo-38142076?_sg%5b0%5d=t2aoqgh8iwt5m5tspfy13cmvznfggmthf3cj3lshkygxaypkubuakur7qcwkxzgiqsgipyk.eca5-4e4nrfgkjotnzaguvdcyxpzcnvnvcnov338thib4z20wxtjk6087abjvj95d5wxpehqurkrbbzp3r9dpw&_sg%5b1%5d=nwhtpxpewoxepjhviyudybznbugpc7ugmgbsrix-3zneuceubymvldaztmmvdrzqz_0idjs._ffwb2dyvandqc9jdx6pb_nchadb2ozztqjzlapuq_tc4-duid1qj5mtocit46_nmrhkb2exnneo2nbzkrlscg https://www.researchgate.net/journal/energy-conversion-and-management-0196-8904 https://www.researchgate.net/journal/energy-conversion-and-management-0196-8904 http://dx.doi.org/10.1016/j.enconman.2016.09.066 https://doi.org/10.1155/2021/6641796 https://doi.org/10.29333/ejosdr/7595 https://doi.org/10.1007/s42452-019-0871-z https://doi.org/10.1007/s42452-019-0871-z https://doi.org/10.14710/ijred.8.3.253-259 https://www.researchgate.net/journal/international-journal-of-water-resources-and-environmental-engineering-2141-6613 https://www.researchgate.net/journal/international-journal-of-water-resources-and-environmental-engineering-2141-6613 https://doi.org/10.5897/ijwree2020.0946 http://dx.doi.org/10.7763/ijesd.2012.v3.195 https://doi.org/10.3390/pr7090600 https://doi.org/10.3390/app9091915 https://doi.org/10.3390/en12061106 http://dx.doi.org/10.1088/1757-899x/601/1/012012 http://dx.doi.org/10.1088/1757-899x/601/1/012012 https://doi.org/10.3390/pr8060703 https://doi.org/10.3390/ijerph19116943 https://doi.org/10.3390/en14020258 https://www.researchgate.net/scientific-contributions/hongguang-zhu-2164610886?_sg%5b0%5d=9jklf-zo9ijoldzkukdukq2mq29jbwe4wegnrx77uedsewspugald0lcbjyroafzoactgpi.oe5nroh4njdp7movvx_tmxuem7cf01kex4vnb02atks77zplhr4dpxbxusdgh9xnqqlgv30numcaqvxfdxkv1a&_sg%5b1%5d=2ayb2uefdqzk7mmmwt3a9oybjifo-_uxoib6hnvydna1vymlehwmcqrgqtdcy7fpncfi_sy.6vyuv_lfg1tohnj-q3stfb8vt6znpc8ajkz7l08ri9pexpmgk5qrjsko179sjcttwrujbj3hp-jbn7sw3tnmmq https://www.researchgate.net/scientific-contributions/jing-yang-2164616274?_sg%5b0%5d=9jklf-zo9ijoldzkukdukq2mq29jbwe4wegnrx77uedsewspugald0lcbjyroafzoactgpi.oe5nroh4njdp7movvx_tmxuem7cf01kex4vnb02atks77zplhr4dpxbxusdgh9xnqqlgv30numcaqvxfdxkv1a&_sg%5b1%5d=2ayb2uefdqzk7mmmwt3a9oybjifo-_uxoib6hnvydna1vymlehwmcqrgqtdcy7fpncfi_sy.6vyuv_lfg1tohnj-q3stfb8vt6znpc8ajkz7l08ri9pexpmgk5qrjsko179sjcttwrujbj3hp-jbn7sw3tnmmq https://www.researchgate.net/scientific-contributions/cheng-xiaowei-2164603562?_sg%5b0%5d=9jklf-zo9ijoldzkukdukq2mq29jbwe4wegnrx77uedsewspugald0lcbjyroafzoactgpi.oe5nroh4njdp7movvx_tmxuem7cf01kex4vnb02atks77zplhr4dpxbxusdgh9xnqqlgv30numcaqvxfdxkv1a&_sg%5b1%5d=2ayb2uefdqzk7mmmwt3a9oybjifo-_uxoib6hnvydna1vymlehwmcqrgqtdcy7fpncfi_sy.6vyuv_lfg1tohnj-q3stfb8vt6znpc8ajkz7l08ri9pexpmgk5qrjsko179sjcttwrujbj3hp-jbn7sw3tnmmq https://www.researchgate.net/journal/e3s-web-of-conferences-2267-1242 https://www.researchgate.net/journal/e3s-web-of-conferences-2267-1242 http://creativecommons.org/licenses/by-sa/4.0/ international journal of renewable energy development int. j. renew. energy dev. 2023, 12(3), 478-487 | 478 https://doi.org/10.14710/ijred.2023.49972 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id prediction of the output power of photovoltaic module using artificial neural networks model with optimizing the neurons number abdulrahman th. mohammad1* , hasanen m. hussen2 , hussein j. akeiber3 1middle technical university (mtu), baqubah technical institute, renewable energy department, baghdad, 10074, iraq 2ministry of higher education and scientific research, department of research and development, baghdad, 10074, iraq 3ministry of interior, directorate of arab and international cooperation, department of educational affairs, baghdad, iraq abstract. artificial neural networks (anns) is an adaptive system that has the ability to predict the relationship between the input and output parameters without defining the physical and operation conditions. in this study, some queries about using ann methodology are simply clarified especially about the neurons number and their relationship with input and output parameters. in addition, two ann models are developed using matlab code to predict the power production of a polycrystalline pv module in the real weather conditions of iraq. the ann models are then used to optimize the neurons number in the hidden layers. the capability of ann models has been tested under the impact of several weather and operational parameters. in this regard, six variables are used as input parameters including ambient temperature, solar irradiance and wind speed (the weather conditions), and module temperature, short circuit current and open circuit voltage (the characteristics of pv module). according to the performance analysis of ann models, the optimal neurons number is 15 neurons in single hidden layer with minimum root mean squared error (rmse) of 2.76% and 10 neurons in double hidden layers with rmse of 1.97%. accordingly, it can be concluded that the double hidden layers introduce a higher accuracy than the single hidden layer. moreover, the ann model has proven its accuracy in predicting the current and voltage of pv module. keywords: photovoltaic, power production, artificial neural networks, neurons @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 31st oct 2022; revised: 27th january 2023; accepted: 16th feb 2023; available online: 30th march 2023 1. introduction the traditional statistical techniques are useful to estimate the behaviour of linear systems in different engineering disciplines. the presence of nonlinearity in some of applications makes the traditional statistical techniques inefficient to predict the relationship between the input and output parameters (nayak et al., 2017). among the soft computing methodologies, the artificial neural networks (anns) are widely used in recent time to predict, optimize and classify the behaviour of many problems in our life (abiodun et al., 2018). anns are relevant to machine learning (ml) models which have the ability to mimic the basic biological neural systems, especially the human brain (mubiru, 2011). anns can be classified into two main algorithms: feed-forward neural network (ffnn) and feedbackward neural network (fbnn) (hertz, 2018). the ffnn is defined as a classification algorithm where each neuron in a layer connects to other neurons in other layers with an equal weight (abiodun et al., 2018). the weight is defined as an indicator for the potential amount of the knowledge in the network. during the ffnn, the information is transmitted in one direction from input layer to output layer throughout the hidden layer. when the network is operated normally and acted as a * corresponding author email: abd20091976@gmail.com (a.th. mohammad) classifier, the fbnn process between the layers is not necessary (hagan and menhaj, 1994). the fbnn is denoted as algorithm for back-preferable propagation training which has the ability to build coordinated graph in sequence from the connections between the neurons. the fbnn can be used to minimize the loss function by adjusting or correcting the weights. in general, the simple architecture of ann consists of input, hidden and output layers (mohammad et al., 2020; mohammad et al., 2013) which composes a number of interconnected elements called neurons. each neuron receives the input signal from external process or from another neuron. the output signal from each neuron produces from a transfer function and passes into other neuron or external outputs (zhang et al., 1998). the following are the key questions that numerous researchers have addressed: a. what is the number of hidden layers and the number of neurons?; b. is there a relationship between the neurons number and the input and output parameters?; c. are there other parameters relate to determine the neurons number?. according to zhang et al. (1998), the accuracy of anns is mostly influenced by the number of hidden layers and their neurons. cybenko, (1989) demonstrated that the single hidden layer is sufficient to investigate any desired accuracy in research article https://doi.org/10.14710/ijred.2023.49972 https://doi.org/10.14710/ijred.2023.49972 https://doi.org/10.14710/ijred.2023.49972 https://www.ncbi.nlm.nih.gov/pubmed/?term=abiodun%20oi%5bauthor%5d&cauthor=true&cauthor_uid=30519653 https://www.ncbi.nlm.nih.gov/pubmed/?term=abiodun%20oi%5bauthor%5d&cauthor=true&cauthor_uid=30519653 mailto:abd20091976@gmail.com https://orcid.org/0000-0002-0487-4464 https://orcid.org/0000-0003-1119-7390 https://orcid.org/0009-0009-2207-0150 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.49972&domain=pdf a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |479 issn: 2252-4940/© 2023. the author(s). published by cbiore any complex nonlinear problems. however, this requires a large number of neurons. this is not desirable in the training time of the optimal number of neurons since it leads to poor generalization ability of anns. several colleagues proved the advantages of double hidden layers over single hidden layer in anns. according to barron (1994), double hidden layers provide greater benefits in some cases. srinivasan et al. (1994) investigated that the higher efficiency in the training stage can be improved using double hidden layers. in addition, zhang (1994) demonstrated that the high accuracy of prediction can be achieved using double hidden layers. on the other hand, some other colleagues demonstrated a relationship between the number of input parameters and neurons in hidden layers and introduced a heuristic constraint on the number of neurons (lachtermacher, 1995). kang (1991), tang and fishwick (1993), wong (1991), and lippmann (1987) and hecht-nielsen (1990), proved the existence of a relationship between the number of neurons and the input parameters which mathematically represented as (j=n/2), (j=n), (j=2n), and (j=2n+1), respectively. j is the number of neurons and n is the number of input parameters. some other colleagues proved the existence of a relationship between the neurons number and input/output parameters. for instance, kalogirou et al. (1996) proposed that the neurons are set according to the formulas: (𝑗𝑗 = 2 3 (𝑛𝑛 + 𝑜𝑜) and 𝑗𝑗 = 3 2 (𝑛𝑛 + 𝑜𝑜)). moreover, mohanraj (2009) proposed that the number of neurons is set according to the formula (𝑗𝑗 = √𝑛𝑛 + 𝑜𝑜 + 𝑎𝑎). where a is a constant from (1 to 5) and o is the output parameters. in this regard, the number of training data has an influential role for determining the number of neurons as mathematically represented as: (𝑗𝑗 = ((𝑛𝑛 + 𝑂𝑂)/2) + √(𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛 𝑜𝑜𝑜𝑜 𝑡𝑡𝑛𝑛𝑎𝑎𝑡𝑡𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 𝑑𝑑𝑎𝑎𝑡𝑡𝑎𝑎 𝑝𝑝𝑜𝑜𝑡𝑡𝑛𝑛𝑡𝑡𝑝𝑝)) (kalogirou, 1996, kalogirou et al., 1997 and 1998, luyao liu et al., 2017, shaft et al., 2006). a critical analysis of the above studies would introduce the shortcoming the of the conducted research. specifically, the number of neurons in hidden layer were randomly selected or by trial and error. furthermore, the compatibility between hidden layers and the number of neurons is not thoroughly addressed and discussed. therefore, the research intends to optimize the neurons number in single and double hidden layers which would enable to predict the output power of pv module. to appropriately approve the contribution of this research, the associated results of optimal neurons will be compared against those neurons that mathematically represented by some other colleagues. in this regard, the comparison between the performance of single and double hidden layers will be presented for the optimum number of neurons. 2. related works the applications of photovoltaic (pv) solar cells technology have become widely used to generate the electrical power over the last two decades. generally, output characteristics of pv modules can be represented in current-voltage (i-v) and powervoltage (p-v) curves (singh and ravindra, 2012). these characteristics are influenced by the ambient conditions such as, solar radiation, ambient temperature, dust and wind speed (ziane, 2021; kidegho et al., 2021). the relationship between the input parameters of ambient conditions and pv output is a complex nonlinear system. in literature, numerous studies were presented to analyse and estimate the output characteristics and performance of pv using an experimental, analytical and numerical models. among of these models, the machine learning using anns has been approved as an active model to predict the output electrical characteristics of pv modules. this section focuses on addressing the most important aspects that can be used to predict the output power of pv module using ann methodology. barhdadi et al. (2019) used an ann model with levenberg-marquardt back-propagation algorithm to predict the output power of pv module. the structure of ann has six input parameters and random neurons number from (5 to 35) with single hidden layer. the results showed that the neuron number 35 has achieved the better prediction in ann model. two topologies of ann named as feed forward and radial basis were investigated by gaur et al. (2018) to predict the performance of five pv module technologies under the influence of solar irradiance and temperature. in each topology, single and multi-hidden layers with 10 and 5 neurons were trained and tested using built-in functions and levenberg-marquardt with resilient back propagation, respectively. the results of the proposed ann models indicated that the mean bias error deviations are less than 1% if compared to the dependent models. in the same context, di-falco et al. (2014) used three types of ann topologies named as multilayer perceptron, a recursive neural network and a gamma memory trained to forecast the production power of pv module under the influence of ambient conditions. additionally, module temperature, open circuit voltage (voc) and short circuit current (isc) were used as an input parameter of the ann model. the result showed that the error ranged between 0.05 to 1% for the predicted and real power of pv module. two architectures of neural networks were used by enachescu et al. (2016) to forecast the production power of pv module. the first architecture is a multilayer perceptron with back propagation and two neurons number (15 and 100). the second architecture is named elman networks with feed forward. according to the obtained results, the elman type with small data has performed better than multilayer perceptron in learning stage. the multilayer perceptron (mlp) topology was used by jumaat et al. (2018) to predict the maximum voltage (vm) and current (im) of pv module. the structure of ann contains seven input parameters named as solar irradiance, ambient temperature, relative humidity, humidity ratio, module temperature, voc, and isc). the ann architecture built as single hidden layer with number of neurons from 1 to 10 and two output parameters. the results elucidated that the ann model is of a high accuracy to predict the vm and im of pv module. kayri and gencoglu (2019) employed the feed-forward ann topology with a back-propagation algorithm to predict the output power of mono-crystalline silicon pv module. six input parameters of weather conditions were considered including the solar irradiance, solar elevation angle, ambient temperature, wind speed, wind direction and relative humidity besides two hidden layers to build the ann structure. the comparison between the estimated and measured results showed that the maximum mean square error has not exceeded 1.4% and the coefficient of determination (r2) ranged between 99.637 to 99.998%. a simple ann structure was proposed by mellit et al. (2013) to estimate the output power of 50 wp pv in turkey. the ann model depends on input data as measured along one year including the solar irradiance, air temperature and output power in cloudy and sunny days. the ann structure considered single hidden layer with one neuron. the results elaborated that the model of sunny days is more accurate than the model of cloudy days. the determination coefficient of the cloudy days recorded between 93% and 97% while recorded between 96% and 97% in sunny days. 3. methods and ann applications a polycrystalline pv module type (frs-165w) selected and installed at the centre of middle technical university a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |480 issn: 2252-4940/© 2023. the author(s). published by cbiore baghdad-iraq to achieve the experimental tests and to collect the dataset. the pv module has the following technical specifications: (pmax=165 wp, isc=9.81 a, voc=22.05 v, imp=9.17 a and vmp=18 v). the i-v tracer (seaward pv200) used to measure the output electrical characteristics of pv module including: voc, isc, vmp and imp. in addition, solar meter (survey 200r) unit used and synchronized with the i-v tracer to measure the ambient temperature, solar radiation and back temperature of the pv module. also, a handle anemometer used to measure the wind speed (va). totally, 326 measured data obtained through the period of tests. these data were stored and reported by solar data logger and displayed by solarcert software. ann has been used effectively in photovoltaic for solving various problems, for example, the effects of atmospheric variables on the production power in pv modules. it is defined as a mapping system that can be used to represent a nonlinear relationship between the input and output parameters (mellit et al., 2013). in general, the simple structure of ann consists of three layers named as: input layer, hidden layer and output layer (figure 1) (pontes et al., 2012). the relationship between the input and output layers can be represented mathematically as: 𝑂𝑂𝑗𝑗 = 𝑜𝑜 ∑ 𝑊𝑊𝑗𝑗,𝑖𝑖 𝑋𝑋𝑖𝑖 + 𝑛𝑛𝑖𝑖 𝑚𝑚 𝑖𝑖=1 (1) where: w, x and b are the weight, input parameter and bias, respectively and f is the activation function. the output neurons (k) in the output layer can be expressed as: 𝑦𝑦𝑘𝑘 = 𝑡𝑡∑ 𝑊𝑊𝑘𝑘,𝑗𝑗 𝑂𝑂𝑗𝑗 𝑛𝑛 𝑗𝑗=1 (2) the relationship between the input parameters and k is represented in eq. 3 (mackay, 1992) 𝑦𝑦𝑘𝑘 = 𝑡𝑡�∑ 𝑊𝑊𝑘𝑘,𝑗𝑗 𝑛𝑛 𝑗𝑗=1 𝑜𝑜�∑ 𝑊𝑊𝑗𝑗,𝑖𝑖 𝑋𝑋𝑖𝑖 + 𝑛𝑛𝑖𝑖 𝑚𝑚 𝑖𝑖=1 �� (3) the training process tries to adjust the connection weight for keeping the predicted output (yk) closed as expectation to the desired ouput (yk���) under the given input parameters (xi). the error fuction between the predicted and desired output can be formuated as a minimum value (elsheikh et al., 2019; raj et al., 2019). 𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡𝑚𝑚𝑛𝑛 1 2 ∑ (yk��� − 𝑦𝑦𝐾𝐾)2𝑙𝑙𝑘𝑘=1 (4) the transfer function is used to obtain and send the signals between the layers. there are three popular types of transfer functions called: linear, sigmoid, and hyperbolic tangent transfer functions (mellit et al., 2013). the mathematical model of these types of transfer functions can be represented as: 𝑜𝑜(𝑆𝑆) = ⎩ ⎨ ⎧ 𝑆𝑆 𝑙𝑙𝑡𝑡𝑛𝑛𝑛𝑛𝑎𝑎𝑛𝑛 𝑜𝑜𝑛𝑛𝑛𝑛𝑓𝑓𝑡𝑡𝑡𝑡𝑜𝑜𝑛𝑛 1 1+𝑒𝑒−𝑆𝑆 𝑝𝑝𝑡𝑡𝑡𝑡𝑛𝑛𝑜𝑜𝑡𝑡𝑑𝑑 𝑜𝑜𝑛𝑛𝑛𝑛𝑓𝑓𝑡𝑡𝑡𝑡𝑜𝑜𝑛𝑛 𝑒𝑒+𝑆𝑆 −𝑒𝑒−𝑆𝑆 𝑒𝑒+𝑆𝑆 +𝑒𝑒−𝑆𝑆 ℎ𝑦𝑦𝑝𝑝𝑛𝑛𝑛𝑛𝑛𝑛𝑜𝑜𝑙𝑙𝑡𝑡𝑓𝑓 𝑡𝑡𝑎𝑎𝑛𝑛𝑡𝑡𝑛𝑛𝑛𝑛𝑡𝑡 𝑜𝑜𝑛𝑛𝑛𝑛𝑓𝑓𝑡𝑡𝑡𝑡𝑜𝑜𝑛𝑛 (5) the sigmoid transfer function is considered as a better method among the transfer functions (mellit et al., 2013). therefore, the current study utilised this as an activation function. the error can be expressed as: 𝐸𝐸 = 1 2 �yk��� − 1 1+𝑒𝑒−𝑆𝑆 � 2 (6) 𝐸𝐸 = 1 2 �yk��� − 1 𝑒𝑒−∑ 𝑊𝑊𝑗𝑗,𝑖𝑖𝑋𝑋𝑖𝑖+𝑏𝑏𝑖𝑖 𝑚𝑚 𝑖𝑖=1 � 2 (7) the chain rule theory is also applied to calculate ( 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 ) as follows: 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 = 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝑥𝑥 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 𝑥𝑥 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 (8) according to the derivation of each component, the final change error of the weight can be represented as: 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 = (yk��� − 𝑦𝑦𝑘𝑘) 1 1+𝑒𝑒−𝑆𝑆 �1 − 1 1+𝑒𝑒−𝑆𝑆 �𝑋𝑋 (9) therefore, if the predicted output 𝑦𝑦𝑘𝑘 is not close to the desired output yk���, the weights must be adapted (rumelhart et al., 1986). 𝑊𝑊𝑛𝑛𝑒𝑒𝑛𝑛 = 𝑊𝑊𝑜𝑜𝑙𝑙𝑜𝑜 − ƞ 𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕 (10) where: ƞ is the learning rate (0 ≤ ƞ ≤ 1). the quantitative variable of neurons is normalized to some standard ranges such as [0 1] or [-1 1] before beginning the training and testing processes (mohammad et al., 2013). the normalization process can be investigated according to (sanjay et al., 2006) as: 𝑋𝑋𝑖𝑖 = 0.8 𝑜𝑜𝑚𝑚𝑚𝑚𝑚𝑚−𝑜𝑜𝑚𝑚𝑖𝑖𝑚𝑚 (𝑑𝑑𝑖𝑖 − 𝑑𝑑𝑚𝑚𝑖𝑖𝑛𝑛) + 0.1 (11) where: dmax, dmin and di are the maximum, minimum and number (ith) of the desired input/ output data, respectively. 3.1. neurons number there is no mathematical formula can determine the number of neurons in the hidden and output layers. therefore, most of the fig. 1 basic design of ann structure. https://www.sciencedirect.com/science/article/pii/s0960148113002279#! https://www.sciencedirect.com/science/article/pii/s0960148113002279#! a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |481 issn: 2252-4940/© 2023. the author(s). published by cbiore colleagues were relied on the trial and error approach to select the neurons number. besides, some of them suggested a relationship between the neurons number and input parameters. for example, kang (1991), tang and fishwick (1993), wong (1991), lippmann (1987) and hecht-nielsen (1990) suggested that the neurons number can be specified by the input parameters: 𝑗𝑗 = � 𝑛𝑛 2� 𝑛𝑛 2𝑛𝑛 2𝑛𝑛 + 1 (12) where: j and n are the neurons number and input parameters, respectively. some other colleagues proved that neurons number can be specified by input and output parameters. mohanraj (2009) proved the following relationship 𝑗𝑗 = √𝑛𝑛 + 𝑜𝑜 + 𝑎𝑎 (13) kalogirou et al. (1996) also introduced eq. 14 to identify the neurons number 𝑗𝑗 = � 2 3 (𝑛𝑛 + 𝑜𝑜) 3 2 (𝑛𝑛 + 𝑜𝑜) (14) where: o is the number output parameters and (a) is a constant from 1 to 5. kalogirou (1996), kalogirou et al. (1997), kalogirou et al. (1998), liu et al. (2017), and shaft et al. (2006) stated that the number of training data plays a key role in determining the neurons number besides considering the input and output parameters. this is clearly represented in eq. 15 𝑗𝑗 = �𝑛𝑛+𝑂𝑂 2 � + �𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛 𝑜𝑜𝑜𝑜 𝑡𝑡𝑛𝑛𝑎𝑎𝑡𝑡𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 𝑑𝑑𝑎𝑎𝑡𝑡𝑎𝑎 𝑝𝑝𝑜𝑜𝑡𝑡𝑛𝑛𝑡𝑡𝑝𝑝 (15) 3.2. structure of proposed ann in the current study, two architectures of ann were proposed. the first one represents ann structure with single hidden layer. the second one represents with double hidden layers. a 326 dataset were used to run the ann models. the block diagram of concept the overlap between the ann model and experimental setup is depicted in figure 2. the minimum and maximum values of the measured parameters are summarized fig. 2 concept of the overlap between the anns model and experimental setup. table 1 minimum and maximum values of measured data. parameters symbol unit min max input parameters solar irradiance g w/m2 169.2 1003 ambient temperature ta oc 32 51 wind speed va m/sec 0.3 2.5 module temperature tc oc 32.6 69.1 open circuit voltage voc v 19.6 20.6 short circuit current isc a 4.116 9.1 output parameters output power p w 0 140 a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |482 issn: 2252-4940/© 2023. the author(s). published by cbiore in table 1. the characteristics of the ann model can be drawn as: • six nodes used in input layer. these include the solar irradiance, ambient temperature, wind speed, module temperature, open circuit voltage and short circuit current) and one node in output layer (output power). • the model used levenberg-marquardt backpropagation technique in training stage. • 70% from dataset was used for training, while 15% for testing and 15% for validation • single and double hidden layers were evaluated in the ann model. • the sigmoid and purlin functions were used as an activation functions in the hidden and output layers. • the neurons number were tested from 1 to 100. • the optimization of neurons number has been evaluated using rmse in validation and training stage. • mean squared error (mse) and coefficient of determination are used to measure the effectiveness of the ann model (r2). 4. results and discussions in this study, two ann architectures with single and double hidden layers were used to predict the production power of pv module in the real weather conditions of iraq. the capability of ann models has been evaluated under the impact of several weather and operational parameters. totally, six variables were used as the input parameters. three of these variables related to the weather conditions (ambient temperature, solar irradiance and wind speed) while, the other variables related to the characteristics of pv module (module temperature, short circuit current and open circuit voltage). 4.1. visualization of weather and operational parameters a scatter plot visualization technique was used to represent and express the data graphically as shown in figure 3. the main goal of visualization process is to acquire insight into the data. the measured output power of pv is plotted and ranged against the solar irradiance, ambient temperature, module temperature, wind speed, open circuit voltage and short circuit current. it is clear that that the power increases linearly with irradiance and open circuit current. on the other hand, the linearity decreases with wind speed. however, the relationship becomes random with ambient temperature, module temperature and open circuit voltage. 4.2. optimization of neurons number two matlab codes were used to optimize the neurons number in single and double hidden layers. the optimization ranged between 1 to 100 and evaluated depending on the minimum value of root mean squared error (rmse) in validation stage. also, the evaluation depends on the minimum difference of rmse between training and validation stages. figures 4 shows the best neuron number of 15 in single layer with rmse value of 2.76 % and 10 in double hidden layer with rmse value of 1.97%. furthermore, the difference between the fig. 3 visualization of measure data with power production of pv module. a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |483 issn: 2252-4940/© 2023. the author(s). published by cbiore rmse in validation and training stages has recorded 0.18% and 0.21% in single and double hidden layers, respectively. generally, the optimal neuron in each single and double hidden layer has showed a better rmse than the neurons that proposed by many colleagues (kang, 1991; liu et al., 2017; shaft et al., 2006; h-nielsen, 1990; mohanraj et al., 2009; kalogirou et al., 1996-1997-1998; kalogirou, 1996; wong, 1991; lippmann, 1987). the comparison of rmse in single hidden layer would also show that the neuron number 5 (according to the formulas of mohanraj et al., 2009 and kalogirou et al., 1996) has represented a lower rmse of 2.4% in validation stage but with difference about 2.31% from training stage. in addition, the neuron number 7 (according to the formula of mohanraj et al., 2009) has elaborated the same value of rmse in the current study but with difference about 0.36% from training stage. however, the neuron number 5 in double hidden layers (according to mohanraj et al., 2009 and kalogirou et al., 1996) has recorded the same value of rmse of the current study but with difference about 2.94%. in more details, table 2 discusses the calculation of neurons number according to the mathematical formulas imposed by the colleagues in the open literature. their values of rmse in validation and training stages (a) single hidden layer (b) double hidden layer fig. 4 optimal neurons number (a) single hidden layer (b) double hidden layers. a) single hidden layer b) double hidden layer fig. 5 best validation performance (a) single hidden layer (b) double hidden layers. a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |484 issn: 2252-4940/© 2023. the author(s). published by cbiore were calculated based on the results of the present ann model when run from (1 to100) neurons number. table 2 has been adopted to clarify the comparison between the present ann model and previous ann models depending on the values of rmse in validation stage and the minimum difference between rmse for validation and training stages. 4.3. performance of ann model figure 5 shows the best performance of ann model for single and double hidden layers. it is clear that the best validation performance in single layer has been investigated at epoch 6 with mse of 1.966×10-3. however, it has been investigated at epoch 4 with mse of 1.950×10-3 in double hidden layer. furthermore, the mse of training stage in single and double hidden layers has recorded lower value than mse of testing and validation stages. this is a great indication that the data of the model were learned very well in training stage. figure 6 defines the regressions curves of all the ann stages including: training, testing, validation and all for single and double hidden layers. the coefficient of determination (r2) is used as an indicator to evaluate the accuracy between the predicted and target of (a) single hidden layer (b) single hidden layer fig. 6 regression curves of all stages (a) single hidden layer (b) double hidden layers fig. 7 measured and predicted values of output power of pv. a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |485 issn: 2252-4940/© 2023. the author(s). published by cbiore production power. in single layer, the values of r2 ranged as 0.9881, 0.981, 0.982 and 0.985 for training, testing, validation, and all, respectively. for double hidden layer, the values of r2 ranged as 0.9887, 0.985, 0.99 and 0.988, respectively. according to r2, it can be observed that the ann model of double hidden layers has the best accuracy if compared to the model of single layer in all the stages of ann model. figure 7 shows the pattern of predicted power of the ann model in single and double table 2 present neurons number compared with literature studies. ann structure reference formula rmse-valid(%) difference rmse% (validtrain) single hidden layer 6-3-1 (kang, 1991) j=n/2 6.4 2.30 6-6-1 (tang et al., 1993) j=n 7.04 1.84 6-12-1 (wong, 1991) and (lippmann,1987) j=2n 2.08 2.18 6-13-1 (h-nielsen, 1990) j=2n+1 8.03 5.90 6-4-1 6-5-1 6-6-1 6-7-1 6-8-1 (liu et al., 2017) 𝑗𝑗 = √𝑛𝑛 + 𝑜𝑜 + 𝑎𝑎 a=1,2,3,4,5 3.2 2.4 7.04 2.76 4.06 0.58 2.31 1.84 0.36 0.68 6-5-1 (shaft et al., 2006) 𝑗𝑗 = 2 3 (𝑛𝑛 + 𝑜𝑜) 2.4 2.31 6-11-1 (shaft et al., 2006) 𝑗𝑗 = 3 2 (𝑛𝑛 + 𝑜𝑜) 8.53 4.88 6-19-1 (mohanraj. et al., 2009), (kalogirou et al., 19961997-1998), (kalogirou., 1996) 𝑗𝑗 = � 𝑛𝑛 + 𝑂𝑂 2 � + �𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛 𝑜𝑜𝑜𝑜 𝑡𝑡𝑛𝑛𝑎𝑎𝑡𝑡𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 𝑑𝑑𝑎𝑎𝑡𝑡𝑎𝑎 𝑝𝑝𝑜𝑜𝑡𝑡𝑛𝑛𝑡𝑡𝑝𝑝 6.23 2.23 6-15-1 present study optimization 2.76 0.18 double hidden layer 6-3-3-1 (kang, 1991) j=n/2 4.03 1.31 6-6-6-1 (tang et al., 1993) n 2.96 2.06 6-12-12-1 (wong, 1991) and (lippmann r.p., 1987) j=2n 6.69 4.96 6-13-13-1 (h-nielsen, 1990) j=2n+1 8.12 4.97 6-4-4-1 6-5-5-1 6-6-6-1 6-7-7-1 6-8-8-1 (liu et al., 2017) 𝑗𝑗 = √𝑛𝑛 + 𝑜𝑜 + 𝑎𝑎 a=1,2,3,4,5 2.67 1.97 2.96 6.22 3.40 1.31 2.94 2.06 3.68 0.13 6-5-5-1 (shaft et al., 2006) 𝑗𝑗 = 2 3 (𝑛𝑛 + 𝑜𝑜) 1.97 2.94 6-11-11-1 (shaft et al., 2006) 𝑗𝑗 = 3 2 (𝑛𝑛 + 𝑜𝑜) 7.93 4.85 6-19-19-1 (mohanraj et al., 2009), (kalogirou et al., 19961997-1998), (kalogirou, 1996) 𝑗𝑗 = � 𝑛𝑛 + 𝑂𝑂 2 � + �𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛 𝑜𝑜𝑜𝑜 𝑡𝑡𝑛𝑛𝑎𝑎𝑡𝑡𝑛𝑛𝑡𝑡𝑛𝑛𝑡𝑡 𝑑𝑑𝑎𝑎𝑡𝑡𝑎𝑎 𝑝𝑝𝑜𝑜𝑡𝑡𝑛𝑛𝑡𝑡𝑝𝑝 5.45 2.02 6-10-10-1 present study optimization 1.97 0.21 (a).measured and predicted (i-v) curves (b)measured and predicted (p-v) curve fig. 8 validation of ann model with measured data (a) i-v curve (b) p-v curve. a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |486 issn: 2252-4940/© 2023. the author(s). published by cbiore hidden layers for training and validation stages. the pattern is similar with measured value of power especially in double hidden layer. however, there is a little deviation in the single hidden layer at training stage. the accuracy of convergence the patterns increases in the validation stage for double hidden layer more than the single hidden layer. . 4.4. case study of (i-v) and (p-v) curves to confirm more reliability of the present ann model, the model was tested to predict the maximum power point current (imp) and voltage (vmp). the output results of current and voltage prediction are represented as: (i-v) and (p-v) curves to compare them against the measured data as shown in figure 8. a good matching can be seen between the predicted and measured (iv) and (p-v) curves. in addition, the ann model with double hidden layer has achieved more convergence with the measured data rather than ann model with single hidden layer. this indicates that the ann model has proven its accuracy in predicting the current and voltage of pv module. 5. conclusion two ann architectures with single and double hidden layer have been used to predict the power production of a polycrystalline pv module in real condition of iraq. the main contribution of ann model was the optimization of the neurons number in the hidden layers. the analysis of results showed that the best prediction of the ann model with single hidden layer was investigated at optimal neuron number 15 with minimum root mean squared error (rmse) of 2.76% and 10 neurons in double hidden layers with rmse of 1.97%. thus, it can be concluded that the present ann model has proven its rapid response in reaching the predicted accuracy more than other models included in the literature. in addition, the ann model with double hidden layers achieved a higher accuracy than the single hidden layer. in a summary, the ann model proved its accuracy to forecast the current and voltage of pv module. references abiodun, o.i., aman j., abiodun e. o., kemi, v. d., nachaat, a. m., humaira, a. (2018). state-of-the-art in artificial neural network applications: a survey. heliyon, 4(11), 1-41. https://doi.org/10.1016/j.heliyon.2018.e00938 barron, a.r. (1994). neural networks: a review from a statistical perspective. statistical science, 9 (1), 33-35. https://doi.org/10.1214/ss/1177010638 cristian-dragos, d., adrian, g. & calin, e. (2016). solar photovoltaic energy production forecast using neural network. procedia technology, 22, 808-815. https://doi.org/10.1016/j.protcy.2016.01.053 cybenko, g. (1989). approximation by super positions of a sigmoidal function. mathematical control signals systems, 2, 303-314. https://link.springer.com/article/10.1007/bf02551274 elsheikh, a.h., sharshir sw, elaziz, m.a., kabeel, a.e., guilan, w. & haiou, z. (2019). modelling of solar energy systems using artificial neural network: a comprehensive review. solar energy, 180 (1), 622-639. https://doi.org/10.1016/j.solener.2019.01.037 gideon, k., francis, n., christopher, m. & robert k. (2021). evaluation of thermal interface materials in mediating pv cell temperature mismatch in pv-teg power generation. energy reports, 7, 16361650. https://doi.org/10.1016/j.egyr.2021.03.015 hagan, m.t. & menhaj, m.b. (1995). training feed forward networks with the marquardt algorithm. ieee trans. neural network, 5(6), 989-993. https://doi.org/10.1109/72.329697 hecht-nielsen, r. (1990) neurocomputing. addison-wesley, longman publishing co., inc. 75 arlington street, suite 300 boston, ma united states. https://dl.acm.org/doi/10.5555/103996 hertz, j., krogh, a. & palmer, r.g. (2018). introduction to the theory of neural computation. crc press-taylor & francis group, 6000 broken sound parkway nw, boca raton, london new york. https://doi.org/10.1201/9780429499661 ismail, k. & gencoglu m.t. (2019). predicting power production from a photovoltaic panel through artificial neural networks using atmospheric indicators. neural computing and applications, 31 (8), 3573-3586. https://doi.org/10.1007/s00521-017-3271-6 kalogirou, s.a, neocleous, c.c. & schizas, c. n. (1998). artificial neural networks for modelling the starting-up of a solar steam generator. applied energy, 60 (2), 89-100. https://doi.org/10.1016/s03062619(98)00019-1 kalogirou, s.a. (1996). artificial neural networks for predicting the local concentration ratio of parabolic trough collectors. freiburg, germany: proceedings of the international conference eurosun’96, 470-475. http://ktisis.cut.ac.cy/handle/10488/822 kalogirou, s.a., neocleous, c.e, & schizas, c.n. (1996). a comparative study of methods for estimating intercept factor of parabolic trough collectors. london, uk: proceedings of the international conference eann’96,5-8. http://users.abo.fi/abulsari/eann96.html kalogirou, s.a., neocleous, c.e., schizas, c.n. (1997). artificial neural networks for the estimation of the performance of a parabolic trough collector steam generation system. stockholm, sweden: proceedings of the international conference eann’97, 227-232. https://ktisis.cut.ac.cy/handle/10488/18149 kang, s. (1991). an investigation of the use of feed forward neural networks for forecasting. ph.d. thesis, kent state university. https://dl.acm.org/doi/10.5555/144978 laarabi, b., may, t. o., dahlioui, d., bassam, a., flota-banuelos, m. & barhdadi, a. (2019). artificial neural network modeling and sensitivity analysis for soiling effects on photovoltaic panels in morocco. superlattices and microstructures, 127, 139-150. https://doi.org/10.1016/j.spmi.2017.12.037 lachtermacher, g., fuller, j.d. (1995). back propagation in time-series forecasting. journal of forecasting, 14 (4), 381393. https://doi.org/10.1002/for.3980140405 lippmann, r.p. (1987). an introduction to computing with neural nets. ieee magazine, 4(2), 4-22. https://doi.org/10.1109/massp.1987.1165576 liu, l., diran l., qie s., hailong, l. & ronald, w. (2017). forecasting power output of photovoltaic system using a bp network method. energy procedia, 142, 780-786. https://doi.org/10.1016/j.egypro.2017.12.126 mackay, d.j.c. (1992). a practical bayesian framework for back propagation networks. neural computation, 4 (3), 448-72. https://doi.org/10.1162/neco.1992.4.3.448 mellit, a., saglam, s. & kalogirou, s.a. (2013). artificial neural networkbased model for estimating the produced power of a photovoltaic module. renewable energy, 60, 71-78. https://doi.org/10.1016/j.renene.2013.04.011 mittal, m. birinchi b., sahaj, s. & anshu, m.g. (2018). performance prediction of pv module using electrical equivalent model and artificial neural network. solar energy, 176, 104-117. https://doi.org/10.1016/j.solener.2018.10.018 mohammad, a.t., al-obaidi, m.a., hameed, e.m. basheer, i.m. & mujtaba i. m. (2020). modelling the chlorophenol removal from wastewater via reverse osmosis process using a multilayer artificial neural network with genetic algorithm. journal of water process engineering, 33, 1-10. https://doi.org/10.1016/j.jwpe.2019.100993 mohammad, a.t., sohif, b. m., sulaiman, m.y., sopian, k., & al-abidi a.a. (2013). artificial neural network analysis of liquid desiccant dehumidifier performance in a solar hybrid air-conditioning system. applied thermal engineering, 59 (1-2), 389397.https://doi.org/10.1016/j.applthermaleng.2013.06.006 mohanraj, m., jayaraj, s. & muraleedharan, c. (2009). performance prediction of a direct expansion solar assisted heat pump using artificial neural networks. appl energy. 86(9):1442-1449. https://doi.org/10.1016/j.apenergy.2009.01.001 mubiru, j. (2011). using artificial neural networks to predict direct solar irradiation. advances in artificial neural systems, 2011, 16877594. https://doi.org/10.1155/2011/142054 nayak, s. c., misra, b. b. & behera, h.s. (2017). artificial chemical reaction optimization of neural networks for efficient prediction https://www.ncbi.nlm.nih.gov/pubmed/?term=abiodun%20oi%5bauthor%5d&cauthor=true&cauthor_uid=30519653 https://www.ncbi.nlm.nih.gov/pubmed/?term=jantan%20a%5bauthor%5d&cauthor=true&cauthor_uid=30519653 https://www.ncbi.nlm.nih.gov/pubmed/?term=omolara%20ae%5bauthor%5d&cauthor=true&cauthor_uid=30519653 https://www.ncbi.nlm.nih.gov/pubmed/?term=dada%20kv%5bauthor%5d&cauthor=true&cauthor_uid=30519653 https://www.ncbi.nlm.nih.gov/pubmed/?term=mohamed%20na%5bauthor%5d&cauthor=true&cauthor_uid=30519653 https://www.ncbi.nlm.nih.gov/pubmed/?term=mohamed%20na%5bauthor%5d&cauthor=true&cauthor_uid=30519653 https://www.ncbi.nlm.nih.gov/pubmed/?term=arshad%20h%5bauthor%5d&cauthor=true&cauthor_uid=30519653 https://www.ncbi.nlm.nih.gov/pmc/articles/pmc6260436/ https://doi.org/10.1016/j.heliyon.2018.e00938 https://doi.org/10.1016/j.protcy.2016.01.053 https://link.springer.com/article/10.1007/bf02551274 https://doi.org/10.1016/j.solener.2019.01.037 https://doi.org/10.1016/j.egyr.2021.03.015 https://doi.org/10.1109/72.329697 https://dl.acm.org/doi/10.5555/103996 https://doi.org/10.1201/9780429499661 https://www.researchgate.net/journal/neural-computing-and-applications-1433-3058?_sg=pkxy8ksdm63fwuf6ovzlfxy_ecpwy2g8309m2ohjqneja-w3iv5sbveceovrgffnjroybjq2gxeicgxi8bqgzjy0dvkfb9m.l0o3sb5etufkk_suxj4ob6kfk4yxrbtyitgrmuhhmwih8aypymy-qydghs1nitlskb9ws63p-dc7mle3vu5jsw https://doi.org/10.1016/s0306-2619(98)00019-1 https://doi.org/10.1016/s0306-2619(98)00019-1 http://ktisis.cut.ac.cy/handle/10488/822 http://users.abo.fi/abulsari/eann96.html https://ktisis.cut.ac.cy/handle/10488/18149 https://doi.org/10.1016/j.spmi.2017.12.037 https://doi.org/10.1002/for.3980140405 https://doi.org/10.1109/massp.1987.1165576 https://www.sciencedirect.com/science/article/pii/s187661021735854x#! https://www.sciencedirect.com/science/article/pii/s187661021735854x#! https://www.sciencedirect.com/author/52364999900/qie-sun https://www.sciencedirect.com/science/article/pii/s187661021735854x#! https://www.sciencedirect.com/science/article/pii/s187661021735854x#! https://www.sciencedirect.com/journal/energy-procedia https://www.sciencedirect.com/journal/energy-procedia/vol/142/suppl/c https://doi.org/10.1016/j.egypro.2017.12.126 https://doi.org/10.1162/neco.1992.4.3.448 https://doi.org/10.1016/j.renene.2013.04.011 https://doi.org/10.1016/j.solener.2018.10.018 https://www.sciencedirect.com/science/article/pii/s2214714419310244 https://www.sciencedirect.com/science/article/pii/s2214714419310244 https://www.sciencedirect.com/science/article/pii/s2214714419310244 https://www.sciencedirect.com/journal/journal-of-water-process-engineering https://www.sciencedirect.com/journal/journal-of-water-process-engineering https://www.sciencedirect.com/journal/journal-of-water-process-engineering/vol/33/suppl/c https://doi.org/10.1016/j.jwpe.2019.100993 https://www.sciencedirect.com/journal/applied-thermal-engineering https://www.sciencedirect.com/journal/applied-thermal-engineering/vol/59/issue/1 https://doi.org/10.1016/j.applthermaleng.2013.06.006 https://doi.org/10.1016/j.apenergy.2009.01.001 https://www.researchgate.net/journal/advances-in-artificial-neural-systems-1687-7608 https://doi.org/10.1155/2011/142054 https://www.sciencedirect.com/science/article/pii/s2090447915001410#! https://www.sciencedirect.com/science/article/pii/s2090447915001410#! https://www.sciencedirect.com/science/article/pii/s2090447915001410#! a. th. mohammad et al int. j. renew. energy dev 2023, 12(3), 478-487 |487 issn: 2252-4940/© 2023. the author(s). published by cbiore of stock market indices. ain shams engineering journal, 8(3), 371390. https://doi.org/10.1016/j.asej.2015.07.015 pontes, f.j., de paiva, a.p., balestrassi, p.p., ferreira, j.r. and da silva, m.b. (2012). optimization of radial basis function neural network employed for prediction of surface roughness in hard turning process using taguchi’s orthogonal arrays. expert systems with applications, 39 (9), 7776-7787. https://doi.org/10.1016/j.eswa.2012.01.058 priyanka, s. & ravindra n.m. (2012). temperature dependence of solar cell performance-an analysis. solar energy materials and solar cells, 101, 36-45. https://doi.org/10.1016/j.solmat.2012.02.019 raj, a.k., kunal, g., srinivas, m. & jayaraj, s. (2019). performance analysis of a double-pass solar air heater system with asymmetric channel flow passages. journal of thermal analysis and calorimetry, 136 (1), 21-38. http://dx.doi.org/10.1007/s10973018-7762-1 rumelhart, d.e., hinton, g.e. & williams, r.j. (1986). learning representations by backpropagating errors. nature, 323, 533-536; https://www.nature.com/articles/323533a0 sanjay, c., jyothi, c. & chin, c.w. (2006). a study of surface roughness in drilling using mathematical analysis and neural networks. international journal advance manufacturing technology, 29, 846852. https://link.springer.com/article/10.1007/s00170-0060717-x shaft, i. a., shah s.i. & kashif, f.m. (2006). impact of varying neurons and hidden layers in neural network architecture for a time frequency application. in proceedings of the 10th ieee international multitopic conference (inmic), islamabad, pakistan, (23-24) 188193. https://doi.org/10.1109/inmic.2006.358160 siti, a. j., flora, c., mohd, h. a.w, nur hanis, m. r. & muhammad, f. o. (2018). prediction of photovoltaic (pv) output using artificial neutral network (ann) based on ambient factors. journal of physics: conference series 1049, (2018) 012088. https://doi.og/10.1088/1742-6596/1049/1/012088 srinivasan, d., liew, a.c. & chang, c.s. (1994). a neural network shortterm load forecaster. electric power systems research, 28 (3), 227234. https://doi.org/10.1016/0378-7796(94)90037-x tang, z., fishwick, p.a. (1993). feed forward neural nets as models for time series forecasting. journal on computing, 5 (4), 374-385. https://doi.org/10.1287/ijoc.5.4.374 valerio, l., brano, g. c. & mariavittoria d.f. (2014). artificial neural networks to predict the power output of a pv panel. international journal of photoenergy, 2014, article id 193083, 1-12. https://doi.org/10.1155/2014/193083 wong, f.s. (1991). time series forecasting using back propagation neural networks. neurocomputing, 2 (4), 147-159. https://doi.org/10.1016/0925-2312(91)90045-d yegnanarayana, b. (2009). artificial neural networks. new york. usa: phi learning pvt. ltd. zhang, g., eddy, p. b. & michael, y. h. (1998). forecasting with artificial neural networks: the state of the art. international journal of forecasting, 14 (1), 35-62. https://doi.org/10.1016/s01692070(97)00044-7 zhang, x., (1994). time series analysis and prediction by neural networks. optimization methods and software, 4(2), 151-170. https://doi.org/10.1080/10556789408805584 ziane, a., necaibia, a., sahouane, n., dabou, r., mostefaoui, m., bouraiou, a., khelifi, s., rouabhia, a. & blal, m. (2021). photovoltaic output power performance assessment and forecasting: impact of meteorological variables. solar energy, 220, 745-75. https://doi.org/10.1016/j.solener.2021.04.004 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://www.sciencedirect.com/journal/ain-shams-engineering-journal https://www.sciencedirect.com/journal/ain-shams-engineering-journal/vol/8/issue/3 https://doi.org/10.1016/j.asej.2015.07.015 https://doi.org/10.1016/j.eswa.2012.01.058 https://www.sciencedirect.com/journal/solar-energy-materials-and-solar-cells https://www.sciencedirect.com/journal/solar-energy-materials-and-solar-cells https://www.sciencedirect.com/journal/solar-energy-materials-and-solar-cells/vol/101/suppl/c https://doi.org/10.1016/j.solmat.2012.02.019 http://dx.doi.org/10.1007/s10973-018-7762-1 http://dx.doi.org/10.1007/s10973-018-7762-1 https://doi.org/10.1109/inmic.2006.358160 https://doi.og/10.1088/1742-6596/1049/1/012088 https://doi.org/10.1016/0378-7796(94)90037-x https://doi.org/10.1287/ijoc.5.4.374 https://doi.org/10.1155/2014/193083 https://doi.org/10.1016/0925-2312(91)90045-d https://doi.org/10.1016/s0169-2070(97)00044-7 https://doi.org/10.1016/s0169-2070(97)00044-7 https://doi.org/10.1080/10556789408805584 prediction of the output power of photovoltaic module using artificial neural networks model with optimizing the neurons number abdulrahman th. mohammad10f , hasanen m. hussen2, hussein j. akeiber3 1. introduction 4. results and discussions liu, l., diran l., qie s., hailong, l. & ronald, w. (2017). forecasting power output of photovoltaic system using a bp network method. energy procedia, 142, 780-786. https://doi.org/10.1016/j.egypro.2017.12.126 mittal, m. birinchi b., sahaj, s. & anshu, m.g. (2018). performance prediction of pv module using electrical equivalent model and artificial neural network. solar energy, 176, 104-117. https://doi.org/10.1016/j.solener.2018.10.018 ziane, a., necaibia, a., sahouane, n., dabou, r., mostefaoui, m., bouraiou, a., khelifi, s., rouabhia, a. & blal, m. (2021). photovoltaic output power performance assessment and forecasting: impact of meteorological variables. solar energy, 220, 745-7... international journal of renewable energy development int. journal of renewable energy development 3 (2) 2014: 145-154 p a g e | 145 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: http://ejournal.undip.ac.id/index.php/ijred influence of the determination methods of k and c parameters on the ability of weibull distribution to suitably estimate wind potential and electric energy ruben m. mouanguea * myrin y. kazetb, alexis kuitcheb and jean-marie ndjakac adepartment of energetic engineering, uit, un, po box 455 ngaoundere, cameroon bdepartment of geea, pai, ensai, university of ngaoundere, cameroon cdepartment of physics, faculty of sciences, university of yaounde i, cameroon abstract: the modeling of the wind speed distribution is of great importance for the assessment of wind energy potential and the performance of wind energy conversion system. in this paper, the choice of two determination methods of weibull parameter s shows theirs influences on the weibull distribution performances. because of important calm winds on the site of ngaoundere airport, we characterize the wind potential using the approach of weibull distribution with parameters which are determined by the modified maximum likelihood method. this approach is compared to the weibull distribution with parameters which are determined by the maximum likelihood method and the hybrid distribution which is recommended for wind potential assessment of sites having nonzero probability of calm. using data provided by the asecna weather service (agency for the safety of air navigation in africa and madagascar), we evaluate the goodness of fit of the various fitted distributions to the wind speed data using the q – q plots, the pearson’s coefficient of correlation, the mean wind speed, the mean square error, the energy density and its relative error. it appears from the results that the accuracy of the weibull distribution with parameters which are deter mined by the modified maximum likelihood method is higher than others. then, this approach is used to estimate the monthly and annual energy productions of the site of the ngaoundere airport. the most energy contribution is made in march with 255.7 mwh. it also appears from the results that a wind turbine generator installed on this particular site could not work for at least a half of the time because of higher frequency of calm. for this kind of sites, the modified maximum likelihood method proposed by seguro and lambert in 2000 is one of the best methods which can be used to determinate the weibull parameters. keywords: frequency of calm, weibull parameters, wind energy, wind modeling, wind potential article history: received december 12, 2013; received in revised form may 20, 2014; accepted june 12, 2014; available online how to cite this article: mouangue, r.m., kazet, m.y., kuitche, a. & ndjaka, j.m. (2014) influence of the determination methods of k and c parameters on the ability of weibull distribution to suitably estimate wind potential and electric energy. int. journal of renewable energy development, 3(2), 145-154. http://dx.doi.org/10.14710/ijred.3.2.145-154 * corresponding author: tel.:+ 237 77 46 10 06 / + 237 95 83 35 64 email: ruben.mouangue@univ-ndere.cm / r_mouangue@yahoo.fr 1. introduction energy is an essential ingredient of socio-economic development and economic growth. the need for implementing new and clean energy technologies became pressing due to the awaited rarefaction of the fossil resources on which our development since nearly two centuries was built (omer 2008; rahman et al. 2014). in the interest of sustainable development, the implementation of renewable solutions in complement of other existing solutions is not any more to prove. because of pollution and greenhouse gas, the wind energy, which is a reliable and promising renewable energy, have attracted increasing attention due to their almost inexhaustible and nonpolluting characteristics (li & li 2005; omer 2008). the conversion of wind http://dx.doi.org/10.14710/ijred.3.2.145-154 mailto:ruben.mouangue@univ-ndere.cm citation: mouangue, r.m., kazet, m.y., kuitche, a. & ndjaka, j.m. (2014) influence of the determination methods of k and c parameters on the ability of weibull distribution to suitably estimate wind potential and electric energy. int. journal of renewable energy development, 3(2), 145-154 doi: 10.14710/ijred.3.2.145-154 p a g e | 146 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved energy for the electrical production or pumping could thus help to solve a certain number of problems of the african populations. before thinking about wind turbine installation, it is necessary to have minimum information on wind characteristics such as the observed frequency distributions of wind speed, the wind energy density, the predominant direction of wind speed, and the daily and seasonal variations of wind. the first part of this paper consists of the presentation of our research motivation and objectives and a summary description of our site of study. secondly, we will make a detailed description of the weibull and hybrid wind distribution models as well as a presentation of the determination methods of k and c parameters including in particular the maximum likelihood method and the modified maximum likelihood method. then, we will evaluate the mean wind speed, the standard deviation and the mean square error using the different distributions described. in order to predict the provided electric energy, we approximate a vestas v82 power curve. the third part is a description of wind conditions data. we will present the obtained results and discuss them in our last part. 1.1 research motivation and objectives the increasingly rising interest in estimating wind power and wind energy potential at a given site highlights the importance of the statistical simulation of wind speed observations. research and studies on wind power assessment and its applications in ngaoundere has not yet been carried out. the realization of this research may enables professionals to identify, clearly and precisely, areas which could be suitable for the establishment of future parks of wind energy for sustainable development. the wind speed distribution is of great importance for the wind energy potential assessment and for the performance of the wind energy conversion system. the weibull distribution with two parameters is versatile and is commonly used for fitting the measured wind speed probability distribution (seguro & lambert 2000; kaldellis 2008; safari & gasore 2010; boudia et al. 2013; kazet et al. 2013). a few years ago, the preferred method of calculating the weibull parameters was a graphical technique which entailed generating the cumulative wind speed distribution, plotting it on a special weibull graph paper, and drawing a line of best fit (takle & brown 1978; seguro & lambert 2000). later, this procedure was implemented by performing a linear regression on a computer (jamil et al. 1995). seguro and lambert (2000), by using sample data which did not have null speed measurements (calm), demonstrated that the maximum likelihood method is a more suitable computer-based method for estimating the weibull parameters. the weibull distribution with parameters which are calculated by this method showed some deficiencies for sample data of sites having calm (salami et al. 2013). the use of hybrid distribution has also been noted many times: takle and brown (1978), salami et al. (2013). for sites having nonzero probability of calm, these authors pointed out the fact that the weibull distribution is not suitable for the modeling of the wind speed distributions and for the wind potential assessment. they recommend the use of the hybrid distribution for these cases. today, there are several ways to estimate k and c parameters. some of the most used are the maximum likelihood method, the least square method and the standard deviation method (ramírez & carta 2005; sathyajith 2006; safari & gasore 2010; morales et al. 2012). the processes of the calculation of k and c parameters being different, disparities in the results could thus influence the accuracy of a distribution in the modeling of the wind speed frequency histogram. time-series wind data collected at the ngaoundere meteorological station show that the calm is very important. in this work, our purpose is to show that the weibull distribution can also be used for sites having nonzero probability of calm, in the condition that the weibull parameters determination method is well selected. using data collected at a site of ngaoundere, we evaluate the observable wind speed distribution and the wind energy density that we model by using the weibull distribution and the hybrid distribution approaches. finally, we provide an estimate of the electric output energy which could be produced on this site. 1.2 brief description of the site ngaoundere is the capital of the adamawa region of cameroon. it lies at the northern end of the railway to yaounde (capital of cameroon) and is also home to an airport. ngaoundere is the most important town of the region and is the highest habitable zone of the country with an elevation of 1104 m (asecna 2012). the asecna weather service (agency for the safety of air navigation in africa and madagascar) is located at the airport and the geographic longitude – latitude coordinates of the wind data collection mast are 13° 33’ 46.26’’ e and 7° 21’ 24.37’’ n. 2. materials and methods 2.1 weibull wind distribution (wwd) model the probability density function f(v) indicates the fraction of time (or probability) for which the wind is at a given speed v. it is given by equation (1).                       kk c v c v c k vf exp)( 1 (1) int. journal of renewable energy development 3 (2) 2014: 145-154 p a g e | 147 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved where:  k is the weibull shape parameter,  c is the scale parameter (m/s),  v is the wind speed (m/s),  f(v) is the probability density function. the cumulative distribution function f(v) gives us the fraction of time (or probability) that the wind speed is equal or lower than v. it is the integral of the probability density function. thus,                 v k c v dvvfvf 0 exp1)()( (2) 2.1.1 determination of weibull parameters a) the maximum likelihood method by using this method, the shape factor k and the scale factor c are estimated solving the following two equations (salami et al. 2013; boudia et al. 2013): 1 1 1 1 )()(                     n vln v vlnv k n i i n i k i n i i k i (3) kn i k i v n c 1 1 1          (4) where:  vi is the wind speed in time step i,  n is the number of nonzero wind speed data points. equation (3) is solved using an iterative procedure. in this work, this is performed using a fortran 90 code and the initial guess used is k = 2 (seguro & lambert 2000). after which, equation (4) is solved explicitly. b) the modified maximum likelihood method according to seguro and lambert (2000), this method most be used when data wind speed are available in the frequency distribution format. the weibull parameters are estimated using equation (5) and (6). 1 1 1 1 )0( )()( )( )()(                      vf vfvln vfv vfvlnv k n i ii n i i k i n i ii k i (5) kn i i k i vfv vf c 1 1 )( )0( 1           (6) where:  vi is the wind speed central to bin i,  n is the number of bins,  f(vi) is the frequency with which the wind speed fall within bin i,  f(v ≥ 0) is the probability that the wind speed equals or exceeds zero. in this work, equation (5) is also solved using an iterative procedure with a fortran 90 code. the initial guess is also k = 2. after which, equation (6) is solved explicitly. 2.1.2 mean speed, standard deviation and energy density a) mean speed the mean wind speed of a regime, following the weibull distribution, is given by equation (7) (sathyajith 2006; kaldellis 2008):          k cdvvvfv m 1 1)( 0 (7) here, γ( ) is the gamma function. b) standard deviation the standard deviation shows how much variation or dispersion from the average exists. a low standard deviation indicates that the data points tend to be very close to the mean. a high standard deviation indicates that the data points are spread out over a large range of values. following the weibull distribution, the standard deviation of the wind speed is calculated by equation (8). 2 1 2 1 1 2 1                    kk c (8) c) energy density the total energy, contributed by all possible speeds in the wind regime, available for unit rotor area and time may be expressed as (sathyajith 2006; sathyajith & geeta 2011):    0 3 )( 2 1 dvvfve d  (9) hence, citation: mouangue, r.m., kazet, m.y., kuitche, a. & ndjaka, j.m. (2014) influence of the determination methods of k and c parameters on the ability of weibull distribution to suitably estimate wind potential and electric energy. int. journal of renewable energy development, 3(2), 145-154 doi: 10.14710/ijred.3.2.145-154 p a g e | 148 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved        k ce d 3 1 2 1 3  (10) ρ = 1.03 kg/m3 is the mean air density at the airport of ngaoundere (asecna 2012). d) relative error and mean square error the relative error expresses the variation, as a percentage, between the predicted value and the awaited value. the relative error (re) on the mean energy density is calculated by the equation (11): 100* dd dpdd e ee re   (11) where:  edd is the energy density obtained from data,  edp is the energy density predicted,  re is the relative error (%). the mean square error (mse) is one of many ways to quantify the difference between values implied by an estimator and the true values of the quantity being estimated. mse here is calculated by the equation (12).     n i ii pv n mse 1 21 (12) here:  vi is the actual wind speed,  pi his predicted value. 2.1.3 electric output energy the question of primary interest in wind power applications is of course: what power production can be expected from a given wind turbine at a given site? to answer this, it is necessary to know the power curve of the wind turbine as well as the probability density function of the wind speed at hub height. the power production by a wind turbine varies with the wind that strikes the rotor. it is common practice to use the wind speed at hub height as a reference for the power response of the wind turbine. the power produced as function of the wind speed at hub height is conventionally called the power curve. the fig. 1 below shows the power curve of vestas v82 wind turbine generator (vestas technology documentation 2005). when the wind speed is less than the cut-in wind speed, the turbine will not be able to produce power. when the wind speed exceeds the cut-in speed, the power output increases with increasing wind speed to a maximum value, the rated power; thereafter the output is almost constant. at wind speeds higher than the cut fig. 1 power curve of vestas v82 wind turbine generator out wind speed the wind turbine is stopped to prevent structural failures. the product of the power curve and the probability density function of the wind speed gives the power density curve, the integral of which is the mean power production pel (bataineh and dalalah, 2013):   n d a n v v v v nel dvvfpdvvfvpp )()()( (13) actual power curves are rather smooth and can be well approximated by a piece-wise linear function with a few nodes (pi, vi) (troen & petersen 1989). it is possible to carry out an approximation which consists of assuming that the variation between two nodes of the power-wind speed curve is linear (troen & petersen 1989; carta et al. 2008). then, given two points ‘‘i” and “i + 1” of the power curve, power as a function of speed can be written as equation (14). ii ii ii pvv vv pp vp       )()( 1 1 (14) hence, the annual energy production is: helel npe * (15) nh is the time period of data collection (in hours). 2.2 hybrid wind distribution (hwd) for data sets having high probability of calm and a low value of scale parameter c, the weibull function is not likely to provide a good fit to the data (takle & brown 1978). the problem of properly including calm periods into the distribution is reduced by defining a int. journal of renewable energy development 3 (2) 2014: 145-154 p a g e | 149 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved hybrid density function (takle & brown 1978; salami et al. 2013). hence,                         kk oh c v c v c k fvf exp1)( 1 (16) fo is the probability of observing zero wind speed. parameters values k and c are determined by solving equation (3) and (4). the corresponding cumulative distribution function is defined by (takle & brown 1978),                          k ooh c v ffvf exp11)( (17) the mean speed, standard deviation and energy density of this distribution are then respectively,          k fcv o h m 1 11 (18)   22 2 11 h moh v k fc        (19)          k fce od 3 11 2 1 3  (20) 3. wind condition data measured wind speed data are commonly available in time-series format, in which each data point represents an average wind speed over some time period. in some instances, wind speed data may instead be available in frequency distribution format. 3.1 measurements data used in this paper was obtained at meteorological airport station of ngaoundere. data were recorded every day by 30 minutes interval (average over 10 minutes around the time of measure) at the standard height of 10 m above the ground. 3.2 vertical extrapolation the estimation of the wind resource at the hub height of a wind turbine is one of the primary goals of the site assessment. because the measurement heights of meteorological towers are typically significantly lower than turbine hub height, a mathematical model is generally needed to extrapolate the measured wind resource at the lower measurement height to the hub height of the turbine. in this work, we used the hellmann exponential law defined by equation (21), which is one of the most commonly used models (lackner et al. 2010; đurišić & mikulović 2012; bataineh & dalalah 2013).           r r h h vv (21) where:  α is the wind shear coefficient, dependent mainly on the terrain roughness and atmospheric stability,  hr is the reference height (m),  h is the height desired (m),  vr is the wind speed at the reference height (m/s),  v is the wind speed at the height desired (m/s). for our site, α = 0.3365 (asecna 2012). 3.3 data processing to succeed in it, we used usual equations below: mean speed      n i i n i ii vf vfv v 1 1 )( )( (22) standard deviation         n i i n i ii vf vvvf 1 1 2 )( )(  (23) energy density      n i i n i ii d vf vfv e 1 1 3 )( )( 2 1  (24) citation: mouangue, r.m., kazet, m.y., kuitche, a. & ndjaka, j.m. (2014) influence of the determination methods of k and c parameters on the ability of weibull distribution to suitably estimate wind potential and electric energy. int. journal of renewable energy development, 3(2), 145-154 doi: 10.14710/ijred.3.2.145-154 p a g e | 150 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved 3.3 picture of the site fig. 2 mast of the wind data collection of the ngaoundere airport 4. result and discussion by using the ngaoundere wind data, the weibull wind distribution model is applied to two different cases: k and c parameters are determined by the modified maximum likelihood method (momalime) (seguro & lambert 2000). k and c parameters are determined by the maximum likelihood method (malime) (seguro & lambert 2000; boudia et al. 2013). to evaluate the accuracy of our approach, the model is investigated and compared to the hybrid wind distribution model (takle & brown 1978; salami et al. 2013). 4.1 wind potential two cases of the weibull distribution and the hybrid distribution are fitted to the observed wind speed frequency histogram. fig. 3 shows the wind speed probability density function estimated by different methods and models. the solid line represents the weibull distribution in which parameters are calculated by the momalime. the dash dot line represents the weibull distribution in which parameters are calculated by the malime. finally, the long dash line represents the hybrid density function. it is observed in fig. 3 that the curve of the weibull probability density function which use momalime match better the histogram. fig. 4 shows the wind speed cumulative distribution functions. the goodness of fit of the various fitted distributions to the wind speed data is evaluated using the quantile – quantile plots, the pearson’s coefficient of correlation r associated with the probability density function (pdf) plots, the mean speed with its related standard deviation and mse, the energy density and its relative error. fig. 3 wind speed probability density functions fig. 4 wind speed cumulative distribution functions fig. 5 quantile–quantile plot of the weibull and hybrid distributions int. journal of renewable energy development 3 (2) 2014: 145-154 p a g e | 151 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved 4.1.1 quantile – quantile (q–q) plot in statistics, a q–q plot is a probability plot, which is a graphical method for comparing two probability distributions by plotting their quantiles against each other (gibbons & chakraborti 2003). from the observed statistical set, one calculates some sample quantiles xi. if the statistical set fit well the chosen theoretical distribution, one should have sample quantiles xi equal to quantiles xi* related to the theoretical model. then, one represent points mi(xi*, xi) with theoretical quantiles on x-axis and sample quantiles on y-axis. if the chosen theoretical distribution is pertinent, fitted points must follow the line y = x (zhang et al. 2013). theoretical quantiles are calculated by the equation (25). )( 1* ii pfx   (25) where:  pi is the probability related to the sample quantile xi,  f -1 is the reciprocal function of the cumulative density function. it is observed that the momalime follows the line y = x more closely than other distributions, as one can see it in fig. 5. 4.1.2 correlation coefficient of pearson the coefficient of correlation is a measure of the agreement between an estimated distribution and the recorded data (zhang et al. 2013). the coefficient of correlation r between the paired pdf data values is evaluated by the following equation (26),                           n i i n i i n i ii yyxx yyxx r 1 2 1 2 1 (26) where:  xi are the observed data values,  yi the predicted data values. the closer the value of r is to one, the more the fitted distribution agrees with the observed data. table 1 shows the comparison of the coefficient of correlation for different models. it is observed that the weibull model with momalime has the largest r value. this observation illustrates the strong potential of this approach to provide accurate representations of wind distribution. regarding the probability density function, the accuracy of fit to the wind speed data of the weibull distribution using momalime is higher than others and this is confirmed by the calculated correlation coefficient. table 1 correlation coefficient of pearson distribution model weibull hybrid momalime malime r 0.908101857 0.781738520 0.781743824 4.1.3 mean speed, standard deviation and mse we applied the previously presented equations to the calculation of the mean speed, the standard deviation and the mse for each month. the results are presented in table 2. it is seen that values of the mean speed obtained by the momalime are more closer to those obtained from data than others approaches. furthermore, most mse values calculated by the momalime are closer to those of data than others mse values (table 2). 4.2 energy density estimation 4.2.1 the height is 10 m for each month of the collection period, the energy density is calculated and the results are presented in table 3. here, the observation is the same as previously. the values of the mean energy density calculated by momalime are closer to those obtained from data than those calculated from malime and hybrid. 4.2.2 relative error on energy density estimated for a better appreciation of the accuracy of momalime, we evaluated the relative errors on the energy density calculated from different models used. fig. 6 shows the estimated relative errors of hybrid distribution and weibull distribution with its two cases (momalime, malime). observation shows that weibull distribution with parameters which are calculated by the momalime performs better than others, the energy density estimations. fig. 6 relative error on the estimated energy density at 10 m of height citation: mouangue, r.m., kazet, m.y., kuitche, a. & ndjaka, j.m. (2014) influence of the determination methods of k and c parameters on the ability of weibull distribution to suitably estimate wind potential and electric energy. int. journal of renewable energy development, 3(2), 145-154 doi: 10.14710/ijred.3.2.145-154 p a g e | 152 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved table 2 monthly results of mean speed, standard deviation and mse calculated from different approaches and models table 3 monthly results of energy density and weibull parameters calculated from different approaches and models months data weibull hybrid momalime malime σ mse vm σ mse vm σ mse vmh σ mse may-11 1.7880 1.3845 0.2130 1.7955 1.3730 0.2094 2.0292 1.3410 0.1998 1.1849 1.4321 0.2563 jun-11 1.4379 1.2108 0.1832 1.4490 1.1499 0.1652 1.8844 1.1898 0.1769 0.8892 1.2461 0.1941 july-11 1.4419 1.1851 0.1755 1.4528 1.1205 0.1569 1.6405 1.1308 0.1598 0.8282 1.1482 0.1648 aug-11 1.6909 1.3071 0.2135 1.6991 1.2865 0.2068 1.8870 1.2614 0.2273 1.0791 1.3348 0.2227 sept-11 1.4060 1.1912 0.1773 1.4169 1.1194 0.1566 1.6571 1.1821 0.1746 0.7820 1.1594 0.1680 oct-11 1.2920 1.1093 0.1538 1.3055 1.0291 0.1323 1.6033 1.1067 0.1531 0.6862 1.0741 0.1442 nov-11 1.4709 1.3214 0.2182 1.4794 1.2633 0.1995 2.0048 1.4039 0.2464 0.8259 1.3365 0.2232 dec-11 1.5239 1.3264 0.2199 1.5327 1.2868 0.2069 1.9201 1.3557 0.2297 0.8352 1.3063 0.2133 janv-12 1.5760 1.4072 0.2475 1.5846 1.3565 0.2300 1.9812 1.4577 0.2656 0.8893 1.3874 0.2406 feb-12 1.7949 1.5303 0.2927 1.8027 1.5159 0.2872 2.1619 1.5638 0.3056 1.1220 1.5612 0.3046 mar-12 1.7869 1.5246 0.2905 1.7929 1.5323 0.2935 2.2623 1.5883 0.3153 1.1039 1.5845 0.3138 apr-12 1.6369 1.2908 0.2082 1.6460 1.2677 0.2008 1.9567 1.2409 0.2009 1.0643 1.3370 0.2234 global 1 year 1.5689 1.3305 0.2212 1.5781 1.2840 0.2061 1.9114 1.3298 0.2211 0.9365 1.3343 0.2225 months data weibull hybrid momalime malime d k c ed k c ed k c fo ed may-11 9.4387 1.320 1.950 9.9149 1.544 2.255 11.1553 1.544 2.255 0.416 6.5165 june-11 5.8483 1.269 1.561 5.5937 1.623 2.104 8.3416 1.623 2.104 0.528 3.9361 july-11 5.7643 1.308 1.575 5.3375 1.475 1.813 6.3043 1.475 1.813 0.495 3.1834 aug-11 7.9364 1.334 1.849 8.2493 1.525 2.094 9.1331 1.525 2.094 0.428 5.2239 sept-11 5.6059 1.275 1.528 5.1851 1.421 1.822 6.8828 1.421 1.822 0.528 3.2502 oct-11 4.4782 1.276 1.406 4.0318 1.473 1.772 5.8992 1.473 1.772 0.572 2.5256 nov-11 6.7628 1.175 1.564 6.9129 1.450 2.211 11.8120 1.450 2.211 0.588 4.8686 dec-11 7.0494 1.196 1.628 7.4172 1.437 2.115 10.5215 1.437 2.115 0.565 4.5807 janv-12 8.2415 1.172 1.674 8.5389 1.375 2.167 12.4148 1.375 2.167 0.551 5.5742 feb-12 11.0278 1.194 1.914 12.1094 1.400 2.372 15.6543 1.400 2.372 0.481 8.1321 mar-12 10.6132 1.174 1.895 12.3265 1.446 2.494 17.0455 1.446 2.494 0.512 8.3222 apr-12 7.4298 1.310 1.785 7.7419 1.615 2.184 9.3991 1.615 2.184 0.456 5.1142 global 1 year 7.5026 1.236 1.690 7.5932 1.460 2.110 10.1252 1.460 2.110 0.510 4.9657 int. journal of renewable energy development 3 (2) 2014: 145-154 p a g e | 153 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved table 4 monthly wind characteristics calculated using weibull distribution with momalime monthly period weibull parameter (momalime) mean speed energy density k c (m/s) (m/s) ed (w/m2) may-11 1.60 4.00 3.58 58.93 jun-11 1.44 3.10 2.82 33.19 jul-11 1.33 3.10 2.89 39.34 aug-11 1.62 3.80 3.38 49.52 sept-11 1.35 3.20 2.92 41.83 oct-11 1.21 2.90 2.75 40.80 nov-11 1.28 4.50 4.15 130.49 dec-11 1.37 4.80 4.43 135.97 jan-12 1.36 4.70 4.30 129.76 fev-12 1.31 5.20 4.84 188.01 mar-12 1.40 5.80 5.28 225.55 apr-12 1.54 3.60 3.26 45.84 4.2.3 extrapolation at 120 m by applying equation (21) on the set of measured data taken at the measurement height, one obtains these synthetic sets of data at the desired height of 120 m (table 4). 4.3 electric output energy the monthly and annual energy productions are calculated using equation (15). the time period of the data collection is one year that is to say 8784 hours. the power curve of the wind turbine generator used is that of vestas v82 (fig. 1) which has the following characteristics: cut-in wind speed (3 m/s), cut-out wind speed (20 m/s) and a rated power of 1650 kw. fig. 7 the monthly electric output energy calculated using the weibull distribution and the wind power curve of vestas v82 wind turbine generator table 5 global wind characteristics calculated using weibull distribution with momalime period of 12 months weibull parameter (momalime) mean speed energy density output energy k c (m/s) (m/s) ed (w/m2) eel (mwh) 05/2011 to 04/2012 1.26 4.00 3.75 95.85 1332.5 fig. 7 shows the monthly energy outputs for the airport site. the most energy contribution is made in march with 255.7 mwh, while the least energy contribution is observed to be 41.1 mwh in june. while using the synthetic set of data from 120 m of height, we obtained some global wind characteristics. table 5 presents the weibull parameters, the mean speed, the energy density and the annual energy production for the whole data used. from the whole of the results, it is observed that quantiles calculated from the weibull distribution in which parameters are determined by the momalime follows the line y = x more closely than other distributions. furthermore, values of the mean speed and energy density calculated from the same distribution are very close to those obtained from data. this is not the case of the malime and the hybrid distribution. low values of mse and relative error obtained by using momalime suggests that the proposed approach could be more usefull for the wind energy assessment of our site than the weibull distribution in which parameters are determined by malime or than the hybrid distribution. for the particular site of ngaoundere airport, it could be concluded that the wind turbine generator doesn’t works for at least a half of the time because of higher frequency of calm. analysis of the observed wind speed distribution for this specific site proves that fact. according to whether one uses the momalime or the malime, it thus appears in an obvious way that the determination method of weibull parameters has an influence on the relevance of this distribution. the calculation principle of k and c parameters by using malime rely on a speed logarithmic calculation. however, 23 % of the velocity measurements of the site are null and of this fact are not taken into account during the calculation of the parameters what has certainly an influence on the awaited result. 5. conclusion our study was to estimate and characterize the observed wind potential and to provide the electric output energy of the site of ngaoundere by using the weibull distribution in which parameters are determined by the modified maximum likelihood method because of the calm wind observed. the maximum likelihood method and the hybrid citation: mouangue, r.m., kazet, m.y., kuitche, a. & ndjaka, j.m. (2014) influence of the determination methods of k and c parameters on the ability of weibull distribution to suitably estimate wind potential and electric energy. int. journal of renewable energy development, 3(2), 145-154 doi: 10.14710/ijred.3.2.145-154 p a g e | 154 © ijred – issn: 2252-4940, 15 july 2014, all rights reserved distribution were also used to compare them to our approach. taking the results obtained into account, one can retain at the end of this study the following:  the maximum likelihood method, which is one of the most used determination method of the weibull parameters, gives results which are not very satisfactory taking data into consideration. that could be due to the important rate of calm observed in measurements of the site.  although the hybrid distribution was proposed by salami et al. (2013) as the best model that fits the frequency histogram of wind speed and which estimate with precision the amounts of the wind energy, it seems to be not quite satisfactory in the case of the site of ngaoundere.  the determination methods of weibull parameters affect the accuracy of this distribution to model the wind potential of a site. the modified maximum likelihood method proposed by seguro and lambert in 2000 which is recommended for use with wind data in frequency distribution format, is one of the best methods which can be used to determinate weibull parameters for sites which have important frequency of calm. by using this method, the weibull distribution appear to be fine for the modeling of the wind potential and for the wind energy assessment, particularly for sites having nonzero probability of calm. acknowledgements authors would like to thank the asecna weather service of ngaoundere to have provided the meteorological data used in this work. references asecna (2012) archives data of the weather station of ngaoundere. bataineh, k.m. & dalalah, d. (2013) assessment of wind energy po tential for selected areas in jordan. renewable energy, 59, 75-81. boudia, s.m., benmansour, a., ghellai, n., benmedjahed, m. & hellal, m.a.t. (2013). temporal assessment of wind energy resource at four locations in algerian sahara. energy conversion and management, 76, 654–664. carta, j.a., ramírez, p. & velázquez, s. (2008) influence of the level of fit of a density probability function to wind-speed data on the wecs mean power output estimation. energy conversion and management, 49, 2647–2655. đurišić, ž. & mikulović, j. (2012) a model for vertical wind speed data extrapolation for improving wind resource assessment using wasp. renewable energy, 41, 407–411. gibbons, j., chakraborti, s. (2003) nonparametric statistical inference. crc press. jamil, m., parsa, s. & majidi, m. (1995) wind power statistics and an evaluation of wind energy density. renewable energy, 6, no. 5-6, 623–628. kaldellis, j.k. (2008) maximum wind potential exploitation in autonomous electrical networks on the basis of stochastic analysis. journal of wind engineering and industrial aerodynamics, 96, 1412– 1424. kazet, m., mouangue, r., kuitche, a., ndjaka, j.m. & takam, s. (2013) modélisation et simulation numérique des données du vent en vue d’une prédiction de l’énergie électrique d’origine éolienne : cas d’un site de la ville de ngaoundere au cameroun. revue des energies renouvelables, 16(3), 527 – 538. lackner, m.a., rogers, a.l., manwell, j.f. & mcgowan, j.g. (2010) a new method for improved hub height mean wind speed estimates using short-term hub height data. renewable energy, 35, 2340– 2347. li, m. & li, x. (2005) mep-type distribution function: a better alternative to weibull function for wind speed distributions. renewable energy, 30, 1221–1240. morales, l., lang, f. & mattar, c. (2012) mesoscale wind speed simulation using calmet model and reanalysis information: an application to wind potential. renewable energy, 48, 57–71. omer, a.m. (2008) on the wind energy resources of sudan. renewable and sustainable energy reviews, 12, 2117–2139. rahman, m.m., mostafiz, s.b., paatero, j.v. & lahdelma, r. (2014) extension of energy crops on surplus agricultural lands: a potentially viable option in developing countries while fossil fuel reserves are diminishing. renewable and sustainable energy reviews, 29, 108–119. ramírez, p. & carta, j.a. (2005) influence of the data sampling interval in the estimation of the parameters of the weibull wind speed probability density distribution: a case study. energy conversion and management, 46, 2419–2438. safari, b. & gasore, j. (2010) a statistical investigation of wind characteristics and wind energy potential based on the weibull and rayleigh models in rwanda. renewable energy, 35, 2874– 2880. salami, a.a., ajavon, a.s.a., kodjo, m.k. & bedja, k.s. (2013) contribution to improving the modeling of wind and evaluation of the wind potential of the site of lome: problems of taking into account the frequency of calm winds. renewable energy, 50, 449– 455. sathyajith, m. (2006) wind energy fundamentals, resource analysis and economics, pp. 68-83. springer-verlag berlin heidelberg. sathyajith, m. & geeta, s.p. (2011) advances in wind energy conversion technology, pp. 74-80. springer-verlag berlin heidelberg. seguro, j.v. & lambert, t.w. (2000) modern estimation of the parameters of the weibull wind speed distribution for wind energy analysis. journal of wind engineering and industrial aerodynamics, 85, 75–84. takle, e.s. & brown, j.m. (1978) note on the use of weibull statistics to characterize wind–speed data. journal of applied meteorology, 17, 556–559. troen, i. & petersen, e.l. (1989) european wind atlas. 1st ed. denmark: risø national laboratory. vestas technology documentation (2005) general specification of v821.65 mw. http://www.vestas.com/files/filer/en/brochures/pro ductbrochurev821_65_uk.pdf. accessed on 12 december 2013. zhang, j., chowdhurya, s., messac, a. & castillo, l. (2013) a multivariate and multimodal wind distribution model. renewable energy, 51, 436–447. http://www.vestas.com/files/filer/en/brochures/productbrochurev821_65_uk.pdf http://www.vestas.com/files/filer/en/brochures/productbrochurev821_65_uk.pdf samsudin_ijred_as_pf_sw int. journal of renewable energy development 10 (3) 2021: 459-469 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 459 contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: http://ejournal.undip.ac.id/index.php/ijred energy analysis and remixing effect of thermal coupling petlyuk column for natural gas liquid (ngl) fractionation train rohani mohd zin*, mohd ammar abidin, munawar zaman shahruddin chemical engineering faculty, universiti teknologi mara (uitm), 40700 shah alam, selangor, malaysia. abstract. in this work, a non-conventional distillation sequence with thermal coupling (petlyuk column) was presented as a technique to perform the separation of the ngl consist of ethane, propane, butane or other higher alkanes. the improvements were investigated through the energy analysis and remixing effect. from the result obtained, it was found that the petlyuk arrangement consumes less amount of energy and able to reduce the remixing effects as compared to the conventional column sequencing. the petlyuk arrangement saved about 44.49% and 12.83% in terms of cooling and heating duty, respectively. the overall annual energy saving shown by this arrangement is 39.22%. this arrangement proved to be able to prevent the remixing effect occurrence that contributes to thermal and separation inefficiency. the desired separation efficiency also obtained by this arrangement as all the product specifications are met. the ability in avoiding remixing effect by the petlyuk column permits a significant reduction in co2 emission with an average of 29.43 % of each equipment involved. hence, it can be concluded that the petlyuk arrangement model is a better alternative to be implemented in the ngl fractionation train. keywords: distillation, fractionation train, thermal coupling, petlyuk column, remixing effect article history: received: 25th sept 2020; revised: 5th february 2021; accepted: 14th february 2021; available online: 18th february 2021 how to cite this article: zin, r.m., abidin, m.a., and shahruddin, m.z. (2021) energy analysis and remixing effect of thermal coupling petlyuk column for natural gas liquid (ngl) fractionation train. international journal of renewable energy development, 10(3), 459-469 https://doi.org/10.14710/ijred.2021.33094 1. introduction distillation is both energy-intensive and efficient processes in the chemical industry, responsible for the largest fraction of an immense amount of energy consumed. distillation operation became a major concern within the sustainability challenge for a primary target of energy-saving efforts in industrially developed (humphrey et al., 1997, matla-gonzález et al., 2013, lucero-robles et al., 2016). although new separation methods are continuously being explored, distillation remains the most frequently used separation process (agrawal et al., 1999; agrawal, 2003). the natural gas processing plant is an example of an industry that dominates these separation processes (manley, 1996; devold. 2009). natural gas liquid (ngl) is a raw material for the petrochemical industry typically consists of ethane, propane, butane or other higher alkanes in a blend (devold 2009). to accomplish the separation of natural gas liquid (ngl), a series of conventional distillation column namely deethanizer, depropanizer and debutanizer columns is used. the separation process requires a series of distillation train because it is impossible to produce more than two products with a single distillation column. due to tight environmental * corresponding author: rohanimz@uitm.edu.my regulations and the high energy cost, there is a need to improve the energy efficiency of distillation systems in a natural gas processing plant. energy analysis is an important approach to understand any process and optimizing the overall energy consumption. various energy analysis works through modelling and optimization of various real-life industry case studies have been conducted by several researchers (hoseinzadeh et al., 2019, kariman et al., 2019, hoseinzadeh et al., 2020 (a), hoseinzadeh et al., 2020(b)). typically, for the conventional distillation arrangements energy integration method has been used to reduce energy costs (shahruddin et al., 2019). however, the scope for energy integration of the conventional distillation columns into an overall process is often limited as the practical constraints prevent integration of distillation columns with the rest of the process (annakou et al., 1996, lucerorobles et al., 2016, egger et al., 2018). since the columns cannot be integrated with the rest of the process and the potential for heat integration is limited by the heat flows in the background process, thus attention must be turned to the distillation operation and look at unconventional arrangement opportunity. research article citation: zin, r.m., abidin, m.a., and shahruddin, m.z. (2021) energy analysis and remixing effect of thermal coupling petlyuk column for natural gas liquid (ngl) fractionation train. int. journal of renewable energy development, 10(3), 459-469, doi: 10.14710/ijred.2021.33094 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 460 the natural gas processing plant is currently using the conventional arrangement that consumes of intensive heat utilization of the overall plant (manley, 1996, matlagonzález et al., 2013). due to a large percentage of heat used in operating the distillation columns, it contributes to a large amount of operating cost and capital cost. alternatively, non-conventional arrangements which is petlyuk columns are the options for these problems. petlyuk column is especially energy-efficient and due to their ability to save capital and energy cost (fidkowski 2006, dwivedi et al., 2013, matla-gonzález et al., 2013, lucero-robles et al., 2016, egger et al., 2018). this unique distillation arrangements have become very attractive although it has rarely been implemented in industries due to the complex arrangement (ramirez-corona et al., 2010), lack of knowledge and understanding and scepticism towards practical issues such as control (hernandez et al., 1999). for the gas processing plant industries, especially in malaysia none of the plants has implemented any of these complex distillation arrangements yet. only in the recent years, the petlyuk arrangement was successfully applied in industry for a limited amount of feed components (matla-gonzález et al., 2013, errico et al., 2015, egger et al., 2018). this indicates that these problems may be resolved by this non-conventional arrangement. thus, in this work, the scope of knowledge extends to the complex distillation arrangements on ngl fractionation train. the present work investigates the energy analysis, purity of the final products (product specifications) and remixing effects of the thermally coupled petlyuk column as the non-conventional arrangement in ngl fractionation train. comparisons are made with the current conventional arrangements. the potential of reducing the co2 emissions through the implementation of petlyuk column is also investigated. comparison analysis is carried out on the potential reduction of co2 mitigation between conventional distillation column and petlyuk column. 2. ngl separation natural gas generally comprises 90% methane and 10% other light alkanes. in its marketable form, natural gas is processed to obtain a specific composition of hydrocarbons, sour and acid ((hydrogen sulfide (h2s) and carbon dioxide(co2)) permissible quantity and energy contents. natural gas liquids are primarily raw materials for a petrochemical industry typically consist of ethane, propane, butane or other higher alkanes in a blend or separately (devold 2009). as the energy demands growing, global oil resources need to be utilized to the fullest and as efficiently as possible (underwood et al, 2015). with the increasing demands and limited resources, it has become a driver for the industry to overcome the problems and find solutions to meet the market demands. one of the methods to overcome the problem is through the optimization of the processing plant and produces high-quality products with low energy consumption and consequently cost-effective. 2.1 conventional ngl fractionation train a typical ngl fractionation train is shown in figure 1 and consists of a series of conventional distillation columns. in this processing plant, the natural gas feed will be separated into individual component to produce methane, ethane propane and butane using a demethanizer, a deethanizer, a depropanizer and a debutanizer column respectively. in some other gas processing plants, the mixed butane product can be further split in a deisobutanizer to produce normal and isobutane. however, the scope of energy analysis of this work will focus on a deethanizer, a depropanizer and a debutanizer column. future work will be done involving the de-methanizer column, as the energy utilization for the column is different from typical distillation operation since it involves a complex cryogenic separation process. in typical gas plant (figure 1), the ngl liquid enters the deethanizer, which separates the ethane from the mixed ngl liquid. the column operates with a partial condenser and produces a cold ethane gas product at the top of the column (as shown in table 1). the bottom product from the deethanizer feeds the depropanizer. a propane stream is taken from the depropanizer reflux drum as the top product as indicated in table 1. the debutanizer removes the butane as the top product from the remaining ngl liquid. pentane plus (pentane with higher alkanes component), or natural gasoline, is taken as a bottom product from the debutanizer as shown in table 1 and figure 1. a train or sequence of separators is required because it is often impossible to produce more than two products with single separation equipment. by proceeding from the lightest hydrocarbons to the heaviest, it is possible to separate the different ngls reasonably easily (essam bahnassi et al., 2005). table 1 summarized the different types of fractionators that are considered in this study for the natural gas liquid (ngl) fractionation train. 2.2 non-conventional arrangements (thermally coupled petlyuk column) one of the most important non-conventional distillation arrangements involves thermal coupling (agrawal et al., 1999; agrawal, 2006). in this work, the distillation sequences with thermal coupling were presented as a technique to perform a separation with less energy consumptions and reduced remixing effects as compared to the conventional simple column sequencing. table 1 types of fractionator types of fractionator feed top product bottom product deethanizer ngl liquid/ (demethanizer bottom) ethane propane plus depropanizer deethanizer bottom propane butanes plus debutanizer depropanizer bottom butane pentanes plus int. journal of renewable energy development 10 (3) 2021: 459-469 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 461 fig 1. conventional fractionation trains of natural gas (essam bahnassi et al., 2005). fig 2. the evolution of distillation arrangements (nath et al., 1981). citation: zin, r.m., abidin, m.a., and shahruddin, m.z. (2021) energy analysis and remixing effect of thermal coupling petlyuk column for natural gas liquid (ngl) fractionation train. int. journal of renewable energy development, 10(3), 459-469, doi: 10.14710/ijred.2021.33094 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 462 a non-conventional column that realizes energy and capital cost savings is the petlyuk column (as shown in figure 2). this type of configuration was discovered and patented by wright in 1949 (underwood, 1946), but only in the last years has become more attractive and its applicability more realistic (matla-gonzález et al., 2013, errico et al., 2015, egger et al., 2018). in this configuration, the columns are thermally-coupled as illustrated in figure 2 (caballero et al., 2004). thermal coupling is used to describe the situation when liquid from a column is used to supply reflux and vapour is used to supply boil-up to another column (petlyuk et al., 1965). several variations of this arrangement are used to eliminate heat exchangers in column designs. in this configuration, three products can be produced from the two columns with one reboiler and one condenser, as compared to a conventional arrangement. figure 2 shows the overall evolution from the distributed sequencing to the fractionator arrangement. then, it evolved finally from the fractionator column to petlyuk column. this figure also shows the evolution from the side-rectifier or side-stripper to the petlyuk column (nath et al., 1981). the principal limitation in employing this structure was the lack of the process design. however, with modern and more suitable mathematical knowledge, high modelling tools and dynamic simulations, the problem can be overcome (thompson et al., 1972). 2.3 co2 emission carbon dioxide (co2) emission is directly related to energy consumption, thus the reduction of energy consumption it will bring significant alleviation in the emission. in the natural gas liquid (ngl) fraction train, the energy consumption is contributed by the operations of reboiler and condensers. higher heat consumption led to higher fuel utilization particularly, in the overall steam usage for the reboilers in which consequently increase the co2 emissions to the environment. similarly, higher condenser duty for a distillation column also indirectly contributes to higher energy (power) consumption in the required pump to transport a large amount of cooling water needed. hence, higher energy (power) consumption directly related to the higher emission of co2. 3. methodology the most important part to be optimized is the distillation arrangement since it is responsible for the largest fraction of a vast amount of energy consumed. there are several process simulations packages available and mostly used for example aspen plus, simsci pro/ii, and unisim (strausa and skogestada, 2016). these simulation packages use sequential-modular approach for solving the flowsheet, in which each unit operation is considered as a separate block and calculated sequentially (biegler et al., 1997). hysys software package was utilized in the modelling and process simulation of natural gas liquids fractionation train in this study. hysys is a market-leading process modelling tool for conceptual design, optimization, performance monitoring for oil & gas production, gas processing, and petroleum refining industries (chowdhury, 2012). the advantages of hysys are it uses best-in-class process technologies and has the ability to provide process design knowledge to improve the profitability and efficiency of the business (sultana and amin 2011). 3.1 modelling and process simulation hysys (an abbreviation for hyprotech systems) is a powerful and efficient tool with a reasonable accuracy that offers a comprehensive thermodynamics model in determining the phase behaviours, transport properties as well as the physical properties of the components. it also offers a comprehensive library of unit operation models (oyegoke and dabai 2018, trupti et al., 2012). hysys software also normally used in process modelling with the capability to provide process design knowledge (sultana and ruhul amin 2011) to improve plant control and operability., the first stage in hysys simulation involves the definition of all components. feed components details were entered into the flowsheet by selecting the material from the component database (chemmangattuvalappil and chong 2017). table 2 shows the compositions of the feed component and its operating conditions for the fractionation of natural gas liquids (ngl) employed in this study. the data used is the typical composition of natural gas feed-in malaysia. whilst table 3 shows the product specifications of natural gas liquids (ngl) component that need to be achieved. the product specification values are based on the market requirements. in this simulation work, consider a typical gas plant as shown in figure 1. in this processing plant, the ngl liquid feed with composition as shown in table 2, enters the deethanizer column. the ethane gas is separated in this column as the top product with the required minimum purity of 0.93 mol fraction (93 mol%) as shown in table 3. the bottom product from the deethanizer feeds the depropanizer and produced propane stream with a minimum purity of 93 mol%. the debutanizer received a feed from the bottom of the depropanizer and separates the butane from the remaining feed liquid. table 2 specifications of feedstock and operating conditions for natural gas liquids (ngl) fractionation. component raw feed (mole percent) methane 0.8080 ethane 0.0566 propane 0.0277 i-butane 0.0094 n-butane 0.0057 i-pentane 0.0042 n-pentane 0.0023 n-hexane 0.0039 n-heptane 0.0032 n-octane 0.0043 carbon dioxide 0.0706 nonane 0.0003 decane 0.0006 nitrogen 0.0032 total 1.0000 inlet flow rate 21420 kmol/hr pressure 23 bar temperature -98.29ºc int. journal of renewable energy development 10 (3) 2021: 459-469 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 463 this butane product stream should be produced typically about 0.98 mol fraction (98 mol%) with allowable maximum propane content of 0.02 mol fraction (2 mol%). pentane plus (pentane with higher alkanes component) is taken out as the bottom product from the debutanizer. the pentane plus product should be produced with a requirement of maximum of 2 mol% of propane content and reid vapor pressure (rvp) of 83 kpa. the corresponding details on these products composition for each component can be referred to table 3. once all the components were computed, appropriate fluid packages were chosen to be assigned to estimate the conditions and properties after the simulation. in this simulation work, non-random two liquid (nrtl) and peng-robinson fluid packages were employed for the liquid and vapour phase respectively for the simulation. peng-robinson fluid package was chosen since the process involves the simulation of hydrocarbon at high pressure (chemmangattuvalappil and chong 2017). the simulation of conventional fractionation train was accomplished by adopting a direct sequence of four distillation columns i.e demethanizer, deethanizer, depropanizer and debutanizer (figure 3). each of these three columns (deethanizer, depropanizer and debutanizer) in this configuration is essentially a conventional distillation column with a reboiler and a condenser being used to carry out the separation. hence, it is simulated using the basic distillation column in hysys. on the other hand, a reboiled absorber is used to simulate the demethanizer as there is no condenser in this unit. for the simulation of non-conventional fractionation thermally coupled, the separations of the individual component are achieved by adopting a sequence of demethanizer, petlyuk column and debutanizer (figure 4). for this simulation, the petlyuk column is simulated using the fractionator (absorber and one main column) as opposed to typical separation. the main column and debutanizer in this configuration are essentially a conventional column with reboiler and condenser. hence, it is simulated using the basic distillation column in hysys. on the other hand, an absorber is used to simulate the fractionator as there is no condenser and reboiler while a reboiled absorber is used to simulate the demethanizer as there is no condenser in this unit. table 3. product specifications of natural gas liquids (ngl) component component (mol fraction) specification requirement metha ne ethane propane butane c5+ methane 0.85 min 0.02 max ethane 0.93 min 0.02 max propane 0.02 max 0.93 min 0.02 max 0.02 max butanes 0.04 max pentanes rvp (kpa) 83 max total product flowrate (kmol/hr) 19670 min : minimum mol fraction content. max : maximum mol fraction content/reid vapor pressure (rvp) the simulations of both conventional (figure 3) and nonconventional with petyluk (figure 4) arrangement were optimized by ensuring each of the natural gas liquids (ngl) component product specifications are achieved as per requirement stated in table 3. 3.2 co2 emission reduction for the analysis of co2 emission, the calculations and analysis are made based on several assumption and conditions. apart from complying with the products specification range, the operating pressure at the bottom of the column should only be in the range of medium or low-pressure steam as the heating media to minimize the operating cost. all calculations on the emissions to the environment resulting from heating and cooling duty are made based on the assumption that fuel used is pure methane and the emission is carbon dioxide. these assumptions are made based on justifications: 1) the heating duty will determine the amount of fuel required, thus directly correlate with the emissions amount to the environment resulting from the combustion process. 2) the cooling duty indirectly determines the power consumption needed by the pump to supply the required cooling water. hence, the power consumption will have a direct correlation with the total co2 emission. according to gadalla (2003), the heating equipment for example boiler and furnace, if combusted when mixed with air, co2 emissions will be produced based on the following: 𝐶𝑥𝐻𝑦 + '𝑥 + 𝑦 4 )𝑂2 → 𝑥𝐶𝑂2 + 𝐻2𝑂2 𝑦 (1) where x and y denote the number of carbons, c, and hydrogen, h, atoms which present in the fuel compositions. in this case, complete oxidation of carbon is assumed. typical fuels used in this heating equipment are light to heavy fuel oils, natural gas, and coal. in this work, methane is considered since it is widely being used in the current market due to more environmentally friendly feature compared to the others. in the combustion of fuels, the air is assumed to be in an excess amount to ensure complete combustion, hence no carbon monoxide is formed. co2 emissions are related to the amount of fuel burnt, qfuel (kw), in heating equipment calculated based on the following equation (2); [𝐶𝑂.]01233 = ' 56789 :;< )'=% ?@@ )𝛼 (2) where (α= 3.67) is the ratio of molar masses of co2 and nhv (kj/kg) represents the net heating value of fuel with a carbon content of c% (dimensionless) (gadalla, 2003). the amount of fuel burnt can be determined as; 𝑄cdef = 5ghij kghij (ℎnopq − 419) v6wxy vz v6wxy v[\]j^ (3) where λproc (kj/kg) and hproc (kj/kg) are the latent heat and enthalpy of steam delivered to the process, respectively tftb (oc) is the flame temperature of the boiler flue gases (smith,1991). citation: zin, r.m., abidin, m.a., and shahruddin, m.z. (2021) energy analysis and remixing effect of thermal coupling petlyuk column for natural gas liquid (ngl) fractionation train. int. journal of renewable energy development, 10(3), 459-469, doi: 10.14710/ijred.2021.33094 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 464 fig 3. rigorous simulation of conventional arrangement. fig 4. rigorous simulation of non-conventional arrangement with petlyuk column. int. journal of renewable energy development 10 (3) 2021: 459-469 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 465 4. results and discussion as the scope of the research focusses on the ngl fractionation trains, the results will be discussed based on the energy analysis, products specifications, remixing effects of the separation involving ethane, propane, butane, and pentane plus products as well as the co2 emission reduction analysis. 4.1 energy analysis and product specification from optimized conventional and non-conventional thermal coupled (petyluk) ngl fractionation arrangement. once the rigorous simulation converged, the conventional and non-conventional thermal coupled (petyluk) ngl fractionation trains were optimized. the optimized results for both configurations are summarized in table 4. based on the optimized results, the total heating duties and cooling duties for conventional columns arrangements are 22,869 kw and 114,686 kw respectively. hence, for this ngl application, 137,555 kw of total energy is required for the separation process. for the petlyuk arrangement, based on optimized results, the total heating duties and cooling duties are 19,936kw and 63,658 kw respectively. for this ngl application, about 131,475kw of total energy is required for the separation process. this amount is much lower than the conventional arrangement. this, in turn would lead to a significant saving in the operating cost (indicated by the energy utilization by condensers and reboilers). the percentage of cooling and heating duty saving is 44.49% and 12.83% respectively. based on the calculations, the overall saving in energy consumption associated with the petlyuk arrangement is 39.22%. in term of product specifications, comparisons of simulated results with the market requirement were performed. the optimized results of the product specification were tabulated in table 5. overall, the product specifications for each of the component in term of mole fraction content and reid vapor pressure (rvp) are achieved for all columns in both conventional and petlyuk arrangement configurations. based on table 5, the composition of methane and ethane products is similar for both arrangements. the composition of propane and butane products for both arrangements is slightly different, however still within the specification range. the purity of propane product for the conventional column is 0.9526 which slightly higher than petlyuk arrangement with corresponding purity of 0.9309. butane products also show a similar condition for both configurations. this is proven where the purity of butane products obtained for conventional and petlyuk arrangement is 0.9398 and 0.9319, respectively. for the rvp requirement, the maximum specification is 83 kpa. table 4 the optimized result of simulated conventional and non-conventional thermal coupling (petlyuk) columns arrangement. columns arrangement conventionala petlyuk columns cooling duty (kw) heating duty (kw) cooling duty (kw) heating duty (kw) deethanizer 6937 9186 n/a n/a depropanizer 65277 8451 8853 13520 debutanizer 42472 5232 54805 6416 sum duty (kw) 114686 22869 63658 19936 total duty: heating & cooling (kw) 137555 83594 *saving cooling duty (%) 44.49 *saving heating duty (%) 12.83 *overall saving: heating & cooling (%) 39.22 a results from zin et al. (2011) *saving: comparison between conventional and petlyuk column. table 5 product specification of simulated conventional arrangement and non-conventional thermally coupled petlyuk arrangement. specification product product methane ethane propane butane c5+ methane ethane propane butane c5+ conventional arrangement petlyuk arrangement methane (mol fraction) 0.8800 0 0.8800 0 ethane (mol fraction) 0.0385 0.9900 0.0060 0.0385 0.9900 0.0050 propane (mol fraction) 0.0099 0.9526 0.0199 0.0099 0.9309 0.0099 butanes (mol fraction) 0.04 0.9398 0.0600 0.9319 rvp (kpa) 61.46 56.83 product flowrate (kmol/hr) 19670 456.5 590.6 300.1 404.4 19670 456.2 602.9 295 393 *rvp (reid vapor pressure) citation: zin, r.m., abidin, m.a., and shahruddin, m.z. (2021) energy analysis and remixing effect of thermal coupling petlyuk column for natural gas liquid (ngl) fractionation train. int. journal of renewable energy development, 10(3), 459-469, doi: 10.14710/ijred.2021.33094 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 466 both arrangements show that this requirement are met which is 61.46 kpa and 56.83 kpa for conventional column and petlyuk column arrangement, respectively. even though there are some similarities and differences in term of product specifications, both simulated configurations show the correct level of design assurance as well as guarantees on the product quality. therefore, it can be concluded that the separation efficiency of petlyuk column arrangement is excellent as the product specifications for each of the component in mole fraction content and rvp are met as per market requirements. hence, it shows that the separation efficiency of the petlyuk arrangement is at par with the conventional column but with a much lower amount of energy being utilized to carry out the required separation. 4.2 remixing effect consider a feed mixture to deethanizer column in the conventional arrangement consisting of mainly ethane (a), propane (b), and butane (c) where a is the lightest and c the heaviest. in the deethanizer column, the concentration of b builds to a maximum at a tray near the bottom. on trays below this point, the amount of the heaviest component c continues to increase, diluting b so that its concentration profile now decreases on each additional tray toward the bottom of the column. energy has been used to separate b to a maximum purity. since b has not been removed at this point, it is remixed and diluted to the concentration at which it is removed in the bottoms. this remixing effect leads to a thermal inefficiency as shown in figure 5. fig 5. composition profile of the conventional arrangement showing remixing effect. fig 6. composition profile of the petlyuk arrangement showing no remixing effect occurring. int. journal of renewable energy development 10 (3) 2021: 459-469 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 467 the remixing effect occurrence is a source of inefficiency in the separation process. in comparison, considering the petlyuk arrangement (figure 6), a split is formed in the fractionator, leading to b being distributed between the top and bottom of the column separating ab from c and bc from a. hence, both sections remove only one component from the product of that column section, and this is also true for all sections of the main column. in this way, the remixing effect which is a feature of the conventional sequence is avoided. besides, it is known that losses occur in distillation operations due to mismatches between the composition of the column feed and the composition on the feed tray. as the fractionator distributes b between top and bottom, this allows greater freedom to match the feed composition with one of the trays in the column to reduce mixing losses at the feed tray. thus, the result of this study is consistent with the previous study by triantafyllou and smith (1992) which explained the occurrence of remixing effect and how it can be avoided using petlyuk column. 4.3 co2 emission reduction analysis table 6 summarizes the amount co2 emission by each of the equipment involved in both conventional distillation and non-conventional (petyluk) arrangement. figure 7 illustrates the amount of co2 emissions based on equipment (boiler, furnace, and gas turbine) involved in the fractionation train operation according to the column arrangement table 6 co2 emission by the boiler, furnace, and gas turbine involved in the fractionation train according to the column arrangement. equipment conventional arrangement non-conventional (petlyuk) arrangement co2 emission (x 107 kg/hr) boiler 4.23 2.99 furnace 18.9 13.4 gas turbine 4.25 3.00 fig 7. co2 emissions by boiler, furnace, and gas turbine for conventional and petyluk column arrangements from table 6 and figure 7, the petlyuk column arrangement exhibits the lowest co2 emissions for each type of equipment involved. on average the percentage of saving for reduction shown by each equipment is about 29.24 %. as the conventional arrangement consists more of distillation columns as compared petlyuk column arrangement, therefore more energy is required to carry out the desired duty. this resulted in more fuel consumption needed and produce a higher emission of co2 as shown in figure 7. due to the feature of thermal coupling of the petlyuk arrangement where the different columns are connected by liquid and vapour countercurrent streams which allow the remixing effect to be avoided, and consequently reducing the amount of heating and cooling energy required. this unique feature is the key reason for the co2 emission reduction by the petlyuk column arrangement 4.4 technology application to other industries. the use of non-conventional distillation column (petyluk) can be extended to other industries that deal with the separation of a multi-component mixture with tight control over purity and consistency. therefore, this technology can be applied in the production of biofuels, pharmaceutical industries, and alcoholic beverage production. 5. conclusion the energy analysis and remixing effects of natural gas liquids (ngl) fractionation process through the conventional and thermally coupled petlyuk arrangement have been successfully carried out. simulations were conducted out using hysys software. through this study, a framework for analyzing and improving the performance of (ngl) fractionation train was provided. synthesis of typical ngl fractionation train, steady-state simulations of conventional sequence and thermally coupled petlyuk arrangement using hysys, analysis on energy consumptions, product specification, and remixing effect have been investigated. the comparisons were made, and it was found that petlyuk arrangement saved about 44.49% and 12.83% in terms of cooling and heating duty, respectively. the overall annual energy consumption saving associated with the petlyuk arrangement is 39.22% as compared to the conventional arrangement. in the aspect of product specifications, petlyuk arrangement able to maintain the purity of the final products within the acceptable ranges. petlyuk arrangement proved to be an attractive alternative column as it can avoid the remixing effect occurrence which is a very important factor to be considered since remixing condition contributes to thermal and separation inefficiency. the unique feature of thermal coupling of petlyuk column permits significant co2 emission reduction due to the ability to avoiding remixing effect. this is one of the key advantages acquired by the petyluk column. from the results collected, it can be preliminary concluded that petlyuk arrangement model is better suited to be implemented in the ngl fractionation train. however, is recommended to have a comprehensive future study to validate the results obtained and to propose solutions to overcome the technological shortcomings. dynamic state of hysys modelling that mimic the real operating conditions and 0.0 2.0 4.0 6.0 8.0 10.0 12.0 14.0 16.0 18.0 20.0 boiler furnace gas turbine co 2 em is si on (x 1 07 k g/ hr ) conventional non-conventional (petlyuk) citation: zin, r.m., abidin, m.a., and shahruddin, m.z. (2021) energy analysis and remixing effect of thermal coupling petlyuk column for natural gas liquid (ngl) fractionation train. int. journal of renewable energy development, 10(3), 459-469, doi: 10.14710/ijred.2021.33094 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 468 the actual equipment sizing should be carried out to verify the practicality of the petlyuk arrangement technology implementation in natural gas fractionation train particularly on the feasibility of scale-up and economic assessment aspect. the current analysis of data in the study conducted can be utilized in the future to understand the process better. it also can become useful information to facilitate in designing and optimizing each step taken in the processing plant as well as the process equipment involved. acknowledgments the authors gratefully like to thank the ministry of higher education (mohe) and universiti teknologi mara (uitm) for grants of fundamental research grant scheme (frgs) 600-irmi/frgs 5/3 (053/2019) in providing financial support and for the facilities provided to conduct this study. references agrawal, r. (1996) synthesis of distillation column configurations for the multicomponent separation. ind. eng. chem. res. 35: 1059-1071; https://doi.org/10.1021/ie950323h agrawal, r. (2003) synthesis of multicomponent distillation column configurations. aiche j. 49(2), 379; https://doi.org/10.1002/aic.690490210 annakou, o., meszaros, a., fonyo, z. and mizsey, p. (1996) operability investigation of energy integrated distillation schemes. hungarian journal of industrial chemistry, 155160. biegler l. t., grossmann i. e, westerberg a.w., (1997). systematic methods of chemical process design. prentice hall. caballero, j. a & grossmann, i.e. (2004) design of distillation sequences: from conventional to fully thermally coupled distillation systems. comp. chem. eng. 28, 2307-2329. https://doi.org/10.1002/aic.690180510 chemmangattuvalappil n. & chong s. (2017) chapter 11 -basics of process simulation with aspen hysys. editor(s): dcy foo, n chemmangattuvalappil, dks ng, r elyas, cl chen, rd elms, hy lee, il chien, s chong, ch chong. chemical engineering process simulation, elsevier, 233-252 chowdhury n.b, hasan z., biplob a. h. m. (2011). hysys simulation of a sulfuric acid plant and optimization approach of annual profit. journal of science 179(2), 4. devold, h. (2009) oil and gas production handbook. an introduction to oil and gas handbook. abb as. second edition. dwivedi d, halvorsen ij, skogestad s. (2013) control structure selection for three-product petlyuk (dividing-wall) column. chemical engineering and processing: process intensification. 64, 57-67; https://doi.org/10.1016/j.cep.2012.11.006 egger t., hiller c., fieg g., (2018) experimental studies of a petlyuk column and validation of a non-equilibrium stage model. chemical engineering & technology 41, 827835.https://doi.org/10.1002/ceat.201700515 errico m., pirellas p., rong bg, segovia-hernandez jg. (2015) design and optimization of intensified quaternary petlyuk configuration. 12th international symposium on process systems engineering and 25th european symposium on computer aided process engineering. elsevier. 1368-1372. https://doi.org/10.1016/b978-0-444-63577-8.50073-5 essam bahnassi, abdul rahman khouri, alderton, p. and fleshman, j. (2005) achieving product specifications for ethane through to pentane plus from ngl fractionation plants. aiche fall conference fidkowski, t.f. (2006) distillation configurations and their energy requirements, aiche j. 52, 2098-2106; https://doi.org/10.1002/aic.10803 gadalla m, retrofit design of heat-integrated crude oil distillation systems, phd thesis, umist, manchester, uk, 2003. hernandez, s. & jimenez, a. (1999) controllability analysis of thermally coupled distillation systems. ind. eng. chem. res. 38: 3957-3963; https://doi.org/10.1021/ie060635s hoseinzadeh s., ghasemi m.h., heyns p. s. (2020a) application of hybrid systems in solution of low power generation at hot seasons for micro hydro systems. renewable energy 160, 323-332, https://doi.org/10.1016/j.renene.2020.06.149 hoseinzadeh s., yargholi r., kariman h., heyns p. s. (2020b) exergoeconomic analysis and optimization of reverse osmosis desalination integrated with geothermal energy. environmental progress & sustainable energy 39(5). https://doi.org/10.1002/ep.13405 hoseinzadeh s., zakeri m.h., shirkhani a., chamkha a.j. (2019) analysis of energy consumption improvements of a zeroenergy building in a humid mountainous area. journal of renewable sustainable energy 11, 015103. https://doi.org/10.1063/1.5046512 humphrey, j.l. and keller, g.e. (1997) separation process technology. mcgraw-hill, new york. kariman h., hoseinzadeh s., heyns p. s. (2019) energetic and exergetic analysis of evaporation desalination system integrated with mechanical vapor recompression circulation. case studies in thermal engineering 16,100548. https://doi.org/10.1016/j.csite.2019.100548 lucero-robles e.,gómez-castro f. i, ramírez-márquez c., segovia-hernández j.g., petlyuk columns in multicomponent distillation trains: effect of its location for the separation of hydrocarbon mixtures. chemical engineering & technology, 39(12), 22072216.https://doi.org/10.1002/ceat.201600152 manley, d. b. (1996) distillation of natural gas liquid. proceedings of the 1996 annual convention, gas processors association: 67-74. matla-gonzález d., urrea-garcía g., alvarez-ramirez j., bolaños-reynoso e., luna-solano g. (2013) simulation and control based on temperature measurements for petlyuk distillation columns. asia-pac. j. chem. eng. 8, 880-894. https://doi.org/10.1002/apj.1733 nath, r. & motard, r. l. (1981) evolutionary synthesis of separation processes. aiche j. 27(4), 578. https://doi.org/10.1002/aic.690270407 oyegoke t & dabai f. (2018) techno-economic feasibility study of bioethanol production from a combined cellulose and sugar feedstock in nigeria: 1-modeling, simulation and cost evaluation. nigerian journal of technology. 37, 913-920; https://doi.org/10.4314/njt.v37i4.8 petlyuk, f.b., platonov, v.m., slavinskii, d.m. (1965) thermodynamically optimal method for separating multicomponent mixtures. international chem. eng. 5, 555561. ramirez-corona n., jimenez-gutierrez a., castro-aquero a., rico-ramirez c. (2010) optimum design of petlyuk and divided-wall distillation systems using a shortcut model. chemical engineering research and design. 88, 1405-1418. https://doi.org/10.1016/j.cherd.2010.02.020 shahruddin, m.z., tan x., rahimi a.n., zubir m.a., islam zahran m.f., ibrahim k.a., abd hamid m.k., (2019) thermal pinch analysis application on distillation columns sequence of 5-component alcohol mixture, chemical engineering transactions, 72, 271-276; https://doi.org/10.3303/cet1972046 int. journal of renewable energy development 10 (3) 2021: 459-469 p a g e | ijred-issn: 2252-4940.copyright © 2021. the authors. published by cbiore 469 smith r, delaby o, 1991. targeting flue gas emissions. transactions of icheme november; part a (69):493–505. issn 0263-8762 strausa j. and skogestada s. (2016). minimizing the complexity of surrogate models for optimization. computer aided chemical engineering, 38, 289-294. https://doi.org/10.1016/b978-0-444-63428-3.50053-9 sultana st & ruhul amin m., (2011) aspen-hysys simulation of sulfuric acid plant. journal of chemical engineering. 26: 47-49; https://doi.org/10.3329/jce.v26i1.10182 thompson, r. w. & king, c. j. (1972) systematic synthesis of separation schemes. aiche j. 18(5): 941 triantafyllou, c. & smith, r. (1992) the design and optimization of fully thermally coupled distillation columns. chem. eng. res. des. 70, 118-132. trupti a., tyagee c., manali k., walker s. (2012) simulation of process equipment by using hysys. international journal of engineering research and applications. 41-44; https://doi.org/10.13140/rg.2.1.4186.9289 underwood t., erastova v., cubillas p., greenwell hc (2015) molecular dynamics simulations of montmorillonite-organic interactions under varying salinity: an insight into enhanced oil recovery. the journal of physical chemistry c., 119: 7282-7294; https://doi.org/10.1021/acs.jpcc.5b00555 underwood, a. j. v. (1946) fractional distillation of multicomponent mixtures calculation of minimum reflux ratio. j. inst. petrol: 32, 614. https://doi.org/10.1021/ie50480a044 zin r.m., salleh r., sazali r.a., kassim n.z., (2011) energy efficiency in natural gas processing plant via adoption of complex column (petlyuk column) for sustainable environment. 3rd international symposium & exhibition in sustainable energy & environment. 36-41. https://doi.org/10.1109/isesee.2011.5977105 © 2021. this article is an open access article distributed under the terms and conditions of the creative commons attributionsharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) international journal of renewable energy development int. j. renew. energy dev. 2023,12(2), 270-276 | 270 https://doi.org/10.14710/ijred.2023.48432 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id utilization of cassava peel (manihot utilissima) waste as an adhesive in the manufacture of coconut shell (cocos nucifera) charcoal briquettes bayu rudiyantoa* , intan rida agustinaa, zeni ulmaa, dafit ari prasetyoa, miftah hijiriawanb , bambang piluhartoc, totok prasetyod aenergy engineering laboratory, departement of renewable energy engineering, politeknik negeri jember, jl. mastrip 164 jember 68121, indonesia bgraduate program of mechanical engineering, universitas sebelas maret, jl. ir. sutami no.36 surakarta, 57126, indonesia cdepartment of chemistry, universitas jember, jl. kalimantan 37 kampus tegalboto, jember 68121, indonesia cdepartment of mechanical engineering, politeknik negeri semarang, jl. prof. h. soedarto s.h. semarang, 50275, indonesia abstract. coconut shells and waste cassava peels could be used as the main raw material for biomass briquettes for alternative energy sources in indonesia. this study aims to analyze the quality of briquettes based on a coconut shell and cassava peel adhesive through proximate analysis with three treatment ratio variations. the ratio of coconut shell to cassava peel used varied from v1 (75%:25%), v2 (70%:30%), and v3 (65%:35%). based on the result, the charcoal briquettes produced have a density of 0.61 gram/cm³-0.66 gram/cm³, water content of 5.51%-7.85%, ash content of 1.50%2.86%, combustion rate of 0.021 gram/s-0.026 gram/s, and the calorific value of 6,161 cal/gram-6,266 cal/gram. however, all the treatment variations appropriate the sni 01-6235-2000, the national standard of indonesia for the quality of charcoal briquette, which includes the calorific value (>5,000 cal/gram), moisture content (<8%), and ash content (<8%). briquettes with the best quality were generated by v1 with a density of 0.66 gram/cm³, water content of 5.51%, ash content of 1.50%, combustion rate of 0.026 gram/s, and calorific value of 6,266 cal/gram. furthermore, briquette material from the coconut shell waste with natural cassava peel adhesive can be feasible as an alternative fuel. keywords: biomass, briquettes, cassava peel waste, coconut shell, proximate analysis @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 20th august 2022; revised: 24th nov 2022; accepted: 2nd jan 2023; available online: 12th jan 2023 1. introduction the amount of energy needed has increased due to indonesia's population growth. in 2021, energy consumption from coal could reach 17% of the total national energy consumption mix (bppt, 2021). this certainly encourages the importance of using alternative and renewable energy sources. in this case, biomass is a renewable energy source that can be used as an alternative fuel to replace fossil fuels with abundant availability (budi surono, 2019; sunardi, djuanda, & mandra, 2019; tzelepi et al., 2020). biomass includes agricultural, plantation, forest waste, and organic components from industry and households (yana, nizar, irhamni, & mulyati, 2022). furthermore, the development of biomass as an alternative energy source has many challenges, and one of them is the production process (cuong et al., 2021; dani & wibawa, 2018; yana et al., 2022). however, the briquette is a biomass product that can be produced through a simple process with economic value, high heat content, and abundant availability of raw materials to compete with other fuels (sunardi et al., 2019). various types of waste can be used as raw materials to produce briquettes while solving the waste management * corresponding author email: bayu_rudianto@polije.ac.id (b. rudiyanto) problem (ardelean et al., 2022; bazhin, kuskov, & kuskova, 2019; ganesan & vedagiri, 2022; vaish, sharma, & kaur, 2022). coconut shell (cocos nucifera) is a waste product that can be utilized to produce charcoal briquettes. in this case, indonesia has quite extensive coconut plantations that can be used. this follows statistical data from the directorate general of plantation (2021) that the total area of coconut plantations is 3,401,893 ha, with a total production of 2,839,852 tons. besides, the coconut shell also contains a high calorific value reaching 7,283.5 cal/gram (nurhilal, suryaningsih, & indrana, 2018), and the coconut shell water content is only 10.03% (ghafar, halidi, & so’aib, 2020). however, in charcoal briquette production, natural adhesives are usually needed to support the quality of the briquettes. the addition of adhesive is meant to reduce the briquette’s pores and give them a solid structure, permitting them to be shipped and stored without being easily destroyed (jiang et al., 2022; kamunur, ketegenov, kalugin, karagulanova, & zhaksibaev, 2022). in addition, the coconut shell charcoal's fine grains are combined with the adhesive substance to be molded as required. research article https://doi.org/10.14710/ijred.2023.48432 https://doi.org/10.14710/ijred.2023.48432 mailto:bayu_rudianto@polije.ac.id https://orcid.org/0000-0002-4708-629x https://orcid.org/0000-0002-4667-2908 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.48432%26domain=pdf b. rudiyanto et al int. j. renew. energy dev 2023, 12(2), 270-276 |271 issn: 2252-4940/© 2023. the author(s). published by cbiore in this case, various adhesive materials in the manufacture of briquettes have been developed (helwani et al., 2020; maulina, sarah, misran, & anita, 2021; suryaningsih, resitasari, & nurhilal, 2019). cassava peel (manihot utilissima) is one of the materials that can be utilized as an alternative due to its availability to assist the production of coconut shell briquettes as a biomass raw material. it can be seen that the production of cassava plants in indonesia can reach 19,053,748 tons (ministry of agriculture, 2021). cassava peel has the potential to be used as an adhesive in the production of briquettes due to its moisture content of 9.93-11.46%, volatile materials of 77.9381.93%, ash content of 1.93-4.36%, fixed carbon content of 13.44-15.51%, lignin content of 6.5-16.0%, cellulose content of 5.5-14.5%, hemicellulose content of 41.0-56.0%, and calorific value of 3,843.84 cal/gram (hirniah, 2020; kayiwa, kasedde, lubwama, & kirabira, 2021a, 2021b). furthermore, cassava peel has a carbohydrate content of about 30.15% that can be used as an adhesive (anggraeni, girsang, nandiyanto, & bilad, 2021; kariuki, muthengia, erastus, leonard, & marangu, 2020). proximate testing is needed to determine the quality of briquettes based on sni. however, due to its ready-to-use product characteristic, proximate testing is required to determine the ability of briquettes as fuel. in charcoal briquette production, it is necessary to consider the value of water content, volatile matters, ash, solid carbon (fixed carbon), and calorific value as the main parameters of the quality of briquettes. the water content indicates the ease of burning, and briquettes are easier to mold when the water content is high. volatile matter, ash, and solid carbon as total fixed carbon refer to the amount of smoke when the briquettes are burned (srisang et al., 2022). besides, the calorific value represents the energy produced from briquettes and the ease of burning (adeleke, odusote, ikubanni, olabisi, & nzerem, 2022; guo et al., 2020; velusamy, subbaiyan, kandasamy, shanmugamoorthi, & thirumoorthy, 2022). the novelty of this research is the composition of raw materials and adhesives for the production of the briquettes. although coconut shell has been commercialized as a raw briquette material, tapioca flour is still used as an adhesive. however, cassava peel is an excellent adhesive material because it has a starch content above 30%. therefore, this research aims to determine the concentration level between coconut shell biomass and cassava peel natural adhesive according to the five aspects based on the sni 01-6235-2000 in indonesia. this is expected to produce a suitable correlation to obtain the development of charcoal briquettes better. as a result, the production of charcoal briquettes as an alternative fuel with high economic value, wide availability, and simplicity of access, can serve in the development of new and ecologically friendly energy sources. 2. method 2.1 development of coconut shell (cocos nucifera) charcoal briquettes material in this research, coconut shell waste is used as raw material for briquettes production, and cassava peel waste is used as an adhesive. the chemical and physical properties of the coconut shell and cassava peel is shown in table 1 and 2. coconut shell as the raw material that has been dried is then pyrolyzed using a furnace at a temperature of 300℃ for 7 hours (rizal et al., 2020; sarkar & wang, 2020; tu et al., 2021). during the pyrolysis process, the raw material of coconut shells is charred evenly. the result of the coconut shell that has been charcoaled is then pounded. table 1 chemical and physical properties of coconut shell (kabir ahmad et al., 2022) parameters properties description proximate analysis moisture content 5.56% volatile matter 70.82% fixed carbon 21.80% ash 1.80% ultimate analysis c 40.08% h 5.22% n 0.22% s 0.17% o 54.31% potential as energy source porosity 24.39% compressibility index 40.24% calorific value 19.4 mj/kg fuel value index 4441 table 2 chemical and physical properties of cassava peel (kayiwa et al., 2021a) no. properties description 1 moisture content 9.77-11.50% 2 volatile matter 78.22-82.31% 3 fixed carbon 13.44-15.51% 4 ash content 1.85-4.40% 5 lignin 6.5-16.0% 6 cellulose 5.5-14.5% 7 hemicellulose 41.0-56.0% then, the coconut shell is sieved using a 40-mesh which aims to produce a fine, uniform particle size, and suitable as a briquette material (abyaz, afra, & saraeyan, 2020; meytij, santoso, rampe, tiwow, & apita, 2021; setter, sanchez costa, pires de oliveira, & farinassi mendes, 2020). the production of cassava peel adhesive begins with cleaning the attached peel dirt. then, the cassava peel is dried and mashed using grinding. when the cassava peel has been processed into flour, it is filtered, combined with hot water in a 1:2 ratio, and stirred thoroughly to remove lumps. the purpose of the hot water addition is to make the mixing process easier. variations in the mixture of briquette raw materials were carried out using coconut shell charcoal which had been mashed using adhesive homogeneously with a predetermined composition, namely variation 1 (v1), variation 2 (v2), and variation 3 (v3), as shown in table 3. furthermore, the finished raw material mixture is placed into the briquette mold in the shape of a cylinder with a material weight of 30 grams. the briquette mixture was flattened to a height of 5.7 cm, then pressed 60% to produce briquettes with a height of 2.3 cm. the briquettes harden during the one-minute pressure hold. the drying process was then continued by heating for 4 hours at 105°c in an oven. the briquettes were consequently stored at room temperature for 24 hours. table 3 composition variations of coconut shell charcoal briquettes variation name briquettes material composition coconut shell charcoal cassava peel adhesive v1 75% (22.5 grams) 25% (7.5 grams) v2 70% (21 grams) 30% (9 grams) v3 65% (19.5 grams) 35% (10.5 grams) b. rudiyanto et al int. j. renew. energy dev 2023, 12(2), 270-276 |272 issn: 2252-4940/© 2023. the author(s). published by cbiore table 4 specifications of instruments used in the study instrument specification wire mesh gb/t6003.1-2012 40 mesh heater unb 400 (oven) 230 vac; 6.1 a; 50/60 hz furnace carbolite elf 11/6b 230 vac; 9.6 a: 2000 watt; max temp 1100℃ m20 universal mill (grinder) 230/115 ±10% vac; 50/60 hz; 550 watt; 20,000 rpm; 250 ml ika © 2000 bomb calorimeter 230/115 vac; 50/60 hz; 1.8 kw; measurement range 40,000 j 2.2 experimental and testing instruments in the manufacturing and analysis performed, this research uses several types of equipment, such as a 40-mesh sieve, mortar, briquette press, heater (oven), grinding, baking sheet, mixing tank, pan, cup, analytical balance, pyrolysis equipment, ika © 2000 bomb calorimeter, stopwatch, and caliper. further details for the instrument used in this study are shown in table 4. 2.3 quality test of briquettes the quality testing of coconut shell charcoal briquettes included density, moisture content, ash, combustion rate, and calorific value. density can be examined by measuring the mass of briquettes and the volume of briquette samples using equation (1): 𝜌 = 𝑚 𝑣 (1) whereas 𝜌 (g/cm3) is density, 𝑚 (g) is the mass of briquettes, and 𝑣 (cm3) is the volume of the briquettes. moisture content can be tested by weighing the sample to determine the initial weight and then heated in an oven at 105℃ for 6 hours. the sample was weighed again to decide its final weight after being dried in the oven for an hour. the water content can be calculated using equation (2): 𝑀𝐶 = 𝑋1−𝑋2 𝑋1 100% (2) where 𝑀𝐶 is moisture content, 𝑋1 (g) is the initial weight of the sample, and 𝑋2 (g) is the final weight of the sample. ash content is the residue from burning briquettes that are not completely burned. the ash content test was carried out by weighing the empty weight of the cup, then 1 gram of the sample in the cup was heated in the furnace gradually at a temperature of 450-950℃ for 1-2 hours and then allowed to stand at room temperature until the temperature was normal. equation (3) can calculate ash content as follow: 𝐴𝐶 = 𝐵−𝐴 𝐶−𝐴 𝑥 100% (3) where 𝐴𝐶 is ash content, 𝐴 is the weight of an empty cup, 𝐵 is the weight of the cup and ash, and 𝐶 is the weight of the cup and the sample. the rate of burning of briquettes is determined by the weight of the briquettes burned over a certain period using equation (4): 𝑉 = 𝑚𝑡 𝑡 (4) where 𝑉 (g/s) is the rate of burning of briquettes, 𝑚𝑡 (g) is the mass of the burned briquettes, and 𝑡 (second) is the required burning time. fig. 1 briquette preparation schematic diagram the heat produced by briquettes and oxygen at a fixed volume can be evaluated using a bomb calorimeter to determine the calorific value. figure 1 shows the method used in this study to produce coconut shell charcoal briquettes using waste cassava peel as an adhesive. 2.4 data analysis in this study, we performed a quantitative analysis of density, moisture content, ash content, combustion rate, and calorific value of the coconut shell charcoal briquettes production using adhesive from waste cassava peel in each variation in the ratio of material composition. the analysis was carried out to determine whether the values of the various parameters complied with the indonesian national standard (sni) 01-62352000. moreover, a one-way analysis of variance (anova) test was conducted to investigate whether variations in the material composition used to create adhesive from waste cassava peel during the production of coconut shell charcoal briquettes affected each of the parameters analyzed in this study. furthermore, posthoc analysis using the tukey method was carried out to determine the significant differences between each variation (aransiola, oyewusi, osunbitan, & ogunjimi, 2019; karimibavani, sengul, & asmatulu, 2020; niño, arzola, & araque, 2020). 3. results and discussion 3.1 density the briquette density test was carried out using the ratio of mass and volume. the homogeneity and size of the charcoal are affected by the density the briquettes produce. the results of the density measurement of charcoal briquettes v1, v2, and v3 are presented in figure 2. fig. 2 density of coconut shell briquettes with cassava peel adhesive 0.66 0.63 0.61 0.54 0.56 0.58 0.60 0.62 0.64 0.66 0.68 0.70 v1 (75%:25%) v2 (70%:30%) v3 (65%:35%) d e n s it y ( g r a m /c m ³) b. rudiyanto et al int. j. renew. energy dev 2023, 12(2), 270-276 |273 issn: 2252-4940/© 2023. the author(s). published by cbiore table 5 density analysis using the tukey method with 95% confidence variation n mean grouping v1 3 0.66456 a v2 3 0.63108 b v3 3 0.61050 c based on figure 2, it can be seen that the highest density value is in v1 of 0.66 g/cm3 with a ratio of coconut shell to cassava peel adhesive of 75%:25%, while the lowest value is in v3 with a ratio of coconut shell to cassava peel adhesive of 65%:35%. however, the values of the three densities are not much different, but the treatment value in v1 shows the results of better briquette density compared to other variations. this is due to the amount of adhesive that meets the void ratio formed by the particle size of 40 mesh. high pressure can also increase the density value. it follows the research conducted by sunardi et al. (2019) about the characteristics of corncob briquettes with a pressure of 44.80 kg/cm3 and a particle size of 60 mesh, which produces a higher density level than corncob briquettes using a pressure of 22.42 kg/cm3 with a particle size of 40 mesh. consequently, the adhesive will tend to fill the surface of the charcoal as the bonds between the molecules of the charcoal become stronger, reducing the cavity filled with water or air (satya, raju, praveena, & jyothi, 2014). therefore, the higher the density value of the briquettes, the smaller the cavity and the rate of combustion is slower (haryanti, wardhana, & suryajaya, 2020). furthermore, in statistical analysis using the one-way anova method to determine the effect of variations in the composition of coconut shell charcoal and cassava peel waste, it is known that the p-value is <0.05, representing that the composition affects the density value of the briquettes. in the post hoc tukey analysis, it is also known that each variable v1, v2, and v3 is significantly different from each other, as shown in table 5. 3.2 moisture content briquettes have hygroscopic properties or easily absorb water, which shows that the value of water content needs to be considered because it can affect the quality of the briquettes produced. in this case, the moisture content of coconut shell briquettes with cassava peel adhesive ranged from 5.51-7.85%, as presented in figure 3. figure 3 shows that the highest water content was obtained in treatment v3 at 7.85%, while the lowest water content was found in v1 at 5.51%. treatment v1 with a ratio of coconut shell to cassava peel adhesive of 75%:25% had better briquette quality than other variations. fig. 3 moisture content of coconut shell briquettes with cassava peel adhesive table 6 moisture content analysis using the tukey method with 95% confidence variation n mean grouping v3 3 7.8525 a v2 3 6.503 b v1 3 5.511 c this is due to the low water content, and the cassava peel adhesive that blends with coconut shell charcoal will be tighter because its pores become smaller. the high and low water content produced can be influenced by the type and percentage of adhesive used to manufacture briquettes (kong, loh, bachmann, rahim, & salimon, 2014). the addition of more adhesive causes the water contained in the adhesive to enter the pores of the charcoal (permatasari & utami, 2015) based on figure 3, it can be seen that the smaller the percentage of adhesive used, the smaller the water content, which means the quality of the briquettes produced will be better. this is in line with the research by maryono et al. (2013) about the quality of coconut shell charcoal briquettes with the addition of higher levels of starch adhesive will produce higher water content as well. the maximum moisture content of charcoal briquettes is 8%, according to sni 01-6235-2000. in this case, the water content in each treatment has met the sni standard because it is below 8%, indicating that the coconut shell briquettes with cassava peel adhesive are suitable for alternative fuels. however, the p-value of the one-way anova test is <0.05, which indicates that the variation in the composition of coconut shell charcoal with cassava peel waste affects the value of the resulting water content. moreover, the post hoc tukey's analysis results show that each variation v1, v2, and v3 is significantly different, as shown in table 6. 3.3 ash content ash content is one of the references to determine the quality of briquettes. ash content can affect the calorific value and carbon. the ash content produced in this study ranged from 1.50-2.86%. the results of the ash content test are presented in figure 4. fig. 4 ash content of coconut shell briquettes with cassava peel adhesive table 7 ash content analysis using the tukey method with 95% confidence variation n mean grouping v3 3 2.8579 a v2 3 2.6569 a v1 3 1.501 b 5.51 6.50 7.85 0.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 v1 (75%:25%) v2 (70%:30%) v3 (65%:35%) m o is tu r e c o n te n t (% ) 1.50 2.66 2.86 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 v1 (75%:25%) v2 (70%:30%) v3 (65%:35%) a s h c o n te n t (% ) b. rudiyanto et al int. j. renew. energy dev 2023, 12(2), 270-276 |274 issn: 2252-4940/© 2023. the author(s). published by cbiore v3 produced the highest ash content with a ratio of coconut shell to cassava peel adhesive of 65%:35%. in contrast, the lowest ash content was obtained in v1 at 1.50%, with a ratio of coconut shell to cassava peel adhesive of 75%:25%. the amount of adhesive applied can influence the high and low levels of ash produced (hasan et al., 2017; modolo et al., 2015). ash content affects the heating value and carbon content. the lower the ash content, the higher the calorific value and the fixed carbon content in the briquettes (lu et al., 2019; román gómez, cabanzo hernández, guerrero, & mejía-ospino, 2018; todaro, rita, cetera, & d’auria, 2015). in addition, the content of inorganic materials in adhesives, such as silica (sio2), mgo, fe2o3, a1f3, mgd3, and fe, can also increase the ash content of briquettes (haryanti et al., 2020). based on these findings, it can be shown that the ash content increases as the adhesive content increases. this is similar to maryono et al. (2013), where the ash content of coconut shell briquettes increased with cassava peel adhesive applied. the higher ash content in briquettes can reduce the calorific value and combustion rate, preventing air voids from penetrating the furnace (sunardi et al., 2019). the maximum permissible ash content in sni 01.6235.2000 is 8%, while the ash content produced in this study ranges from 1.50-2.86%. it shows that the briquettes produced had good quality. the composition of the comparison of coconut shell with cassava peel adhesive is best produced by v1. it has the lowest ash content compared to other variations. the one-way anova analysis obtained a pvalue <0.05, it can be seen that the composition of coconut shell charcoal and cassava peel waste affects the ash content results. furthermore, in the post hoc tukey analysis, it is known that the variations v2 and v3 are not significantly different from each other, while the variations of v1 are significantly different from each other with v2 and v3, as shown in table 7. 3.4 combustion rate five briquettes were used to heat 700 ml of water using three iterations of each variation in the combustion rate test to measure the rate of briquette combustion starting at the speed of the briquette flame. the calculation of the briquette burning rate result in this study ranged from 0.021-0.026 gram/s, as shown in figure 5. fig. 5 combustion rate of coconut shell briquettes with cassava peel adhesive table 8 combustion rate analysis using the tukey method with 95% confidence variation n mean grouping v1 3 0.071000 a v2 3 0.063000 b v3 3 0.058000 c the fastest burning rate is produced by v1 at 0.026 gram/s, while v3 has the slowest burning rate at 0.021 gram/s. figure 5 shows that the percentage ratio of the adhesive composition can affect the rate of combustion produced. this is in line with syarief et al. (2021), that the higher the percentage of adhesive added, the slower the burning rate, and vice versa. the high percentage of adhesive addition will make the granules on the briquettes stick firmly. it makes the briquette pores smaller and difficult for air to enter to speed up the combustion process. comparison of the composition of the variations of the resulting material did not differ much, but the v1 showed better briquette results than other variations. this is due to the faster rate of combustion, which makes it easier for the briquettes to ignite and burn away without producing a lot of smoke. the v1 shows a more effective and efficient result to be used as an alternative fuel. the results from the one-way anova analysis obtained a p-value <0.05. this indicates that the composition of coconut shell charcoal and cassava peel waste affects the rate of combustion that occurs in briquettes. based on the results of post hoc tukey analysis, it is known that the respective variations of v1, v2, and v3 are significantly different from each other, as shown in table 8. 3.5 calorific value the calorific value is the main parameter in determining the quality of briquettes. the calorific value produced in this study ranged from 6,161 to 6,266 cal/gram. the results of the heat test using the ika © 2000 bomb calorimeter are shown in figure 6. figure 6 shows that the highest calorific value produced by v1 is 6,266 cal/gram, while v3 of 6,161 cal/gram has the lowest calorific value. the higher the calorific value, the better the quality of the briquettes (haryanti et al., 2020). the calorific value is related to the amount of water and ash in the briquettes. the percentage of adhesive given influences the amount of water and ash produced. the higher the adhesive added, the higher the water and ash produced. thus, the calorific value created is low and vice versa (sulistyaningkarti and utami, 2017). in this case, the results of the one-way anova analysis show the p-value >0.05. it can be seen that the composition of coconut shell charcoal and cassava peel waste does not affect the resulting calorific value. based on these results, a post hoc tukey analysis is not required. the minimum standard calorific value of briquettes, according to sni 01-6235-2000, is 5,000 cal/gram. however, the calorific value of briquettes in v1, v2, and v3, as shown in figure 6, they have a value of over 5,000 cal/gram. the highest calorific value was shown by v1 of 6,266 cal/gram with the coconut shell and cassava peel adhesive ratio at 75%:25%. this is influenced by the value of water content and ash content. moreover, the briquettes produced by v1 offer better quality than other variations. fig. 6 calorific value of coconut shell briquettes with cassava peel adhesive 0.071 0.063 0.058 0.000 0.010 0.020 0.030 0.040 0.050 0.060 0.070 0.080 v1 (75%:25%) v2 (70%:30%) v3 (65%:35%) c o m b u s ti o n r a te (g r a m /s ) 6266 6234 6161 6000 6050 6100 6150 6200 6250 6300 6350 v1 (75%:25%) v2 (70%:30%) v3 (65%:35%) c a lo r if ic v a lu e (c a l/ g r a m ) b. rudiyanto et al int. j. renew. energy dev 2023, 12(2), 270-276 |275 issn: 2252-4940/© 2023. the author(s). published by cbiore in this study, v1, with a composition of 75% coconut shell and 25% cassava peel adhesive, is the best composition in terms of 4 parameters: density, moisture content, ash content, and calorific value. this is as a result that applying too much adhesive can reduce the briquettes' quality. therefore the addition of adhesive must be carried out appropriately (saputra at. al, 2021). however, the best result for the combustion rate is sample v2, with a composition of 65%:35%, but this is not very influential because the difference in the combustion rate between v1 and v2 has a slight difference. 4. conclusion based on the research results, it can be seen that the use of coconut shells as raw material for briquettes has an excellent ability to become a renewable energy source in the form of biomass. in this case, the percentage variation of adhesive material such as cassava peel can produce characteristics as fuel for alternative energy sources. it can be seen that the v1 treatment with a ratio of coconut shell with cassava peel adhesive of 75%:25% can produce charcoal briquettes that have better quality than other variations, with a density value of 0.66 gram/cm3, water content of 5.51%, ash content of 1.50%, combustion rate of 0.026 gram/s and calorific value of 6,266 cal/gram. the one-way anova analysis shows that the composition of coconut shell charcoal and cassava peel waste affects the resulting density, moisture content, ash content, and burning rate. in this case, the heating value is not affected by variations in the composition of the raw materials. however, charcoal briquettes from coconut shell waste and natural adhesives from cassava peel waste are feasible to be used as alternative fuels because of their economic value, easy to obtain, abundantly available, and have complied with sni 01-62352000. besides, further identification of the starch content in cassava peel, volatile matter content and carbon content in briquettes is required to improve research findings and develop solutions using alternative energy sources with higher quality and more environmentally friendly. this is because the natural adhesive content can reduce the calorific value of briquettes. the addition of adhesive is carried out without a carbonization process, and it is necessary to analyze the value of volatile matter and fixed carbon. considering their ability to marge, it is important to know the volatile and fixed carbon content. moreover, if the volatile content is too high and the fixed carbon is too low, it will significantly affect the decrease in the heating value of the briquettes. author contributions: br; supervision, resources, project administration, ira; conceptualization, original draft, zu; methodology, dap; formal analysis, mh; writing—review and editing, project administration, bp; supervision, validation, tp: supervision, validation —. all authors have read and agreed to the published version of the manuscript. conflicts of interest: the authors declare no conflict of interest. references abyaz, a., afra, e., & saraeyan, a. (2020). improving technical parameters of biofuel briquettes using cellulosic binders. energy sources, part a: recovery, utilization and environmental effects, 00(00), 1–12. https://doi.org/10.1080/15567036.2020.1806955 adeleke, a. a., odusote, j. k., ikubanni, p. p., olabisi, a. s., & nzerem, p. (2022). briquetting of subbituminous coal and torrefied biomass using bentonite as inorganic binder. scientific reports, 12(1), 1–11. https://doi.org/10.1038/s41598-022-12685-5 anggraeni, s., girsang, g. c. s., nandiyanto, a. b. d., & bilad, m. r. (2021). effects of particle size and composition of sawdust/carbon from rice husk on the briquette performance. journal of engineering science and technology, 16(3), 2298–2311. aransiola, e. f., oyewusi, t. f., osunbitan, j. a., & ogunjimi, l. a. o. (2019). effect of binder type, binder concentration and compacting pressure on some physical properties of carbonized corncob briquette. energy reports, 5, 909–918. https://doi.org/10.1016/j.egyr.2019.07.011 ardelean, e., socalici, a., lupu, o., bistrian, d., dobrescu, c., & constantin, n. (2022). recovery of waste with a high iron content in the context of the circular economy. materials, 15(14), 1–18. https://doi.org/10.3390/ma15144995 bazhin, v. y., kuskov, v. b., & kuskova, y. v. (2019). processing of lowdemand coal and other carbon-containing materials for energy production purposes. inzynieria mineralna, 2019(1), 195–198. https://doi.org/10.29227/im-2019-01-37 bppt. (2021). indonesia energy outlook 2021: perspective of indonesian energy technology solar power for charging station energy supply. jakarta. budi surono, u. (2019). biomass utilization of some agricultural wastes as alternative fuel in indonesia. journal of physics: conference series, 1175(1). https://doi.org/10.1088/17426596/1175/1/012271 cuong, t. t., le, h. a., khai, n. m., hung, p. a., linh, l. t., thanh, n. v., … huan, n. x. (2021). renewable energy from biomass surplus resource: potential of power generation from rice straw in vietnam. scientific reports, 11(1), 1–10. https://doi.org/10.1038/s41598-020-80678-3 dani, s., & wibawa, a. (2018). challanges and policy for biomass energy in indonesia. international journal of business, economic and law, 15(5), 41047. https://www.ijbel.com/wpcontent/uploads/2018/04/ijbel15_212.pdf directorate general of plantation. (2021). national leading plantation statistics 2019-2021 (d. gartina & r. l. l. sukriya, eds.). jakarta: sekterariat direktorat jenderal perkebunan. ganesan, s., & vedagiri, p. (2022). production of sustainable biomass briquettes from de-oiled cashewnut shell. materials today: proceedings. https://doi.org/10.1016/j.matpr.2022.09.179 ghafar, h., halidi, s. n. a. m., & so’aib, m. s. (2020). coconut shell: thermogravimetric analysis and gross calorific value. proceedings of mechanical engineering research day, 206–207. https://www3.utem.edu.my/care/proceedings/merd20/pdf/0 6_energy_engineering_and_management/089-p206_207.pdf guo, z., wu, j., zhang, y., wang, f., guo, y., chen, k., & liu, h. (2020). characteristics of biomass charcoal briquettes and pollutant emission reduction for sulfur and nitrogen during combustion. fuel, 272(april), 117632. https://doi.org/10.1016/j.fuel.2020.117632 haryanti, n. h., wardhana, h., & suryajaya. (2020). effect of pressure on alaban charcoal briquettes small particle size. jurnal risalah fisika, 4(1), 19–26. https://doi.org/https://doi.org/10.35895/rf.v4i1.170 hasan, e. s., jahiding, m., mashuni, ilmawati, w. o. s., wati, w., & sudiana, i. n. (2017). proximate and the calorific value analysis of brown coal for high-calorie hybrid briquette application. journal of physics: conference series, 846(1). https://doi.org/10.1088/1742-6596/846/1/012022 helwani, z., ramli, m., rusyana, a., marlina, m., fatra, w., idroes, g. m., … idroes, r. (2020). alternative briquette material made from palm stem biomass mediated by glycerol crude of biodiesel byproducts as a natural adhesive. processes, 8(7). https://doi.org/10.3390/pr8070777 hirniah, f. e. (2020). energy analysis in making charcoal briquettes from cassava peel with tapioca flour as adhesive. universitas jember, jember. jiang, x., wu, c., zhou, h., gao, b., fang, x., han, j., & gao, w. (2022). relationship between thermal properties and structure, composition of briquette through grey relational analysis. journal of applied geophysics, 206(november 2021), 104786. https://doi.org/10.1016/j.jappgeo.2022.104786 kabir ahmad, r., anwar sulaiman, s., yusup, s., sham dol, s., inayat, m., & aminu umar, h. (2022). exploring the potential of coconut shell biomass for charcoal production. ain shams engineering journal, 13(1), 101499. https://doi.org/10.1016/j.asej.2021.05.013 kamunur, k., ketegenov, t., kalugin, s., karagulanova, a., & zhaksibaev, m. (2022). the role of the alkaline promoter on the formation of strength and burning of coal briquettes. south african journal of chemical engineering, 42(may), 156–161. https://doi.org/10.1080/15567036.2020.1806955 https://doi.org/10.1038/s41598-022-12685-5 https://doi.org/10.1016/j.egyr.2019.07.011 https://doi.org/10.3390/ma15144995 https://doi.org/10.29227/im-2019-01-37 https://doi.org/10.1088/1742-6596/1175/1/012271 https://doi.org/10.1088/1742-6596/1175/1/012271 https://doi.org/10.1038/s41598-020-80678-3 https://www.ijbel.com/wp-content/uploads/2018/04/ijbel15_212.pdf https://www.ijbel.com/wp-content/uploads/2018/04/ijbel15_212.pdf https://doi.org/10.1016/j.matpr.2022.09.179 https://www3.utem.edu.my/care/proceedings/merd20/pdf/06_energy_engineering_and_management/089-p206_207.pdf https://www3.utem.edu.my/care/proceedings/merd20/pdf/06_energy_engineering_and_management/089-p206_207.pdf https://doi.org/10.1016/j.fuel.2020.117632 https://doi.org/https:/doi.org/10.35895/rf.v4i1.170 https://doi.org/10.1088/1742-6596/846/1/012022 https://doi.org/10.3390/pr8070777 https://doi.org/10.1016/j.jappgeo.2022.104786 https://doi.org/10.1016/j.asej.2021.05.013 b. rudiyanto et al int. j. renew. energy dev 2023, 12(2), 270-276 |276 issn: 2252-4940/© 2023. the author(s). published by cbiore https://doi.org/10.1016/j.sajce.2022.08.009 karimibavani, b., sengul, a. b., & asmatulu, e. (2020). converting briquettes of orange and banana peels into carbonaceous materials for activated sustainable carbon and fuel sources. energy, ecology and environment, 5(3), 161–170. https://doi.org/10.1007/s40974-020-00148-4 kariuki, s. w., muthengia, j. w., erastus, m. k., leonard, g. m., & marangu, j. m. (2020). characterization of composite material from the copolymerized polyphenolic matrix with treated cassava peels starch. heliyon, 6(7), e04574. https://doi.org/10.1016/j.heliyon.2020.e04574 kayiwa, r., kasedde, h., lubwama, m., & kirabira, j. b. (2021a). characterization and pre-leaching effect on the peels of predominant cassava varieties in uganda for production of activated carbon. current research in green and sustainable chemistry, 4(february), 100083. https://doi.org/10.1016/j.crgsc.2021.100083 kayiwa, r., kasedde, h., lubwama, m., & kirabira, j. b. (2021b). the potential for commercial scale production and application of activated carbon from cassava peels in africa: a review (elsevier ltd; vol. 15). elsevier ltd. https://doi.org/10.1016/j.biteb.2021.100772 kong, s. h., loh, s. k., bachmann, r. t., rahim, s. a., & salimon, j. (2014). biochar from oil palm biomass: a review of its potential and challenges. journal renewable and sustainable energy reviews, 39, 729–739. https://doi.org/10.1016/j.rser.2014.07.107 lu, z., chen, x., yao, s., qin, h., zhang, l., yao, x., … lu, j. (2019). feasibility study of gross calorific value, carbon content, volatile matter content and ash content of solid biomass fuel using laserinduced breakdown spectroscopy. fuel, 258(september), 116150. https://doi.org/10.1016/j.fuel.2019.116150 maryono, sudding, & rahmawati. (2013). preparation and quality analysis of coconut shell charcoal briquette observed by starch concentration. journal chemical, 14(1), 74–83. maulina, s., sarah, m., misran, e., & anita, m. f. (2021). the correlation of ultimate analysis and calorific value on palm oil briquettes using durian seed adhesives. iop conference series: materials science and engineering, 1122(1), 012079. https://doi.org/10.1088/1757-899x/1122/1/012079 meytij, j. r., santoso, i. r. s., rampe, h. l., tiwow, v. a., & apita, a. (2021). infrared spectra patterns of coconut shell charcoal as result of pyrolysis and acid activation origin of sulawesi, indonesia. e3s web of conferences, 328, 08008. https://doi.org/10.1051/e3sconf/202132808008 ministry of agriculture. (2021). agricultural statistics 2021 (a. a. susanti & m. a. supriyatna, eds.). jakarta: center for agricultural data and information systems, ministry of agriculture, republic of indonesia. modolo, r. c. e., silva, t., senff, l., tarelho, l. a. c., labrincha, j. a., ferreira, v. m., & silva, l. (2015). bottom ash from biomass combustion in bfb and its use in adhesive-mortars. fuel processing technology, 129, 192–202. https://doi.org/10.1016/j.fuproc.2014.09.015 niño, a., arzola, n., & araque, o. (2020). experimental study on the mechanical properties of biomass briquettes from a mixture of rice husk and pine sawdust. energies, 13(5). https://doi.org/10.3390/en13051060 nurhilal, o., suryaningsih, s., & indrana, i. (2018). study of thermal efficiency of biomass carbonizing by direct method. journal of physics: conference series, 1080. https://doi.org/10.1088/17426596/1080/1/012024 rizal, w. a., nisa, k., maryana, r., prasetyo, d. j., pratiwi, d., jatmiko, t. h., … suwanto, a. (2020). chemical composition of liquid smoke from coconut shell waste produced by sme in rongkop gunungkidul. iop conference series: earth and environmental science, 462(1). https://doi.org/10.1088/17551315/462/1/012057 román gómez, y., cabanzo hernández, r., guerrero, j. e., & mejíaospino, e. (2018). ftir-pas coupled to partial least squares for prediction of ash content, volatile matter, fixed carbon and calorific value of coal. fuel, 226(april), 536–544. https://doi.org/10.1016/j.fuel.2018.04.040 sarkar, j. k., & wang, q. (2020). different pyrolysis process conditions of south asian waste coconut shell and characterization of gas, bio-char, and bio-oil. energies. satya, m., raju, c. a. i., praveena, u., & jyothi, k. r. (2014). studies on development of fuel briquettes using locally available waste. journal of engineering research and applications, 4(3), 553–559. https://www.ijera.com/papers/vol4_issue3/version%201/ct 4301553559.pdf setter, c., sanchez costa, k. l., pires de oliveira, t. j., & farinassi mendes, r. (2020). the effects of kraft lignin on the physicomechanical quality of briquettes produced with sugarcane bagasse and on the characteristics of the bio-oil obtained via slow pyrolysis. fuel processing technology, 210(august), 106561. https://doi.org/10.1016/j.fuproc.2020.106561 srisang, s., phetpan, k., ruttanadech, n., limmun, w., youryon, p., kongtragoul, p., … chungcharoen, t. (2022). charcoal briquette production from waste in the coffee production process using hydrothermal and torrefaction techniques: a comparative study with carbonization technique. journal of cleaner production, 372(august), 133744. https://doi.org/10.1016/j.jclepro.2022.133744 sulistyaningkarti, l., & utami, b. (2017). making charcoal briquettes from corncob organic waste using variations in type and percentage of adhesives. jurnal kimia dan pendidikan kimia, 2(1), 43–53. sunardi, djuanda, & mandra, m. a. s. (2019). characteristics of charcoal briquettes from agricultural waste with compaction pressure and particle size variation as alternative fuel. international energy journal, 19, 139–148. suryaningsih, s., resitasari, r., & nurhilal, o. (2019). analysis of biomass briquettes based on carbonized rice husk and jatropha seed waste by using newspaper waste pulp as an adhesive material. journal of physics: conference series, 1280(2). https://doi.org/10.1088/1742-6596/1280/2/022072 syarief, a., nugraha, a., ramadhan, m. n., fitriyadi, & supit, g. g. (2021). effect of variation in composition and type of adhesive on physical properties and burning characteristics of alaban wood charcoal waste briquettes (vitex pubescens vahl) rice husk (oryza sativa l). proceedings of the national wetland environment seminar. banjarmasin. todaro, l., rita, a., cetera, p., & d’auria, m. (2015). thermal treatment modifies the calorific value and ash content in some wood species. fuel, 140, 1–3. https://doi.org/10.1016/j.fuel.2014.09.060 tu, w., liu, y., xie, z., chen, m., ma, l., du, g., & zhu, m. (2021). a novel activation-hydrochar via hydrothermal carbonization and koh activation of sewage sludge and coconut shell for biomass wastes: preparation, characterization and adsorption properties. journal of colloid and interface science, 593, 390–407. https://doi.org/10.1016/j.jcis.2021.02.133 tzelepi, v., zeneli, m., kourkoumpas, d. s., karampinis, e., gypakis, a., nikolopoulos, n., & grammelis, p. (2020). biomass availability in europe as an alternative fuel for full conversion of lignite power plants: a critical review. energies, 13(13). https://doi.org/10.3390/en13133390 vaish, s., sharma, n. k., & kaur, g. (2022). a review on various types of densification/briquetting technologies of biomass residues. iop conference series: materials science and engineering, 1228(1), 012019. https://doi.org/10.1088/1757-899x/1228/1/012019 velusamy, s., subbaiyan, a., kandasamy, s., shanmugamoorthi, m., & thirumoorthy, p. (2022). combustion characteristics of biomass fuel briquettes from onion peels and tamarind shells. archives of environmental and occupational health, 77(3), 251–262. https://doi.org/10.1080/19338244.2021.1936437 yana, s., nizar, m., irhamni, & mulyati, d. (2022). biomass waste as a renewable energy in developing bio-based economies in indonesia: a review. renewable and sustainable energy reviews, 160(5), 112268. https://doi.org/10.1016/j.rser.2022.112268 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1016/j.sajce.2022.08.009 https://doi.org/10.1007/s40974-020-00148-4 https://doi.org/10.1016/j.heliyon.2020.e04574 https://doi.org/10.1016/j.crgsc.2021.100083 https://doi.org/10.1016/j.biteb.2021.100772 https://doi.org/10.1016/j.rser.2014.07.107 https://doi.org/10.1016/j.fuel.2019.116150 https://doi.org/10.1088/1757-899x/1122/1/012079 https://doi.org/10.1051/e3sconf/202132808008 https://doi.org/10.1016/j.fuproc.2014.09.015 https://doi.org/10.3390/en13051060 https://doi.org/10.1088/1742-6596/1080/1/012024 https://doi.org/10.1088/1742-6596/1080/1/012024 https://doi.org/10.1088/1755-1315/462/1/012057 https://doi.org/10.1088/1755-1315/462/1/012057 https://doi.org/10.1016/j.fuel.2018.04.040 https://www.ijera.com/papers/vol4_issue3/version%201/ct4301553559.pdf https://www.ijera.com/papers/vol4_issue3/version%201/ct4301553559.pdf https://doi.org/10.1016/j.fuproc.2020.106561 https://doi.org/10.1016/j.jclepro.2022.133744 https://doi.org/10.1088/1742-6596/1280/2/022072 https://doi.org/10.1016/j.fuel.2014.09.060 https://doi.org/10.1016/j.jcis.2021.02.133 https://doi.org/10.3390/en13133390 https://doi.org/10.1088/1757-899x/1228/1/012019 https://doi.org/10.1080/19338244.2021.1936437 https://doi.org/10.1016/j.rser.2022.112268 international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (4), 702-710 |702 https://doi.org/10.14710/ijred.2023.51314 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id development of biobr/tio2 nanotubes electrode for conversion of nitrogen to ammonia in a tandem photoelectrochemical cell under visible light prita amelia and jarnuzi gunlazuardi* department of chemistry, faculty of mathematics and natural sciences, university of indonesia, indonesia abstract. ammonia (nh3) is one of the important chemicals for human life. the demand for ammonia is expected to increase every year. conventionally, the fixation process of n2 to produce nh3 in the industrial sector is carried out through the haber−bosch process, which requires extreme temperature and pressure conditions that consume a high amount of energy and emit a considerable amount of co2. therefore, it is necessary to develop alternative technology to produce ammonia using environmentally friendly methods. many studies have developed the photoelectrochemical conversion of nitrogen to ammonia in the presence of semiconductor materials, but the resulting efficiency is still not as expected. in this research, the development of the tandem system of dye-sensitized solar cell photoelectrochemistry (dssc pec) was carried out for the conversion of nitrogen to ammonia. the dssc cell was prepared using n719/tio2 nanotubes as photoanode, pt/fto as cathode, and electrolyte i/i3-. the dssc efficiency produced in this research was 1.49%. pec cell at the cathode and anode were prepared using biobr/tio2 nanotubes synthesized by the silar (successive ionic layer adsorption and reaction) method. the resulting ammonia levels were analyzed using the phenate method. in this study, ammonia levels were obtained at 0.1272 µmol for 6 hours of irradiation with an scc (solar to chemical conversion) percentage of 0.0021%. keywords: ammonia, biobr/tio2 nanotubes, dssc, nitrogen fixation, photo-electrochemistry @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 29th dec 2022; revised: 26th march 2023; accepted: 27th may 2023; available online: 23rd june 2023 1. introduction ammonia (nh3) has a significant role in human life, including in industry and agriculture (lan et al., 2021). ammonia is used as the main raw material in a variety of chemical products, such as fertilizers, nitric acid, and freezing agents (feng et al., 2021). in addition, nh3 is an essential material for hydrogen storage with a high-energy hydrogen layer and is environmentally friendly. the global output of ammonia is about 15 billion tons per year (huang et al., 2021). one important source of ammonia is the nitrogen (n2) atmosphere (cao et al. 2018). although about 78% of the earth's atmosphere contains nitrogen, most organisms find its molecular form unusable due to the non-polar covalent triple bonds of nn. this bond is highly resistant to dissociation and has a high ionization energy and negative electron affinity. traditionally, the industrial fixation of n2 to nh3 was carried out via the haber−bosch process, which requires extreme conditions (400–600oc, 20–40 mpa) in the presence of an ironbased catalyst to overcome the kinetic limitations (wang et al., 2022). this process involves about 2% of the world's energy supply and emits 2.3 tons of co2 annually. in addition, the hydrogen gas used to synthesize nh3 is mainly produced from the reformation of methane vapor, which requires about 3−5 % of the world's annual natural gas production and emits large * corresponding author email: jarnuzi@ui.ac.id (j. gunlazuardi) amounts of co2 (li et al., 2015). given the shortage of fossil fuels and global climate change, a fixation process of nitrogen that requires less energy consumption is required for long-term goals. the catalytic process that produces nh3 from n2 at room temperature and atmospheric pressure is a promising method for the sustainable, safe, and clean synthesis of nh3 (olabi et al., 2023). the photo-electrocatalytic reaction to convert n2 to nh3 has attracted a lot of attention due to the availability of light energy, making it more environmentally friendly and energy efficient. many researchers have developed photo-electrocatalytic techniques using semiconductor materials to utilize solar energy (humayun et al., 2018). one of the semiconductor materials that has been widely developed is tio2 because it is relatively efficient, inexpensive, inert, non-toxic, and has good photocatalytic activity (hoang et al., 2021). however, the use of tio2 with relatively large band gap energy (3.0 ev for the rutile phase and 3.2 ev for the anatase phase) requires ultraviolet (uv) light for electron-hole separation, so tio2 is not efficient in sunlight, which only contains ~5% uv light (moghni et al., 2022). therefore, it is very important to develop photocatalysts that can be used in both visible light (400-700 nm) and ultraviolet (uv) light (290-400 nm) to increase photo-electrocatalytic efficiency. research article https://doi.org/10.14710/ijred.2023.51314 https://doi.org/10.14710/ijred.2023.51314 mailto:jarnuzi@ui.ac.id https://orcid.org/0009-0008-7497-7603 https://orcid.org/0000-0001-5991-7110 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.51314&domain=pdf p. amelia and j.gunlazuardi int. j. renew. energy dev 2023, 12(4), 702-710 | 703 issn: 2252-4940/© 2023. the author(s). published by cbiore several modifications that can be made to increase the photo-electrocatalytic activity of tio2 in visible light are doping with metal and non-metallic elements, deposition with noble metals, combining other semiconductor materials to form heterojunctions and surface modification with inorganic acids (zhang et al., 2016). on the other hand, bismuth-based semiconductor materials have been widely developed for fixing nitrogen into ammonia because they show good activity as photocatalysts (huang et al., 2021). shiraishi et al. (2020) developed a bismuth oxyhalide semiconductor [biox (x = cl, br, i)] with the consideration that biox is a non-toxic, inexpensive, and environmentally friendly semiconductor (shiraishi et al., 2020). on the other hand, tio2 has a high conduction band, so a semiconductor with an appropriate conduction band position is required. biobr is a semiconductor that responds to visible light and has a more negative conduction band than tio2. the use of biobr has attracted attention in composites with tio2, as this facilitates the injection of electrons from the conduction band of biobr into tio2 and reduces the recombination of photoelectrons and holes. (jia et al., 2018). in this research, a tandem system of dye-sensitized solar cell (dssc) and photo-electrochemical (pec) cell was developed for the conversion of nitrogen to ammonia, aiming to reduce electrical energy consumption. the pec cell served as a reaction site, utilizing visible light for the conversion process, while the dssc cell will provide a source of electrons. the cathode and anode in the pec cell was made of biobr/tio2 nanotubes, synthesized using the silar method (sreedev et al., 2019). biobr served as the catalyst for the photoelectrocatalytic conversion of n2 to nh3 on the cathode, while on the photoanode, biobr facilitated the efficient oxidation of h2o. the dssc zone utilized n719/tio2 nanotubes as the photoanode, pt/fto as the cathode, and electrolyte i-/i3-. this research aims to produce ammonia with good solar to ammonia efficiency using environmentally friendly methods with low energy consumption. 2. materials and methods 2.1 materials the ti plate, acetone (c3h6o), ethanol p.a (c2h5oh), ethylene glycol (c2h6o), ammonium fluoride (nh4f), bismuth nitrate pentahydrate (bi(no3)3.5h2o), ammonium chloride (nh4cl), sodium sulfate (na2so4), trisodium citrate (na3c6h5o7), sodium hypochlorite (naocl), sodium hydroxide (naoh), sodium nitroprusside (snp), sodium bromide (nabr), fto, dye n719, nafion membrane 117, phenol 99% and deionized water were purchased commercially. all materials were obtained from sigma-aldrich, except for the titanium plate (99.6% purity) obtained from baoji jinsheng metal material co. ltd and deionized water from onemed. 2.2 methods 2.2.1 preparation of tio2 nanotubes (tio2nts) the titanium plate (6 cm x 1.5 cm x 0.02 cm) was cleaned through a sonication process at room temperature in acetone, ethanol, and water for 15 minutes each, then dried in air. all anodization experiments were carried out in an electrochemical cell of two electrodes. a ti plate and stainless steel were used as the anode and cathode, respectively. an ethylene glycol solution containing 2% h2o and 0.3% nh4f was used as the electrolyte. the distance between the two electrodes was set at about 1.5 cm. the anodization process was performed for 60 minutes at a potential of 40 v. after the anodization process, the sample was rinsed with deionized water and dried in the open air, then calcined for 2 hours at 450oc with a temperature increase rate of 5oc/minute (surahman et al., 2015). 2.2.2 preparation of biobr/tio2 nanotubes (biobr/tio2nts) biobr/tio2nts were prepared by the silar (successive ionic layer adsorption and reaction) method, where 5 mm bi(no3)3⋅5h2o and 5 mm nabr were dissolved in 100 ml of 0,1 m mannitol solution and deionized water, respectively. tio2nts were first immersed in bi-solution for 1 minute, followed by rinsing with deionized water, then immersed in br-solution for 1 minute, followed by rinsing with deionized water again. these experiments were carried out in a 40oc water bath to produce a depositing of biobr onto tio2nts (ma et al., 2021). the abovementioned process is called one cycle of the silar process and was repeated 5, 10, and 10 times. this variation silar cycle was denoted as biobr/tio2nts (5), biobr/tio2nts (10), and biobr/tio2nts (15), respectively. 2.2.3 preparation of dye-sensitized solar cell (dssc) there are three main components in dssc, and those are n719/tio2nts, pt/fto, and i-/i3that were used as photoanode, cathode, and electrolyte solution, respectively. for n719/tio2nts preparation, tio2nts was immersed in 300 µm n719 dye solution (ethanol solvent) for 24 hours. after 24 hours of immersion, the n719 sensitized tio2nts plate was cleaned with ethanol and dried in the air. for pt/fto preparation, the fto glass (4 x 1.5 cm) was cleaned by sonication in ethanol for 10 minutes. the conducting part of the fto glass dripped slowly with 20 mm h2ptcl6 solution in ethanol, dried in the open air, and then heated at 380℃ for 30 minutes (neetu et al., 2017). the i-/i3electrolyte solution was prepared by dissolving 0.13 g of i2 crystals in a mixed solvent of 5 ml ethylene glycol and 20 ml acetonitrile, then adding 0.18 g of ki continuing by stirring for 30 minutes (gu et al., 2017). the arrangement of the dssc cells followed a sandwich cell configuration, where electrolyte i-/i3was dripped on the surface of the photoanode (n719/tio2nts), then parafilm was placed as a separator between the photoanode and cathode to avoid short circuit current, and then closed with a cathode (pt/fto). the photovoltaic performance of the dssc was investigated by current–voltage (j–v) measurement under 40 watt phillips tungsten lamp as the source of visible light. by doing this measurement, a graph of current vs. potential was obtained. as the material was in the form of a plate, the current density was determined by dividing the measured current by the surface area of the plate that was immersed in the electrolyte. furthermore, the filling factor (ff) and  (% efficiency of dssc) were calculated using the following equation (chougala et al., 2017): 𝐹𝐹 = 𝑉𝑚𝑎𝑥 × 𝐽𝑚𝑎𝑥 𝑉𝑜𝑐×𝐽𝑠𝑐 (1) 𝜂 (% 𝑒𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦) = 𝐹𝐹 ×𝐽𝑠𝑐×𝑉𝑜𝑐 𝐼0 × 100% (2) where jsc = short circuit current density, jmax = maximum power point current density, voc = open circuit voltage, and vmax = maximum power point voltage. p. amelia and j.gunlazuardi int. j. renew. energy dev 2023, 12(4), 702-710 | 704 issn: 2252-4940/© 2023. the author(s). published by cbiore 2.2.4 electrochemical cell measurement the preparation of the photo-electrochemical cell was carried out using a 3-electrode system potentiostat. the working electrodes were tio2nts, biobr/tio2nts, and n719/tio2nts, while the counter electrode was pt and the reference electrode was ag/agcl. a 40 watt phillips tungsten lamp was used as the source of visible light. electrochemical measurement for tio2nts and biobr/tio2nts was carried out using 0.1 m na2so4 electrolyte, and n719/tio2nts was carried out using 0.01 m ki electrolyte in acetylacetone. measurements were conducted using two methods, those are linear sweep voltammetry (lsv) with a scan rate of 25 mv/s and multi pulsed amperometry (mpa) with a constant potential of 0 v in dark and light conditions. 2.2.5 preparation of the tandemsystem of dssc-pec for the conversion of nitrogen to ammonia the conversion of nitrogen to ammonia was carried out in an h-type reactor connected with a nafion membrane. biobr/tio2nts served as the photoanode in the pec cell for the water oxidation reaction to produce proton and electron. this electron will flow to the dssc cathode (pt/fto). meanwhile, the dssc photoanode (n719/tio2nts) acted as a source of electrons to the pec cathode, where the reduction reaction of nitrogen to ammonia takes place, using biobr/tio2nts as well. the pec photoanode and dssc cathode, as well as the dssc photoanode and pec cathode, were connected with cu wires to facilitate the electron flow, as illustrated in figure 1. philips 400 watt tungsten lamp was used as a source of visible light radiation. the electrolyte used in the reactor was 0.1 m na2so4 solution with a reaction time of 6 hours. the nh3 gas formed was absorbed in a 0.01 m hcl solution to form nh4+ and then analyzed using spectrophotometric methods through phenate methods. ammonia levels were determined using the phenate method with a uv-visible spectrophotometer at a wavelength of 640 nm which refers to sni 06-6989.30-2005. samples containing ammonia will form a blue indophenol complex when reacted with phenol, alkaline citrate solution (trisodium citrate, naoh, and naocl), and sodium nitroprusside. a total of 5 ml of sample or standard solution of nh4cl at various concentrations was added 0.2 ml of phenol solution; 0.2 ml nitroprusside; and 0.5 ml of oxidizing reagent (alkaline citrate solution), then homogenized and stored in a dark place for 2 hours. then a blue color was formed and measured using a uvvisible spectrophotometer at a wavelength of 640 nm. furthermore, the percent efficiency of conversion to ammonia was calculated using the solar to chemical conversion (scc) equation as follows (shiraishi et al., 2018): 𝑆𝐶𝐶 𝐸𝑓𝑓𝑖𝑐𝑖𝑒𝑛𝑐𝑦 (%) = [∆𝐺 𝑓𝑜𝑟 𝑁𝐻3𝑓𝑜𝑟𝑚𝑎𝑡𝑖𝑜𝑛 (𝐽𝑚𝑜𝑙 − 1)] × [𝑁𝐻3𝑓𝑜𝑟𝑚𝑒𝑑 (𝑚𝑜𝑙)] [𝑡𝑜𝑡𝑎𝑙 𝑖𝑛𝑝𝑢𝑡 𝑒𝑛𝑒𝑟𝑔𝑦 (𝑊)] × [𝑟𝑒𝑎𝑐𝑡𝑖𝑜𝑛 𝑡𝑖𝑚𝑒 (𝑠)] × 100% (3) where g is the free energy of ammonia formation from nitrogen gas and water, which is 399 kj/mol. 2.3 characterization tio2nts and biobr/tio2nts were characterized using xrd (x'pert pro merk panalitycal mpd paw3040/60), ftir (shimadzu ir prestige 21), raman microscopes (horiba the labram hr evolution), uv-vis drs (shimadzu uv-2450), sem-edx (quanta 650 of thermo scientific), and potentiostat (par-versastat ii). while n719/tio2nts was characterized using ftir, uv-vis drs, and potentiostat. characterization of xrd was used to determine the crystallite size of the synthesized material. the crystallite size was estimated using the fwhm (full width at half maximum) of the high intensity peak appearing at 2θ = 25.31° (101) using the scherer formula (alkorbi et al., 2022): 𝐷 = 𝑘𝜆 𝛽𝑐𝑜𝑠𝜃 (4) where d is the crystallite size (nm), k is a constant with the value of 0.89, λ is the x-ray wavelength (nm), θ is the bragg angle (radians), and β is the fwhm (radians). meanwhhile, the characterization using uv-vis drs was used to determine the energy band gap value. the energy band gap was determined by using the diffuse reflectance spectra and the kubelka-munk method according to the following equation (garzon-roman et al., 2020), 𝐹(𝑅) = (1 − 𝑅)2 2𝑅 (5) where r is the diffuse reflectance. in this case, f(r) becomes an absorbance function, and the following equation was used to determine the bandgap: (𝐹(𝑅)ℎ𝑣) 1 2 𝑣𝑠 𝐸 (6) 3. results and discussion 3.1 anodic oxidation of tio2nts figure 2 shows the three stages of the formation of tio2nts, the first stage (i) shows a very sharp decrease in current density from 22.2 ma/cm2 to 4.48 ma/cm2 in 50 seconds. the decrease in current is due to the oxidation of the ti metal surface by releasing ti4+ ions and electrons, which causes the formation of a tio2 oxide layer on the ti metal surface. in this case, there was an interaction between water and oxidized ti metal (broens et al., 2023). the tio2 oxide layer that was formed electrochemically acts as a barrier layer and obstructs the flow fig. 1 the schematic diagram of the dssc-pec cell for conversion of nitrogen to ammonia p. amelia and j.gunlazuardi int. j. renew. energy dev 2023, 12(4), 702-710 | 705 issn: 2252-4940/© 2023. the author(s). published by cbiore of ions and electrons so that the potential between titanium metal and the electrolyte decreases, which is indicated by a decrease in current during the anodization process (qian et al., 2022). according to (indira et al., 2015), the reactions that occur in the first stage (stage i) are as follows: oxidation of ti: ti → ti4+ + 4e− (7) oxide layer formation: ti4+ + 2h2o → tio2 + 4h + (8) in stage ii, the current increased from 4.48 ma/cm2 to 7.11 ma/cm2 within 40 seconds. the increase in current was caused by forming small holes and pores in the tio2 oxide layer through electric field dissolution reactions and chemical dissolution so that the anode was more conducive (qin et al., 2021). small holes in the oxide layer were formed due to an electric field breaking the ti and o bonds. the oxide layer with pinholes reacted with f(fluoride) ions and caused the current density to increase, as seen in stage ii. this stage produced a porous oxide layer with a particular size and depth. the reaction of fions with the oxide layer produced a complex compound [tif6]2 with the following reaction (indira et al., 2015): 𝑇𝑖𝑂2 + 6𝐹 − + 4𝐻+ → [𝑇𝑖𝐹6] 2− + 2𝐻2𝑂 (9) in stage iii, there was a relatively stable decrease in current density. as the anodization time increased, the current decreased slightly due to the change in the pore depth of the holes formed. the pore growth rate and nanotube length were determined by the competition between electrochemical oxide formation and chemical dissolution by fluoride ions (yoo et al., 2018). this process produced a structure of vertically growing nanotubes. the tio2 material formed from the anodization process was amorphous. next, a calcination process was carried out at 450oc for 2 hours to change from the amorphous phase to the anatase phase (fang et al., 2011). 3.2 xrd patterns, structure, and morphology observations figure 3 shows the diffractogram of tio2nts and biobr/ tio2nts. the 2θ value of the synthesized material was compared with the icdd database. based on icdd data no. 01089-4921 the typical peak of anatase tio2 is at position 25.36° ; 37.85° ; 48.15° ; 53.97° ; 55.19° ; 62.81° ; 68.88° ; 74.18° ; 75.21° ; 76.22°. based on icdd data no. 00-044-1294 the typical peak of ti metal is at position 35.09° ; 38.42° ; 40.17° ; 53.01° ; 70.66° ; 82.29° ; 86.76° ; 92.73°. the typical peak produced by the synthesized material shows the peak according to the icdd data. so, it can be known that the crystalline phase formed on the synthesized tio2nts is the anatase phase. the estimated crystallite size of tio2nts calculated using equation (4) was 26.23 nm. in the biobr/tio2nts diffraction pattern, several additional peaks were observed compared to the tio2nts diffraction pattern. based on icdd data with reference code 00-003-0733, there are additional peaks at positions 2θ of 32.41°, 46.53°, 53.55°, and 57.56° respectively, according to the crystal plane (1 1 0), (2 0 0), (2 1 1), and (2 1 2) in the tetragonal phase of biobr (yu et al., 2023). as the silar cycle increased, the peak observed at position 2θ became clearer. in this case, the 15cycle treatment showed a peak of 2θ for biobr, which was clearer than the 5 and 10-cycle treatments. this is because the more cycle variations, the more biobr will be deposited on the surface of tio2nts so that more dominant biobr facets will be formed compared to fewer cycles. figure 4 shows the ftir spectra of tio2nts and biobr/tio2nts to analyze functional groups. based on the ftir spectra, there were five main absorptions in biobr/tio2nts. ti-o stretching vibrations could be shown at 455-861 cm-1, ti-o-ti stretching vibrations at 1630 cm-1 (singh & dutta, 2018), oh bending vibrations at 1437-1600 cm-1, oh stretching vibrations at 3000-3612 cm-1, and bi-o stretching vibration at 512 cm-1 (mera et al., 2018). figure 5 shows the raman spectra of biobr/tio2nts compared to tio2nts. it could be observed that there are several peaks which are typical peaks of anatase tio2 at raman shifts of 147, 197, 395, 517, and 637 cm-1, which correspond respectively to eg(1), eg(2), b1g, a1g and eg(3) for the anatase phase mode. the eg mode could appear due to the o–ti–o fig. 2 current density profile of tio2nts formations fig. 3 xrd patterns of tio2nts and biobr/tio2nts fig. 4 ftir spectra of tio2nts and biobr/tio2nts p. amelia and j.gunlazuardi int. j. renew. energy dev 2023, 12(4), 702-710 | 706 issn: 2252-4940/© 2023. the author(s). published by cbiore symmetrical stretching vibration, the b1g mode was assigned to the o–ti–o symmetric bending vibration, and the a1g mode corresponded to the ti–o–ti antisymmetric bending vibration in tio2. these results were consistent with research conducted by (boda & shah, 2017) which showed that the raman spectra of tio2nts could be observed in the range of 245 – 351cm−1 which corresponds to the covalent interaction of o–o. the active mode at 147 and 197 cm−1 were assigned to the unit cell's ti interaction. the strongest mode that appears at 147 cm−1 is due to external symmetrical vibrations confirming the anatase phase's formation. in addition, there was an additional peak at 125 cm-1 which corresponds to the a1g stretching vibration of the bi–br bond. based on the literature (wang et al., 2016), three characteristic bands in biobr could be observed at 112, 162, and 385 cm−1 which correspond respectively to the a1g internal stretching vibration of the bi−br bond, the eg internal stretching vibration of the bi−br bond and the b1g band generated by the vibration oxygen atom from the bi-o bond. however, in the results of this study, only the a1g band could be clearly observed in the raman spectra of biobr/tio2nts, while the two characteristic bands could not be observed clearly. this might be caused by the two other bands overlapping with the characteristic raman band of tio2nts, whose intensity was very strong so that the characteristic band of biobr/tio2nts could not be observed clearly. sem images of biobr/tio2nts (figure 6a) show the morphology of the surface of biobr/tio2nts (15) at 100,000x magnification. based on these results, it could be observed that tio2 was formed with nanotube morphology deposited by biobr on the surface, which is marked by a white color on the surface of tio2nts, and the tube of tio2nts was still visible. this result shows that the deposition of biobr on the surface of tio2nts did not block the mouth of the tube. figure 6(b) shows the morphology of the biobr/tio2nts cross-section with a magnification of 15,000 times showing a tube height of 4.4 µm and it could be observed that the cross section showed several small particles indicating the biobr compound. in addition to characterization using sem, characterization was also carried out using edx which aims to determine the composition of each constituent element. figure 7(a) shows the point edx spectra of biobr/tio2nts (15). these results indicated the presence of bi and br elements, in addition to ti and o from tio2 compounds which allows the formation of biobr on the surface of tio2. figure 7(b) shows the edx mapping of biobr/tio2nts (15) indicating that the elements bi, o, br, and ti have been formed and distributed evenly. 3.3 optical absorption and photo-electrochemical performance figure 8 shows the results of the characterization using uvdrs. the more silar cycles indicate a shift in the wavelength absorption toward the visible region. for biobr/tio2nts (15), biobr/tio2nts (10), and biobr/tio2nts (5) the uv–vis diffuse reflectance spectroscopy (drs) spectra present a slight red shift of adsorption edge and a higher visible light absorbance. the kubelka-munk model, presented in equation (5), was used to estimate the band gap energy of all samples by plotting (f(r)ℎv)1/2 versus the energy of absorbed light (landi et al., 2022). the band gap energy values obtained for tio2nts fig. 5 raman spectra of tio2nts and biobr/tio2nts fig. 6 sem images of biobr/tio2nts 15 silar cycles (a) surface with 100,000x magnification and (b) cross section with 15,000x magnification fig. 7 edx point (a) and mapping (b) of biobr/tio2nts p. amelia and j.gunlazuardi int. j. renew. energy dev 2023, 12(4), 702-710 | 707 issn: 2252-4940/© 2023. the author(s). published by cbiore was 3,20 ev, and for each variation of biobr/tio2nts, were 2.96 ev, 2.94 ev, and 2.92 ev, respectively for 5x, 10x, and 15x cycles of silar process. these results indicated that the more silar cycles, the more biobr would be deposited on the tio2nts surface, making it more active under visible light. this was indicated by shifting the absorption to the visible region and the decrease in the band gap energy, which becomes smaller. figure 9 shows electrochemical performance results using the lsv and mpa methods. under visible light irradiation, measurements using the lsv method were carried out over a potential range of -1v to +1v. these results indicated that the deposition of biobr on the surface of tio2nts could increase the resulting current density compared to tio2nts without adding biobr (figure 9a). the highest current density was produced in biobr/tio2nts (15), with a variation of 15x silar cycles, because more biobr might be deposited on the surface of tio2nts. as a result, the current density was higher in irradiation using visible lamps compared to 5 and 10 cycles. figure 9(b) shows the measurement results using the mpa method on biobr/tio2nts within 100 seconds in dark and light conditions alternately with a duration of 10 seconds for each condition. these results indicated that when the biobr/tio2nts were irradiated using a visible lamp, there was an increase in current density, and when the visible light was turned off, there was a significant decrease in current density. the results of this measurement indicated that biobr/tio2nts had a higher current density response when compared to tio2nts, where biobr/tio2nts produced the highest current density with 15 silar cycles. based on measurements using the lsv and mpa methods, it shows that the deposition of biobr on the surface of tio2nts could increase the current density, which makes it active in the visible region, compared to tio2nts without modification with biobr. this result was observed because the addition of biobr could produce oxygen vacancies on the surface, increasing photocatalytic activity and exhibiting higher intensity of photocurrent. it suggests that the photo-induced hole-electron pairs in biobr are efficiently separated (wang et al., 2021). 3.4 dssc efficiency the dssc components were arranged like a sandwich structure using n719/tio2nts as the anode, pt/fto as the cathode, and i-/i3as the electrolyte solution. in addition, a parafilm spacer was used as a barrier between the anode and cathode to avoid short circuits. dssc performance was carried out using a potentiostat by connecting the anode and reference wires to the n719/tio2nts plate and the cathode wires to the pt/fto. figure 10 shows the photocurrent-to-potential curve given the irradiation conditions using a visible lamp. efficiency was determined based on data from the photocurrent-to-potential change curve. based on this curve, several parameters were obtained that could be used to calculate the filling factor and dssc efficiency, and those are short circuit current density (jsc), maximum power point current density (jmax), open circuit voltage (voc), maximum power point voltage (vmax) with the resulting values respectively 0.2028 ma/cm2, 0.1202 ma/cm2, 0.5280 v, and 0.320 v. based on the results of the calculation using equation (1) and (2), by using the total incident irradiance (i0) of 2.58 mw/cm2, fig. 8 uv-drs spectra of tio2nts and biobr/tio2nts fig. 9 electrochemical performance of tio2nts and biobr/tio2nts using (a) linear sweep voltammetry (lsv) method and (b) multi pulse amperometry (mpa) method under visible light irradiation fig. 10 photocurrent density vs. potential of dssc p. amelia and j.gunlazuardi int. j. renew. energy dev 2023, 12(4), 702-710 | 708 issn: 2252-4940/© 2023. the author(s). published by cbiore the ff value is 0.3592, and the dssc efficiency is 1.49%. this dssc cell will be used in a dssc-pec tandem system for the conversion of nitrogen to ammonia to increase the electrons that would be entering the catalysis zone so that more ammonia will be produced. 3.5 conversion of nitrogen to ammonia in a dssc-pec tandem system the conversion of nitrogen to ammonia was carried out in an h-type reactor that utilizes the dssc-pec system. the conversion reaction of nitrogen was a reduction reaction that occurs at the pec cathode. the pec cathode was connected to n719/tio2nts, which functions as a dssc anode for visible light harvesting. meanwhile, the pec anode was where the water oxidation reaction occurs, producing electrons (e-) and protons (h+). these electrons would go to the dssc cathode (pt/fto) through an external circuit which would later be used to reduce i3to iions in dssc. at the same time, the protons would go to the pec cathode through the nafion membrane, which would involve converting nitrogen into ammonia. in the photoanode part of the pec cell, biobr/tio2nts (15) were used, which is expected to facilitate the oxidation reaction of water into electrons (e-) and protons (h+) under visible light radiation. while at the pec cathode, biobr/tio2nts (15) were used to facilitate the conversion process of nitrogen to ammonia through the reduction reaction. the reaction is as follows: reaction at the photoanode: 2h2o (l) → o2 (g) + 4h+ + 4e (10) reaction at the cathode: n2 (g) + 6h+ + 6e→ 2nh3 (g) (11) furthermore, the resulting ammonia flowed into a glass containing 0.01 m hcl solution to produce ammonium chloride solution with the following reaction: nh3 (g) + hcl (aq) → nh4cl (aq) (12) the experiment was carried out for 6 hours, where samples were taken every 2 hours. the samples were analyzed spectrophotometrically using the phenate method. table 1 shows the amount of ammonia produced within 2, 4, and 6 hours. this data shows that the longer the irradiation time, the more ammonia will be produced. when irradiated by visible light in the anode, holes transferred from tio2 to biobr for the oxidation reaction of water, while in the cathode, electrons on biobr transferred to the tio2 surface for nitrogen fixation reaction. the photo-electrocatalytic activity of biobr/tio2nts heterojunction could be ascribed to effective interfacial interaction (wang et al., 2021). furthermore, the percent efficiency of conversion to ammonia was calculated using equation (3). the total energy input (w) is the amount of light given from the irradiation source to the active area of the dssc. the light intensity used in this study was 312.5 w/m2 with an irradiated dssc area of 0.0003 m2, so the total power input generated was 0.09375 watts. based on the calculations, the solar to chemical conversion (scc) percentage was 0.0028%, 0,0021%, and 0,0021% for a reaction time of 2, 4, and 6 hours, respectively. in comparison to hirakawa et al.'s research (2017), where the scc was reported to be 0.02%, our result is lower. this discrepancy may be due to the fact that they used tio2 in powder form, which increased the contact area between the catalyst and the reactants. they also used a higher lamp power, which excited more electrons from the valence band to the conduction band (hirakawa et al., 2017). however, our results are consistent with an'nur et al.'s (2020) research, where the scc value for ammonia production using a dssc-pec system was reported to be 0.005% after 24 hours of irradiation (an'nur et al., 2020). it should be noted that the difference in scc values between our study and an'nur et al.'s study may also be attributed to the difference in the duration of the experiment. while our study was conducted for 6 hours, an'nur et al. carried out their experiment for 24 hours. the longer duration of their experiment might have allowed for more ammonia production, leading to a higher scc value. however, despite the shorter duration of our experiment, our study still provides valuable insights into the potential use of dssc-pec systems for solardriven ammonia synthesis. 4. conclusion in summary, we have developed the combined system of dsscpec to convert nitrogen to ammonia under visible light without external input energy. biobr/tio2nts heterojunction in the pec cell was obtained through a facile synthesis using the silar method. the characterization showed that biobr was well deposited on the surface of the tio2 nanotubes to form the desired heterojunction structure. in the dssc cell, n719/tio2nts-based material converted light energy to produce active electrons to be injected into the pec cathode for nitrogen reduction. the ammonia produced was 0.1272 µmoles for 6 hours, with the percentage of scc being 0.0021%. this work provides new prospects for developing the construction of the combined system of dssc-pec using heterojunction catalysts for efficient photo-electrocatalytic nitrogen conversion to ammonia. acknowledgments the authors would like to thank the ministry of education, culture, research, and technology, which has provided research funding through penelitian tesis magister (ptm) in 2022. author contributions: p.a.: performed the experiments and data analysis, methodology, writing—original draft, review and editing. j.g.; supervision, project administration, writing—review and editing, table 1 the amount of ammonia produced irradiation time (hour) concentration (ppm) amount of ammonia (µmol) scc (%) 2 0.0947 0.0557 0.0028 4 0.1393 0.0819 0.0021 6 0.2162 0.1272 0.0021 p. amelia and j.gunlazuardi int. j. renew. energy dev 2023, 12(4), 702-710 | 709 issn: 2252-4940/© 2023. the author(s). published by cbiore validation. all authors have read and agreed to the published version of the manuscript. funding: this research was funded by the ministry of education, culture, research, and technology, contract number 987/un2.rst/hkp.05.00/2022. conflicts of interest: the authors declare no conflict of interest. references alkorbi, a. s., muhammad asif javed, h., hussain, s., latif, s., mahr, m. s., mustafa, m. s., alsaiari, r., & alhemiary, n. a. (2022). solar light-driven photocatalytic degradation of methyl blue by carbon-doped tio2 nanoparticles. optical materials, 127, 112259. https://doi.org/10.1016/j.optmat.2022.112259 an’nur, f. k., wihelmina, b. v., gunlazuardi, j., & wibowo, r. (2020). tandem system of dyes sensitized solar cell-photo electro chemical (dssc-pec) employing tio2 nanotube/biobr as dark cathode for nitrogen fixation. aip conference proceedings, 2243(june 2020). https://doi.org/10.1063/5.0001100 boda, m. a., & shah, m. a. (2017). fabrication mechanism of compact tio2 nanotubes and their photo-electrochemical ability. materials research express, 4(7). https://doi.org/10.1088/20531591/aa7cd2 broens, m. i., ramos cervantes, w., asenjo collao, a. m., iglesias, r. a., teijelo, m. l., & linarez pérez, o. e. (2023). tio2 nanotube arrays grown in ethylene glycol-based media containing fluoride: understanding the effect of early anodization stages on the morphology. journal of electroanalytical chemistry, 935, 117314. https://doi.org/10.1016/j.jelechem.2023.117314 chougala, l. s., yatnatti, m. s., linganagoudar, r. k., kamble, r. r., & kadadevarmath, j. s. (2017). a simple approach on synthesis of tio2 nanoparticles and its application in dye sensitized solar cells. journal of nanoand electronic physics, 9(4). https://doi.org/10.21272/jnep.9(4).04005 fang, d., luo, z., huang, k., & lagoudas, d. c. (2011). effect of heat treatment on morphology, crystalline structure and photocatalysis properties of tio 2 nanotubes on ti substrate and freestanding membrane. applied surface science, 257(15), 6451– 6461. https://doi.org/10.1016/j.apsusc.2011.02.037 feng, j., zhang, x., zhang, g., li, j., song, w., & xu, z. (2021). improved photocatalytic conversion of high−concentration ammonia in water by low−cost cu/tio2 and its mechanism study. chemosphere, 274. https://doi.org/10.1016/j.chemosphere.2021.129689 garzon-roman, a., zuñiga-islas, c., & quiroga-gonzález, e. (2020). immobilization of doped tio2 nanostructures with cu or in inside of macroporous silicon using the solvothermal method: morphological, structural, optical and functional properties. ceramics international, 46(1), 1137–1147. https://doi.org/10.1016/j.ceramint.2019.09.082 gu, p., yang, d., zhu, x., sun, h., wangyang, p., li, j., & tian, h. (2017). influence of electrolyte proportion on the performance of dyesensitized solar cells. aip advances, 7(10). https://doi.org/10.1063/1.5000564 hirakawa, h., hashimoto, m., shiraishi, y., & hirai, t. (2017). photocatalytic conversion of nitrogen to ammonia with water on surface oxygen vacancies of titanium dioxide. journal of the american chemical society, 139(31), 10929–10936. https://doi.org/10.1021/jacs.7b06634 hoang, n. t.-t., tran, a. t.-k., le, t.-a., & nguyen, d. d. (2021). enhancing efficiency and photocatalytic activity of tio2-sio2 by combination of glycerol for mo degradation in continuous reactor under solar irradiation. journal of environmental chemical engineering, 9(5), 105789. https://doi.org/10.1016/j.jece.2021.105789 huang, r., li, x., gao, w., zhang, x., liang, s., & luo, m. (2021). recent advances in photocatalytic nitrogen fixation: from active sites to ammonia quantification methods. rsc advances, 11(24), 14844– 14861. https://doi.org/10.1039/d0ra10439f humayun, m., raziq, f., khan, a., & luo, w. (2018). modification strategies of tio2 for potential applications in photocatalysis: a critical review. green chemistry letters and reviews, 11(2), 86–102. https://doi.org/10.1080/17518253.2018.1440324 indira, k., mudali, u. k., nishimura, t., & rajendran, n. (2015). a review on tio2 nanotubes: influence of anodization parameters, formation mechanism, properties, corrosion behavior, and biomedical applications. journal of bioand tribocorrosion, 1(4), 1–22. https://doi.org/10.1007/s40735-0150024-x jia, l., tan, x., yu, t., & zhang, z. (2018). enhanced photoelectrocatalytic performance of temperature-dependent 2d/1d biobr/tio2-x nanotubes. materials research bulletin, 105, 322–329. https://doi.org/10.1016/j.materresbull.2018.05.005 lan, m., zheng, n., dong, x., ma, h., & zhang, x. (2021). one-step insitu synthesis of bi-decorated biobr microspheres with abundant oxygen vacancies for enhanced photocatalytic nitrogen fixation properties. colloids and surfaces a: physicochemical and engineering aspects, 623, 126744. https://doi.org/10.1016/j.colsurfa.2021.126744 landi, s., segundo, i. r., afonso, c., lima, o., costa, m. f. m., freitas, e., & carneiro, j. (2022). evaluation of band gap energy of tio2 precipitated from titanium sulphate. physica b: condensed matter, 639, 10–13. https://doi.org/10.1016/j.physb.2022.414008 li, h., shang, j., ai, z., & zhang, l. (2015). efficient visible light nitrogen fixation with biobr nanosheets of oxygen vacancies on the exposed {001} facets. journal of the american chemical society, 137(19), 6393–6399. https://doi.org/10.1021/jacs.5b03105 ma, b., xin, s., xin, y., ma, x., zhang, c., & gao, m. (2021). optimized fabrication of biobr/tio2nanotube arrays for efficient degradation of organic pollutant under visible light irradiation. journal of environmental chemical engineering, 9(2), 104833. https://doi.org/10.1016/j.jece.2020.104833 mera, a. c., rodríguez, c. a., valdés, h., jaramillo, a. f., rojas, d., & meléndrez, m. f. (2018). solvothermal synthesis and photocatalytic activity of biobr microspheres with hierarchical morphologies. acta chimica slovenica, 65(2), 429–437. https://doi.org/10.17344/acsi.2018.4181 moghni, n., boutoumi, h., khalaf, h., makaoui, n., & colón, g. (2022). enhanced photocatalytic activity of tio2/wo3 nanocomposite from sonochemical-microwave assisted synthesis for the photodegradation of ciprofloxacin and oxytetracycline antibiotics under uv and sunlight. journal of photochemistry and photobiology a: chemistry, 428(february). https://doi.org/10.1016/j.jphotochem.2022.113848 neetu, maurya, i. c., singh, s., gupta, a. k., srivastava, p., & bahadur, l. (2017). n/al-incorporated tio2 nanocomposites for improved device performance of a dye-sensitized solar cell. chemistryselect, 2(15), 4267–4276. https://doi.org/10.1002/slct.201700550 olabi, a. g., abdelkareem, m. a., al-murisi, m., shehata, n., alami, a. h., radwan, a., wilberforce, t., chae, k. j., & sayed, e. t. (2023). recent progress in green ammonia: production, applications, assessment; barriers, and its role in achieving the sustainable development goals. energy conversion and management, 277, 116594. https://doi.org/10.1016/j.enconman.2022.116594 qian, q., lin, y., xiong, z., su, p., liao, d., dai, q., chen, l., & feng, d. (2022). internal anodization of porous ti to fabricate immobilized tio2 nanotubes with a high specific surface area. electrochemistry communications, 135, 107201. https://doi.org/10.1016/j.elecom.2022.107201 qin, j., cao, z., li, h., & su, z. (2021). formation of anodic tio2 nanotube arrays with ultra-small pore size. surface and coatings technology, 405, 126661. https://doi.org/10.1016/j.surfcoat.2020.126661 shiraishi, y., hashimoto, m., chishiro, k., moriyama, k., tanaka, s., & hirai, t. (2020). photocatalytic dinitrogen fixation with water on bismuth oxychloride in chloride solutions for solar-tochemical energy conversion. journal of the american chemical society, 142(16), 7574–7583. https://doi.org/10.1021/jacs.0c01683 shiraishi, y., shiota, s., kofuji, y., hashimoto, m., chishiro, k., hirakawa, h., tanaka, s., ichikawa, s., & hirai, t. (2018). nitrogen fixation with water on carbon-nitride-based metalhttps://doi.org/10.1016/j.optmat.2022.112259 https://doi.org/10.1063/5.0001100 https://doi.org/10.1088/2053-1591/aa7cd2 https://doi.org/10.1088/2053-1591/aa7cd2 https://doi.org/10.1016/j.jelechem.2023.117314 https://doi.org/10.21272/jnep.9(4).04005 https://doi.org/10.1016/j.apsusc.2011.02.037 https://doi.org/10.1016/j.chemosphere.2021.129689 https://doi.org/10.1016/j.ceramint.2019.09.082 https://doi.org/10.1063/1.5000564 https://doi.org/10.1021/jacs.7b06634 https://doi.org/10.1016/j.jece.2021.105789 https://doi.org/10.1039/d0ra10439f https://doi.org/10.1080/17518253.2018.1440324 https://doi.org/10.1007/s40735-015-0024-x https://doi.org/10.1007/s40735-015-0024-x https://doi.org/10.1016/j.materresbull.2018.05.005 https://doi.org/10.1016/j.colsurfa.2021.126744 https://doi.org/10.1016/j.physb.2022.414008 https://doi.org/10.1021/jacs.5b03105 https://doi.org/10.1016/j.jece.2020.104833 https://doi.org/10.17344/acsi.2018.4181 https://doi.org/10.1016/j.jphotochem.2022.113848 https://doi.org/10.1002/slct.201700550 https://doi.org/10.1016/j.enconman.2022.116594 https://doi.org/10.1016/j.elecom.2022.107201 https://doi.org/10.1016/j.surfcoat.2020.126661 https://doi.org/10.1021/jacs.0c01683 p. amelia and j.gunlazuardi int. j. renew. energy dev 2023, 12(4), 702-710 | 710 issn: 2252-4940/© 2023. the author(s). published by cbiore free photocatalysts with 0.1% solar-to-ammonia energy conversion efficiency [research-article]. acs applied energy materials, 1(8), 4169–4177. https://doi.org/10.1021/acsaem.8b00829 singh, r., & dutta, s. (2018). synthesis and characterization of solar photoactive tio2 nanoparticles with enhanced structural and optical properties. advanced powder technology, 29(2), 211–219. https://doi.org/10.1016/j.apt.2017.11.005 sreedev, p., rakhesh, v., roshima, n. s., & shankar, b. (2019). preparation of zinc oxide thin films by silar method and its optical analysis. journal of physics: conference series, 1172(1). https://doi.org/10.1088/1742-6596/1172/1/012024 surahman, h., krisnandi, y. k., & gunlazuardi, j. (2015). modification of mixed structure tio2 nanoporousnanotube arrays with cds nanoparticle and their photoelectrochemical properties. jurnal sains materi indonesia, 16(3), 118–125. https://jurnal.batan.go.id/index.php/jsmi/article/view/4229 wang, j., fang, y., zhang, w., yu, x., wang, l., & zhang, y. (2021). tio2/biobr 2d-2d heterostructure via in-situ approach for enhanced visible-light photocatalytic n2 fixation. applied surface science, 567, 150623. https://doi.org/10.1016/j.apsusc.2021.150623 wang, l., wang, s., li, m., yang, x., li, f., xu, l., & zou, y. (2022). constructing oxygen vacancies and linker defects in mil-125 @tio2 for efficient photocatalytic nitrogen fixation. journal of alloys and compounds, 909, 164751. https://doi.org/10.1016/j.jallcom.2022.164751 wang, x. j., zhao, y., li, f. t., dou, l. j., li, y. p., zhao, j., & hao, y. j. (2016). a chelation strategy for in-situ constructing surface oxygen vacancy on {001} facets exposed biobr nanosheets. scientific reports, 6, 1–11. https://doi.org/10.1038/srep24918 yoo, h., kim, m., kim, y. t., lee, k., & choi, j. (2018). catalyst-doped anodic tio2 nanotubes: binder-free electrodes for (photo)electrochemical reactions. catalysts, 8(11), 1–25. https://doi.org/10.3390/catal8110555 yu, h., li, j., lin, y., wang, z., peng, h., tao, h., lai, x., & huang, y. (2023). biobr cluster spheres grown on conductive glass for a recyclable and efficient photocatalytic reactor. materials letters, 331, 133536. https://doi.org/10.1016/j.matlet.2022.133536 zhang, x., yang, h., zhang, b., shen, y., & wang, m. (2016). bioi-tio2 nanocomposites for photoelectrochemical water splitting. advanced materials interfaces, 3(1), 3–7. https://doi.org/10.1002/admi.201500273 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1021/acsaem.8b00829 https://doi.org/10.1016/j.apt.2017.11.005 https://doi.org/10.1088/1742-6596/1172/1/012024 https://jurnal.batan.go.id/index.php/jsmi/article/view/4229 https://doi.org/10.1016/j.apsusc.2021.150623 https://doi.org/10.1016/j.jallcom.2022.164751 https://doi.org/10.1038/srep24918 https://doi.org/10.3390/catal8110555 https://doi.org/10.1016/j.matlet.2022.133536 https://doi.org/10.1002/admi.201500273 international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (3), 459-466 |459 https://doi.org/10.14710/ijred.2023.51353 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id utilization of the spent catalyst as a raw material for rechargeable battery production: the effect of leaching time, type, and concentration of organic acids tabita kristina mora ayu panggabeana , ratna frida susantia , widi astutib , himawan tri bayu murti petrusc , anastasia prima kristijartia , kevin cleary wantaa,* adepartment of chemical engineering, faculty of industrial technology, parahyangan catholic university, bandung, indonesia bresearch unit for mineral technology, national research and innovation agency, tanjung bintang, indonesia cdepartment of chemical engineering, faculty of engineering, universitas gadjah mada, yogyakarta, indonesia abstract. this study examines the potential use of the spent catalyst as a raw material for rechargeable batteries. the spent catalyst ni/γ-al2o3 still contains relatively high amounts of nickel. this indicates the potential use of the spent catalyst to be leached and purified for synthesizing nickelbased compounds so that it can be applied to rechargeable battery cathodes. in this study, the spent catalyst leaching process employed four types of organic acids: citric acid, lactic acid, oxalic acid, and acetic acid. the spent catalyst was leached under atmospheric conditions and room temperature. organic acid concentrations were also varied at 0.1, 0.5, 1, and 2 m. the leaching process took place for 240 minutes, where sampling was conducted periodically at 30, 60, 120, 180, and 240 minutes. experimental results showed that ni (ii) and al (iii) ions were successfully leached to the maximum when using 2m citric acids at a leaching time of 240 minutes. the conditions succeeded in leaching ni (ii) and al (iii) ions of 357.8 and 1,975.4 ppm, respectively. organic acid, notably citric acid, has excellent potential for further development. citric acid, as a solvent, has the ability to leach metal ions with high recovery. in addition, this acid is categorized as an eco-friendly and green solvent compared to inorganic acid. thus, the leaching process can take place without harming the environment. keywords: batteries, organic acid, spent catalyst, nickel, aluminium @ the author(s). published by cbiore. this is an open-access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 31st dec 2022; revised: 3rd feb 2023; accepted: 24th feb 2023; available online: 24th march 2023 1. introduction the rapid development and production of electric vehicles worldwide have caused a rapid increase in rechargeable battery production. this has come about because electric vehicles require an energy storage system, like a rechargeable battery, to ensure that the vehicle can operate adequately (maddu et al., 2022; thiruvonasundari & deepa, 2021). this rechargeable battery comprises various components, such as an electrode (cathode and anode), electrolyte, and separator; those components are made of various compounds, for instance, metal alloys for the electrode, organic liquid for the electrolyte (liang et al., 2019; cheng et al., 2011). one of the critical elements in producing those batteries is nickel. in rechargeable batteries, nickel, in the form of an oxide, hydroxide, or other compounds, has a role as one of the materials that compose the cathode of rechargeable batteries (ash et al., 2020; kiani et al., 2010; yan et al., 2018; zhang et al., 2018). the choice of nickel as the cathode is due to the ability to provide higher storage capacity, higher energy density, and lower cost (benayed et al., 2021; cui et al., 2021; y. liu et al., 2011; xie et al., 2019). because of the importance of nickel in producing those batteries, it is * corresponding author email: kcwanta@unpar.ac.id (k.c. wanta) necessary to pay attention to supplying nickel for the rechargeable battery industry so that there is no scarcity or shortage of nickel sources in the future. nickel can be obtained from primary and secondary mineral sources (meshram et al., 2018; wanta et al., 2020a). so far, the demand for nickel has been dominated by ore or concentrate, classified as primary mineral sources. however, a number of these major mineral sources will eventually be depleted, so the ore or concentrate will only be available temporarily. to maintain the sustainability of this nickel source, secondary mineral sources, such as spent catalysts or spent batteries, need to be utilized well (garole et al., 2020). the secondary mineral source is feasible because its metal components are still at high concentrations. for example, the nickel content in the spent catalyst amounts to 13–38% (goel et al., 2009; sheik et al., 2013; wanta et al., 2021). this value is higher than the nickel content in ore or concentrate. thus, the effort of leaching nickel from the spent catalyst has excellent potential to be carried out, especially for nickel sources in producing these rechargeable batteries. nickel sources are one of many things to consider. the method of leaching nickel from those mineral sources also research article https://doi.org/10.14710/ijred.2023.51353 https://doi.org/10.14710/ijred.2023.51353 http://creativecommons/ mailto:kcwanta@unpar.ac.id https://orcid.org/0009-0007-5970-7712 https://orcid.org/0000-0002-4619-0300 https://orcid.org/0000-0001-9364-4291 https://orcid.org/0000-0001-9372-0784 https://orcid.org/0009-0006-4458-7845 https://orcid.org/0009-0009-3780-820x http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.51353&domain=pdf t.k.m.a. panggabean et al int. j. renew. energy dev 2023, 12(3), 459-466 |460 issn: 2252-4940/© 2023. the author(s). published by cbiore needs to be observed. like the recovery of other metals, the leaching method is a proven method capable of leaching metal with nearly perfect effectiveness (esmaeili et al., 2020; meshram et al., 2016; li et al., 2015). so far, the application of this method has been dominated by using inorganic acid solvents such as sulphuric acid (hosseini et al., 2017; ucyildiz & girgin, 2017), hydrochloric acid (le & lee, 2010; oediyani et al., 2019), and nitric acid (khalid & athraa, 2017; ramos-cano et al., 2016). however, this acid type has its drawbacks, primarily related to the environmental aspects involved (astuti et al., 2016). demarco (2020) revealed that the use of inorganic acids in the metal leaching process allows the release of so3, cl2, and nox gases, which pose a risk to the environment. furthermore, the acid waste generated after leaching is also harmful to the environment (li et al., 2010). as a result, greener solvents need to be applied so that the metal leaching process will not damage the environment in the future. one of the green solvents that are feasible for application is organic acid one. organic acids are classified as mild, generally biodegradable, and generate little or no environmental pollution (pathak et al., 2020). in addition, these organic acids can be produced using biotechnology, where the production of acids is carried out through fermentation pathways with the help of microorganisms (angumeenal & venkappayya, 2013; chen & nielsen, 2016). it is the forerunner of the emergence of a biological leaching process known as bioleaching (srichandan et al., 2019). metallurgically, organic acid solvents are desirable and promising for application. the ability to remove metal ions is reasonably tested because, as an acid, this organic acid can still provide hydrogen ions which have a prominent role in the mechanism of the leaching process. in several studies that have been conducted, the performance and effectiveness of organic acids as solvents have been well-tested (golmohammadzadeh et al., 2018; wanta et al., 2022). in the metal leaching process, liu et al. (2021) have proven that organic acids can effectively leach various metals, namely mn, cd, zn, and pb, from plant combustion ashes. their study also stated that citric acid was the best solvent for that particular process. another study by astuti et al. (2022) even explicitly proved that citric acid could operate better than inorganic acids in extracting lanthanum from the spent hydroprocessing catalyst. citric acid is not the only type of organic acid that can be applied in the leaching process. several similar studies yielded effective results in metal recovery from various secondary mineral sources when formic acid (arslanoğlu & yaraş, 2019), acetic acid (behera & parhi, 2016), and edta (gaber, 2019) were applied as solvents. this study focuses on the potential of the spent catalyst as a nickel source for rechargeable batteries. more specifically, leaching metal ions for this research study employs organic acid solvents. many leaching studies like this have already been conducted. however, in the metallurgical context, differences in the characteristics of raw materials lead to different mechanisms and treatments for the leaching process. each mineral resource has a uniqueness depending on its characteristics. a catalyst is an artificial mineral resource. it means that the catalyst comprises metal in the form of elements and compounds with a different mechanism to leach the metal ion. it is impossible for the metal in the form of elements to leach directly without additive compounds. therefore, if metal elements dominate the catalyst composition, it will affect the mechanism of the leaching process. thus, the potency of this spent catalyst needs to be studied further in terms of leaching the metal ion using organic acid. this study emphasizes sustainable processes and green technology. the parameters studied are leaching time, acid type, and concentration. the operating condition of this process was conducted under operating conditions classified as a simple and safe process, such as atmospheric and low temperatures. the main objective of this research study is to observe the effect of leaching time, organic acid types (citric, oxalic, acetic, and lactic acids), and acid concentrations in recovering ni (ii) and al (iii) ions contained in the spent catalysts ni/γ-al2o3. 2. material and methods 2.1 materials the spent catalyst ni/γ-al2o3 employed as a source of metal ions in this study originated at pt. petrokimia gresik, indonesia. this company is a fertilizer manufacturer; there is a reforming unit and applied ni/γ-al2o3 catalyst in that production. as a solvent, this leaching process applied four types of organic acids, namely citric acid (merck), oxalic acid (merck), acetic acid (merck), and lactic acid (merck). in addition, for analytical purposes, eriochrome cyanine r (ecr, merck) and dimethylglyoxime (dmg, merck) were used as complexing agents. all chemicals were dissolved using demineralized water. the spent catalyst was analyzed using an x-ray fluorescence (xrf, epsilon xle panalytical) instrument to determine the metal composition contained in the catalyst. the results of the analysis are presented in table 1. this table shows that there are three elements, namely aluminium (al), nickel (ni), and calcium (ca), which are the elements that dominate as well as compose this spent catalyst. these three elements in the catalyst are clearly visible in the mineral phases. mineral phase testing was also conducted by analyzing the spent catalyst using an x-ray diffraction instrument (xrd, bruker d8 advance, germany). the test results are presented in fig. 1. basically, this figure confirms the results of the xrf analysis and generates information about the mineral phase contained in the catalyst. the results of these characteristics still show that al, ni, and ca dominate the mineral phase contained. other components in table 1 define various heavy metals contained in the spent catalyst, such as titanium (ti), vanadium (v), zinc (zn), gallium (ga), strontium (sr), yttrium (y), zirconium (zr), molybdenum (mo), tellurium (te), and iridium (ir). the concentration of those elements is in part per million (ppm). this catalyst forms nickel in the pure nickel (nio) and nickel oxide (nio) crystalline phase. this leaching process is carried out without the use of additives that act as oxidizing agents. thus, the nickel that can be leached only comes from the mineral nio phase. fig. 1 xrd analysis results on the spent catalyst (raw material) t.k.m.a. panggabean et al int. j. renew. energy dev 2023, 12(3), 459-466 |461 issn: 2252-4940/© 2023. the author(s). published by cbiore table 1 xrf analysis results on the spent catalyst (raw material) component composition, %wt aluminium (al) 38.2 nickel (ni) 37.7 calcium (ca) 22.6 phosphorus (p) 0.4 iron (fe) 0.4 potassium (k) 0.2 silicon (si) 0.2 others 0.3 in the nio phase, nickel will not be leached in this process because the characteristic of the phase is stable and cannot be dissolved in water. nickel can only dissolve in water in the form of ni (ii) ions. this form will be produced from nio compounds that react with hydrogen ions from acids. meanwhile, for aluminium, the dominant form of aluminium that will react is aluminium oxide (al2o3). this compound will react with hydrogen ions from acids and produce al (iii) ions, which dissolve in water. 2.2 procedures this leaching process was carried out using a set of equipment consisting of a 1 l glass bottle (as an extractor), a horizontal shaker incubator, and a sampler. first, the spent catalyst was milled and sieved to obtain the spent catalyst powder with a size of less than 149 microns. seventy grams of spent catalyst powder were put into the extractor and mixed with 700 ml of the organic acid solution. after that, the extractor was conditioned on a shaker and was operated at 140 rpm. this leaching process was conducted at room temperature. the four types of organic acids were varied in this experiment. in addition, the concentration of each acid was varied and studied. for citric acid, acetic acid, and lactic acid, the studied concentration of those acid solutions was 0.1, 0.5, 1, and 2 m. meanwhile, the concentration of the oxalic acid solution was only varied at 0.1, 0.5, and 1 m due to the maximum solubility of oxalic acid in water. by way of comparison, sulphuric acid was also used in this leaching process. this experiment was conducted to observe the potential of organic acid as a solvent compared to inorganic acid, which dominates the metal ions leaching process. in this experiment, the sulphuric acid's leaching process was designed at an acid concentration of 0.1 m for 240 minutes. the sampling process for the metal ions analysis was carried out periodically at 30, 60, 120, 180, and 240 minutes. the sample was taken out from the extractor. the sample was separated between the solid and liquid phases using a centrifuge that operated at 6,000 rpm for 15 minutes. the formed supernatant (liquid phase) was taken to the amount of 10 ml and diluted to a solution volume of 20 ml. after that, the diluted sample was analyzed for the content of ni (ii) and al (iii) ions using a uv–vis spectrophotometer (mapada uv–6100 pc). the analysis process was carried out using complexing agents, namely dmg for ni (ii) ions and ecr for al (iii) ions (wanta et al., 2020b). 2.3 analysis of ni (ii) ions this analysis method was adapted from the article written by haar & westerveld (1948). ten milliliters of sample solution were mixed with 1 ml of 2 n sulphuric acid solution, 2 ml of 20% sodium tartrate solution, 8 ml of 5% potassium persulphate solution, and 0.5 ml of 1% dmg solution. to this mixture was added 5 n of sodium hydroxide solution until the total volume of the mixture amounted to 50 ml. this sample was subsequently moved to a cuvette and was analyzed using a uv–vis spectrophotometer at a wavelength of 560 nm. the blank solution was produced by mixing 10 ml of demineralized water, a complexing agent, and 5 n sodium hydroxide until the total volume amounted to 50 ml. 2.4 analysis of al (iii) ions this analysis method followed the procedure from the general chemistry laboratory, eastern michigan university (2018). ten milliliters of sample solution were mixed with 2 ml of 0.02 n sulphuric acid solution, 20 ml of buffer solution, 2 ml of the ascorbic acid solution, and 10 ml of ecr solution. to this mixture was added demineralized water until the total volume amounted to 100 ml. this sample was subsequently moved to a cuvette and was analyzed using a uv–vis spectrophotometer at a wavelength of 535 nm. the blank solution was produced by mixing 10 ml of demineralized water and a complexing agent until the total volume amounted to 100 ml. 3. results and discussion 3.1 effect of leaching time theoretically, the leaching time critically affects recovering metal ions in the spent catalyst. in this research study, the leaching time was varied from 0 to 240 minutes. the organic acid concentration was kept constant at 1 m. the experimental results on leaching time are presented in fig. 2. this figure shows that the increase in ni (ii) ion recovery was accompanied by an increase in leaching time. the results show that from 30 to 240 minutes, the leaching process increased the recovery of ni (ii) ions between 2.41–3.91 times for the four types of acids. the profile of the experimental results that occur in the recovery of ni (ii) ions also occurs for the recovery of al (iii) ions. this can be observed in fig. 3. the longer the leaching time, the bigger the opportunity that each molecule in the system remains in contact, especially the reactant molecules. as a result, the chemical reactions during the leaching process will react and produce more product molecules. fig. 2 effect of leaching time on the ni (ii) ions recovery at an acid concentration of 1 m 0 50 100 150 200 250 0 50 100 150 200 250 300 350 c on ce nt ra tio n of n i(i i) io ns (p pm ) time (minutes) citric acid lactic acid oxalic acid acetic acid t.k.m.a. panggabean et al int. j. renew. energy dev 2023, 12(3), 459-466 |462 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 3 effect of leaching time on the al (iii) ions recovery at an acid concentration of 1 m time affects not only the step of chemical reactions but also molecular diffusion. the diffusion step experienced by each reactant and product molecule also takes time to move. the reactant molecules will diffuse from the bulk liquid to the surface of the liquid and the reaction site. on the other hand, the product molecules will move and diffuse from the reaction site to the liquid body. these two steps of diffusion will take place simultaneously so that in the pores/paths that tend to be narrow, all molecules will collide with each other. thus, this diffusion step will require a long time to produce an optimal leaching process. fig. 2 and 3 provide an overview of the effect of time on metal ion recovery. these two figures explain how the behavior occurs during the leaching process. in the recovery of al (iii) ions (fig. 3), there was a very significant increase in aluminium recovery in the first 30 minutes of the leaching process. after 30 minutes, the tendency to increase metal ion concentration occurred gradually. this phenomenon indicates that the leaching rate is swift at the beginning of the leaching time. this condition is related to the role of aluminium in catalyst production. aluminium, in the form of γ-al2o3 (γ-alumina), supports the catalyst (trueba & trasatti, 2005). it causes aluminium will occupy most of the area on the catalyst, especially on the surface of the catalyst powder. as a result, retrieving aluminium will be much easier and require faster leaching times, especially at the early stage of the leaching process. the opposite condition occurs in the uptake of ni (ii) ions. fig. 2 shows that the concentration of ni (ii) ions in the bulk solution increases gradually. it proves that the recovery of ni (ii) ions takes longer. as the active site, nickel is scattered in a deeper position, so this recovery process takes longer, which is especially necessary for the diffusion process, both reactant molecules and product molecules. in addition, the experimental results shown in fig. 2 and 3 show the potential to obtain more and more metal ions. in this condition, it can be observed that there is no sign of equilibrium in the two graphs. this means that the maximum time used in this experiment differs from the time that gives optimal results. one of the main reasons this phenomenon occurs is that the temperature used in this research experiment is room temperature. using this specific temperature will yield a slower process rate and result in a longer leaching time to obtain optimal leaching results. in this study, the main focus of learning is on room temperature. in the future, studies related to leaching temperature deserve to be considered and studied in greater depth so that its effect on the length of time for the leaching process can also be observed. 3.2 effect of the type of organic acid in the leaching process, the choice of solvent is necessary. one type of solvent that can be used is an acid solution. acid solutions are solvents that have been tested for leaching metal ions. it is due to hydrogen ions (h+) which can be produced from the dissociation process when the acid is dissolved in water. this hydrogen ion will displace the metal ion from its anion, such as the oxide ion, so the metal ion can dissolve and form molecular compounds according to the acid's anion. the chemical reaction that occurs during the leaching process using organic acid is as follows (behera & mulaba-bafubiandi, 2015; golmohammadzadeh et al., 2018; simate et al., 2010). organic acid dissociation step citric acid : c6h8o7(aq) ↔ (c6h5o7)3–(aq) + 3h+(aq) (1) lactic acid : c3h6o3(aq) ↔ (c3h5o3)3–(aq) + h+(aq) (2) oxalic acid : c2h2o4(aq) ↔ (c2o4)2–(aq) + 2h+(aq) (3) acetid acid : c2h4o2(aq) ↔ (c2h3o2)– (aq) + h+(aq) (4) proton attack step mxoy(s) + 2yh+(aq) → xm2y/x+(aq) + yh2o(l) (5) where m stands for the metal element. complexation or chelation step xm2y/x+(aq) + az–(aq) ↔ mza2y/x(aq) (6) where a is an anion from the dissociation step, for instance, citrate, lactate, oxalate, acetate ions. this research study focuses on using organic acids as a solvent for metal ions in the spent catalyst. four types of organic acids were studied for their effect on metal ion recovery: citric acid, lactic acid, oxalic acid, and acetic acid. the experimental results can be seen in fig. 4, where the results are obtained when the acid concentration is 1 m and the extraction process takes place for 240 minutes. the experimental result shown in fig. 4 provides information that the order of the types of organic acids from the best to the worst for leaching ni (ii) and al (iii) ions are citric acid – lactic acid – oxalic acid – acetic acid. citric acid is the most suitable solvent due to the properties of the hydrogen ions. equation (1) shows that hydrogen ions are produced in greater quantities than the other three organic acids. fig. 4 effect of leaching time on the metal ions recovery at an acid concentration of 1 m for 240 minutes 0 50 100 150 200 250 0 100 200 300 400 500 600 700 800 900 1000 c on ce nt ra tio n of a l(i ii) io ns (p pm ) time (minutes) citric acid lactic acid oxalic acid acetic acid 344.64 321.68 305.92 182.56 953.13 873.75 807.96 620.85 citric acid lactic acid oxalic acid acetic acid 0 100 200 300 400 500 600 700 800 900 1000 c o n ce n tr a tio n o f m e ta l i o n s (p p m ) type of acid ni(ii) ions al(iii) ions t.k.m.a. panggabean et al int. j. renew. energy dev 2023, 12(3), 459-466 |463 issn: 2252-4940/© 2023. the author(s). published by cbiore as a tricarboxylic acid, citric acid has the ability to donate three protons (h+) per molecule when it dissociates in water. this means that citric acid also has three pka values and can be observed in the following equation (golmohammadzadeh et al., 2018). h3c6h5o7(aq) = h2c6h5o7–(aq) + h+(aq) (pka1 = 2.79) (7) h2c6h5o7– (aq) = hc6h5o72–(aq) + h+(aq) (pka2 = 4.30) (8) hc6h5o72– (aq) = c6h5o73–(aq) + h+(aq) (pka3 = 5.65) (9) the more hydrogen ions, the greater the probability of a proton attack. thus, this organic acid can perform the metal leaching process better than other acids. instead of this, the pka value of each acid can explain the phenomenon in the other three organic acids. the pka values for lactic acid, oxalic acid, and acetic acid were 3.86, 4.19, and 4.76, respectively (golmohammadzadeh et al., 2018). the lower the pka, the stronger the acidity. this indicates that the ability to donate protons (h+ ions) in the system is also improving. this phenomenon explains why lactic acid can leach more ni (ii) and al (iii) ions compared to oxalic acid and acetic acid. however, the use of oxalic acid leads to another phenomenon where it also forms nickel oxalate precipitate. this phenomenon follows the following chemical reaction equation (allen, 1953). 2niso4(aq) + 2h2c2o4(aq) ↔ [ni(c2o4)2]– ni2+(aq) + 2h2so4(aq) (10) [ni(c2o4)2]– ni2+(aq) → 2nic2o4.2h2o(s) (11) as evidence of the formation of those compounds, testing of the residual solids resulting from leaching with 1 m oxalic acid solution was carried out to determine the mineral phase that occurred. the results of the analysis are presented in fig. 5. the formation of nickel oxalate compounds during the leaching process is undesirable. nickel oxalate compounds have very low solubility in water, so the nickel oxalate precipitation process will also coincide with the leaching process. indeed, this phenomenon looks advantageous because it does not require precipitation or product crystallization at a later stage. however, this condition will be detrimental because the formed precipitate compound is mixed directly with the spent catalyst in the system. this will lead to new problems separating spent catalyst residues and nickel oxalate products. this separation process can be complicated or even classified as an impossible thing to accomplish. thus, oxalic acid as a solvent in the metal leaching process is not recommended for use as a solvent. fig. 5 xrd pattern on residue solid from leaching process using oxalic acid 1 m fig. 6 effect of acid concentration on the ni (ii) ions recovery for 240 minutes 3.3 effect of organic acid concentration in studying the effect of organic acid concentrations, the utilization of citric acid, lactic acid, and acetic acid was carried out at concentrations of 0.1, 0.5, 1, and 2 m. meanwhile, oxalic acid could not be carried out at concentrations of 2 m because of the solubility of that particular compound in water. the experimental results are presented in fig. 6. these results showed similar results for the four types of organic acids. a higher concentration of acid will produce a more significant number of hydrogen ions. the hydrogen ions will attack the metal compound, for example, nickel and aluminium oxide, on the catalyst to produce water-soluble ni (ii) ions. thus, a high acid concentration will increase the nickel's recovery in the solution. the phenomenon that occurs in ni (ii) ions is the same as in the leaching al (iii) ions. the experimental results for al (iii) ions can be observed in fig. 7. 3.4 organic acid vs. sulphuric acid employed for the leaching process in this section, it between organic acid and sulphuric acid will be compared as a solvent for this leaching process. in hydrometallurgy, sulphuric acid is the most widely used type of solvent. the results of the comparison between the five types of acids are presented in fig. 8, where the concentration used was 0.1 m. sulphuric acid as a solvent gave much better leaching results than all the organic acids used in previous studies. as previously explained, the leaching process is strongly influenced by the strength of the acid, as seen from the number of hydrogen ions in the system. sulphuric acid in water dissociates entirely according to the following chemical reaction equation (vynnycky & assunção, 2020). h2so4(aq) → 2h+(aq) + so42–(aq) (12) compared with the four organic acids (equation 1–4), the hydrogen ion in sulphuric acid will ideally form according to stoichiometry. this is because the reaction is irreversible. on the other hand, in organic acids, the dissociation reaction is reversible, indicating that there is a balance factor in the formation of hydrogen ions. this results in citric acid, although it forms three hydrogen ions; in terms of the strength of sulphuric acid, which produces two hydrogen ions, it still has a higher level of acidity. as a result, in the leaching process with the same acid concentration, the performance of the sulphuric acid solvent is still superior. 0.0 0.5 1.0 1.5 2.0 50 75 100 125 150 175 200 225 250 275 300 325 350 375 n i(i i) io ns c on ce nt ra tio n (p pm ) acid concentration (m) citric acid lactic acid oxalic acid acetic acid t.k.m.a. panggabean et al int. j. renew. energy dev 2023, 12(3), 459-466 |464 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 7 effect of acid concentration on the al (iii) ions recovery for 240 minutes fig. 8 the performance of organic and sulphuric acid in the leaching process of metal ions at an acid concentration of 0.1 m for 240 minutes in addition, the use of organic acids (weak acids) raises other phenomena that will affect the mechanism of the metal leaching process. equation (6) shows how complex compounds (chelates/ligands) are formed when metal ions bind to anions from organic acids. meanwhile, this phenomenon does not occur using inorganic acids (strong acids) during the leaching process. the complex compound has a larger molecular size than the product molecules formed using inorganic acids (wanta et al., 2022). this condition will affect the mechanism and the leaching process's total rate. large molecules will hinder the diffusion process of reactant and product molecules inside the solid. thus, the rate of the leaching process with organic acids will be controlled by the diffusion step of the product molecules in the solid. this experiment needs to be designed by applying various kinetic models and will be the focus of further research in our future studies. even though sulphuric acid gives better results, using organic acids does not mean they do not have the potential and opportunities for further development. moreover, this organic acid is an environmentally friendly solvent compared to sulphuric acid. citric acid is a type of organic acid whose leaching behavior is similar to sulphuric acid. astuti et al. (2016) conducted a comparative study between sulphuric acid and organic acids in the leaching process of nickel from indonesian saprolite ore. in their study, the citric acid solvent was able to match the performance of sulphuric acid in their leaching process. this condition is based on the hope that organic acids, especially citric acid, deserve to be considered a suitable solvent for leaching metal ions. indeed, several efforts need to be made to improve the performance of the leaching process using organic acids. the improvement effort in question can be in the form of increasing the operating temperature, reducing particle size, extending the operating time, and many other efforts. 4. conclusion spent catalyst has excellent potential to be explored and utilized for its nickel resources as part of the raw material for manufacturing rechargeable battery cathodes. the metal from the catalyst can be leached through an extraction process using organic acid solvents. this experiment proves that citric acid, lactic acid, oxalic acid, and acetic acid can be applied as solvents. these four acids are believed to be green solvents compared to sulphuric acid, which is more widely used as a solvent. based on the experimental results, citric acid is the solvent that gives the best leaching results compared to the other three organic acids. when the leaching process took place for 4 hours, the 2 m citric acid solution succeeded in leaching ni (ii) and al (iii) ions of 357.8 and 1,975.4 ppm, respectively. the recovery of metal ions with citric acid results are still lower than the ones with sulphuric acid as a solvent. however, citric acid can still be developed further to have the same or better performance than sulphuric acid. various alternatives for manipulating the operating parameters of the leaching process, such as temperature, particle size, the addition of additives, and others, can be conducted to increase the process rate. thus, the extraction of metal ions from the spent catalyst is suitable for further study to obtain maximum results and meet the demand for metal ions as a raw material for producing rechargeable batteries. the results obtained from this research study are expected to form a solution in supplying the need for nickel as a raw material for producing rechargeable batteries. acknowledgments the authors gratefully acknowledge the institute for research and community service, parahyangan catholic university (lppm unpar) for their financial support. the authors would like to thank brin’s science services and mr. gelar panji gemilar from pt. petrokimia gresik because of their support, especially the analysis and raw materials. author contributions: t.k.m.a.p.: conceptualization, methodology, formal analysis, writing—original draft; r.f.s.: supervision, validation, writing—review and editing; w.a.: formal analysis, writing—review and editing; h.t.b.m.p.: writing—review and editing, resources; a.p.k.: conceptualization, methodology, supervision, writing—review and editing, project administration; k.c.w.: conceptualization, methodology, supervision, validation, writing—original draft, project administration. funding: this research was funded by institute for research and community service, parahyangan catholic university. conflicts of interest: the authors declare no conflict of interest. references allen, j. a. (1953). the precipitation of nickel oxalate. j. phys. chem., 57(7), 715–716. https://doi.org/10.1021/j150508a027 angumeenal, a. r., & venkappayya, d. (2013). an overview of citric acid production. lwt food science and technology, 50(2), 367– 370; https://doi.org/10.1016/j.lwt.2012.05.016 0.0 0.5 1.0 1.5 2.0 200 400 600 800 1000 1200 1400 1600 1800 2000 a l(i ii) io ns c on ce nt ra tio n (p pm ) acid concentration (m) citric acid lactic acid oxalic acid acetic acid 237.6 212.01 155.92 74.54 361.03342.28 306.66 221.87 168.39 932.8 citric acid lactic acid oxalic acid acetic acid sulphuric acid 0 100 200 300 400 500 600 700 800 900 1000 c on ce nt ra tio n of m et al io ns (p pm ) type of acid ni(ii) ions al(iii) ions https://doi.org/10.1021/j150508a027 https://doi.org/10.1016/j.lwt.2012.05.016 t.k.m.a. panggabean et al int. j. renew. energy dev 2023, 12(3), 459-466 |465 issn: 2252-4940/© 2023. the author(s). published by cbiore arslanoğlu, h., & yaraş, a. (2019). recovery of precious metals from spent mo–co–ni/al2o3 catalyst in organic acid medium: process optimization and kinetic studies. petroleum science and technology, 37(19), 2081–2093; https://doi.org/10.1080/10916466.2019.1618867 ash, b., nalajala, v. s., popuri, a. k., subbaiah, t., & minakshi, m. (2020). perspectives on nickel hydroxide electrodes suitable for rechargeable batteries: electrolytic vs. chemical synthesis routes. nanomaterials, 10(9), 1–22; https://doi.org/10.3390/nano10091878 astuti, w., hirajima, t., sasaki, k., & okibe, n. (2016). comparison of effectiveness of citric acid and other acids in leaching of lowgrade indonesian saprolitic ores. minerals engineering, 85, 1–16; https://doi.org/10.1016/j.mineng.2015.10.001 astuti, w., mufakhir, f. r., setiawan, f. a., wanta, k. c., & petrus, h. t. b. m. (2022). leaching characteristics of lanthanum from a secondary resource using inorganic and organic acids: emphasizing the citric acid kinetics. circular economy and sustainability; https://doi.org/10.1007/s43615-022-00183-9 behera, s. k., & mulaba-bafubiandi, a. f. (2015). advances in microbial leaching processes for nickel extraction from lateritic minerals a review. korean journal of chemical engineering, 32(8), 1447– 1454; https://doi.org/10.1007/s11814-015-0085-z behera, s. s., & parhi, p. k. (2016). leaching kinetics study of neodymium from the scrap magnet using acetic acid. separation and purification technology, 160, 59–66. https://doi.org/10.1016/j.seppur.2016.01.014 benayed, a., gasbaoui, b., bentouba, s., & soumeur, m. a. (2021). movement of a solar electric vehicle controlled by ann-based dtc in hot climate regions. international journal of renewable energy development, 10(1), 61–70; https://doi.org/10.14710/ijred.2021.18596 chen, y., & nielsen, j. (2016). biobased organic acids production by metabolically engineered microorganisms. current opinion in biotechnology, 37, 165–172; https://doi.org/10.1016/j.copbio.2015.11.004 cheng, f., liang, j., tao, z., & chen, j. (2011). functional materials for rechargeable batteries. advanced materials, 23(15), 1695–1715; https://doi.org/10.1002/adma.201003587 cui, z., xie, q., & manthiram, a. (2021). zinc-doped high-nickel, lowcobalt layered oxide cathodes for high-energy-density lithiumion batteries. acs applied materials and interfaces, 13(13), 15324– 15332; https://doi.org/10.1021/acsami.1c01824 demarco, j., cadore, j. s., veit, h. m., madalosso, h. b., tanabe, e. h., & bertuol, d. a. (2020). leaching of platinum group metals from spent automotive catalysts using organic acids. minerals engineering, 159; https://doi.org/10.1016/j.mineng.2020.106634 esmaeili, m., rastegar, s. o., beigzadeh, r., & gu, t. (2020). ultrasoundassisted leaching of spent lithium ion batteries by natural organic acids and h2o2. chemosphere, 254, 126670; https://doi.org/10.1016/j.chemosphere.2020.126670 gaber, m. a. f. m. (2019). extraction of nickel from spent catalyst of primary reformer. recent advances in petrochemical science, 6(3), 64–69; https://doi.org/10.19080/rapsci.2019.06.555690 garole, d. j., hossain, r., garole, v. j., sahajwalla, v., nerkar, j., & dubal, d. p. (2020). recycle, recover and repurpose strategy of spent li-ion batteries and catalysts: current status and future opportunities. chemsuschem, 13(12), 3079–3100; https://doi.org/10.1002/cssc.201903213 general chemistry laboratory, eastern michigan university. (2018). experiment 3: analysis of aluminum (iii) in water. https://canvas.emich.edu/courses/48894 goel, s., pant, k. k., & nigam, k. d. p. (2009). extraction of nickel from spent catalyst using fresh and recovered edta. journal of hazardous materials, 171(1–3), 253–261; https://doi.org/10.1016/j.jhazmat.2009.05.131 golmohammadzadeh, r., faraji, f., & rashchi, f. (2018). recovery of lithium and cobalt from spent lithium ion batteries (libs) using organic acids as leaching reagents: a review. resources, conservation and recycling, 136, 418–435; https://doi.org/10.1016/j.resconrec.2018.04.024 haar, k. t., & westerveld, w. (1948). the colorimetric determination of nickel, as ni(4) dimethylglyoxime. recueil, 67, 71–81; https://doi.org/10.1002/recl.19480670112 hosseini, s. a., raygan, s., rezaei, a., & jafari, a. (2017). leaching of nickel from a secondary source by sulfuric acid. journal of environmental chemical engineering, 5(4), 3922–3929; https://doi.org/10.1016/j.jece.2017.07.059 khalid, m. m., & athraa, b.a. (2017). experimental study on factors affecting the recovery of nickel from spent catalyst. journal of powder metallurgy & mining, 6(1); https://doi.org/10.4172/21689806.1000146 kiani, m. a., mousavi, m. f., & ghasemi, s. (2010). size effect investigation on battery performance: comparison between microand nano-particles of β-ni(oh)2 as nickel battery cathode material. journal of power sources, 195(17), 5794–5800; https://doi.org/10.1016/j.jpowsour.2010.03.080 le, m. n. l. & lee, m. s. (2020). separation of al(iii), mo(vi), ni(ii), and v(v) from model hydrochloric acid leach solutions of spent petroleum catalyst by solvent extraction. journal of chemical technology & biotechnology, 95(11), 2886–2897; https://doi.org/10.1002/jctb.6448 li, l., ge, j., chen, r., wu, f., chen, s., & zhang, x. (2010). environmental friendly leaching reagent for cobalt and lithium recovery from spent lithium-ion batteries. waste management, 30(12), 2615–2621; https://doi.org/10.1016/j.wasman.2010.08.008 li, l., qu, w., zhang, x., lu, j., chen, r., wu, f., & amine, k. (2015). succinic acid-based leaching system: a sustainable process for recovery of valuable metals from spent li-ion batteries. journal of power sources, 282, 544–551; https://doi.org/10.1016/j.jpowsour.2015.02.073 liang, y., zhao, c.-z., yuan, h., chen, y., zhang, w., huang, j.-q., yu, d., liu, y., titirici, m.-m., chueh, y.-l., yu, h., & zhang, q. (2019). a review of rechargeable batteries for portable electronic devices. infomat, 1(1), 6–32; https://doi.org/10.1002/inf2.12000 liu, y., pan, h., gao, m., & wang, q. (2011). advanced hydrogen storage alloys for ni/mh rechargeable batteries. journal of materials chemistry, 21(13), 4743–4755; https://doi.org/10.1039/c0jm01921f liu, r., tian, z., cheng, h., zhou, h., & wang, y. (2021). organic acid leaching was an efficient approach for detoxification of metalcontaining plant incineration ash. environmental science and pollution research, 28(25), 32721–32732; https://doi.org/10.1007/s11356-021-13027-0 maddu, a., sulaeman, a. s., wahyudi, s. t., & rifai, a. (2022). enhancing ionic conductivity of carboxymethyl cellulose-lithium perchlorate with crosslinked citric acid as solid polymer electrolytes for lithium polymer batteries. international journal of renewable energy development, 11(4), 1002–1011; https://doi.org/10.14710/ijred.2022.40090 meshram, p., pandey, b. d., & mankhand, t. r. (2016). process optimization and kinetics for leaching of rare earth metals from the spent ni-metal hydride batteries. waste management, 51, 196–203; https://doi.org/10.1016/j.wasman.2015.12.018 meshram, p., abhilash, & pandey, b. d. (2018). advanced review on extraction of nickel from primary and secondary sources. in mineral processing and extractive metallurgy review, 40(3), 157– 193; https://doi.org/10.1080/08827508.2018.1514300 oediyani, s., ariyanto, u., & febriana, e. (2019). effect of concentration, agitation, and temperature of pomalaa limonitic nickel ore leaching using hydrochloric acid. iop conf. series: materials science and engineering, 478(012013), 1–8; https://doi.org/10.1088/1757-899x/478/1/012013 pathak, a., vinoba, m., & kothari, r. (2020). emerging role of organic acids in leaching of valuable metals from refinery-spent hydroprocessing catalysts, and potential techno-economic challenges: a review. critical reviews in environmental science and technology, 51(1), 1–43; https://doi.org/10.1080/10643389.2019.1709399 ramos-cano, j., gonzález-zamarripa, g., carrillo-pedroza, f. r., soriaaguilar, m. d. j., hurtado-macías, a., & cano-vielma, a. (2016). kinetics and statistical analysis of nickel leaching from spent catalyst in nitric acid solution. international journal of mineral processing, 148, 41–47; https://doi.org/10.1016/j.minpro.2016.01.006 sheik, a. r., ghosh, m. k., sanjay, k., subbaiah, t., & mishra, b. k. (2013). dissolution kinetics of nickel from spent catalyst in nitric https://doi.org/10.1080/10916466.2019.1618867 https://doi.org/10.3390/nano10091878 https://doi.org/10.1016/j.mineng.2015.10.001 https://doi.org/10.1007/s43615-022-00183-9 https://doi.org/10.1007/s11814-015-0085-z https://doi.org/10.1016/j.seppur.2016.01.014 https://doi.org/10.14710/ijred.2021.18596 https://doi.org/10.1016/j.copbio.2015.11.004 https://doi.org/10.1002/adma.201003587 https://doi.org/10.1021/acsami.1c01824 https://doi.org/10.1016/j.mineng.2020.106634 https://doi.org/10.1016/j.chemosphere.2020.126670 https://doi.org/10.19080/rapsci.2019.06.555690 https://doi.org/10.1002/cssc.201903213 https://canvas.emich.edu/courses/48894 https://doi.org/10.1016/j.jhazmat.2009.05.131 https://doi.org/10.1016/j.resconrec.2018.04.024 https://doi.org/10.1002/recl.19480670112 https://doi.org/10.1016/j.jece.2017.07.059 https://doi.org/10.4172/2168-9806.1000146 https://doi.org/10.4172/2168-9806.1000146 https://doi.org/10.1016/j.jpowsour.2010.03.080 https://doi.org/10.1002/jctb.6448 https://doi.org/10.1016/j.wasman.2010.08.008 https://doi.org/10.1016/j.jpowsour.2015.02.073 https://doi.org/10.1002/inf2.12000 https://doi.org/10.1039/c0jm01921f https://doi.org/10.1007/s11356-021-13027-0 https://doi.org/10.14710/ijred.2022.40090 https://doi.org/10.1016/j.wasman.2015.12.018 https://doi.org/10.1080/08827508.2018.1514300 https://doi.org/10.1088/1757-899x/478/1/012013 https://doi.org/10.1080/10643389.2019.1709399 https://doi.org/10.1016/j.minpro.2016.01.006 t.k.m.a. panggabean et al int. j. renew. energy dev 2023, 12(3), 459-466 |466 issn: 2252-4940/© 2023. the author(s). published by cbiore acid medium. journal of the taiwan institute of chemical engineers, 44(1), 34–39; https://doi.org/10.1016/j.jtice.2012.08.003 simate, g. s., ndlovu, s., & walubita, l. f. (2010). the fungal and chemolithotrophic leaching of nickel laterites challenges and opportunities. hydrometallurgy, 103(1–4), 150–157; https://doi.org/10.1016/j.hydromet.2010.03.012 srichandan, h., mohapatra, r. k., parhi, p. k., & mishra, s. (2019). bioleaching approach for extraction of metal values from secondary solid wastes: a critical review. hydrometallurgy, 189; https://doi.org/10.1016/j.hydromet.2019.105122 thiruvonasundari, d., & deepa, k. (2021). evaluation and comparative study of cell balancing methods for lithium-ion batteries used in electric vehicles. international journal of renewable energy development, 10(3), 471–479; https://doi.org/10.14710/ijred.2021.34484 trueba, m., & trasatti, s. p. (2005). γ-alumina as a support for catalysts: a review of fundamental aspects. european journal of inorganic chemistry, 17, 3393–3403; https://doi.org/10.1002/ejic.200500348 ucyildiz, a., & girgin, i. (2017). high pressure sulphuric acid leaching of lateritic nickel ore. physicochemical problems of mineral processing, 53(1), 475–488; https://doi.org/10.5277/ppmp170137 vynnycky, m., & assunção, m. (2020). on the significance of sulphuric acid dissociation in the modelling of vanadium redox flow batteries. journal of engineering mathematics, 123(1), 173–203; https://doi.org/10.1007/s10665-020-10061-8 wanta, k. c., astuti, w., perdana, i., & petrus, h. t. b. m. (2020). kinetic study in atmospheric pressure organic acid leaching: shrinking core model versus lump model. minerals, 10(7), 1–10; https://doi.org/10.3390/min10070613 wanta, k. c., gunawan, w. t., susanti, r. f., gemilar, g. p., petrus, h. t. b. m., & astuti, w. (2020). subcritical water as a solvent for extraction of nickel and aluminum ions from reforming spent catalysts. iop conference series: materials science and engineering, 742(1); https://doi.org/10.1088/1757-899x/742/1/012025 wanta, k. c., natapraja, e. y., susanti, r. f., gemilar, g. p., astuti, w., & petrus, h. t. b. m. (2021). increasing of metal recovery in leaching process of spent catalyst at low temperature: the addition of hydrogen peroxide and sodium chloride. metalurgi, 36(2); https://doi.org/10.14203/metalurgi.v36i2.591 wanta, k. c., astuti, w., petrus, h. t. b. m., & perdana, i. (2022). product diffusion-controlled leaching of nickel laterite using low concentration citric acid leachant at atmospheric condition. international journal of technology, 13(2), 410–421; https://doi.org/10.14716/ijtech.v13i2.4641 xie, q., li, w., & manthiram, a. (2019). a mg-doped high-nickel layered oxide cathode enabling safer, high-energy-density li-ion batteries. chemistry of materials, 31(3), 938–946; https://doi.org/10.1021/acs.chemmater.8b03900 yan, p., zheng, j., liu, j., wang, b., cheng, x., zhang, y., sun, x., wang, c., & zhang, j. g. (2018). tailoring grain boundary structures and chemistry of ni-rich layered cathodes for enhanced cycle stability of lithium-ion batteries. nature energy, 3(7), 600–605; https://doi.org/10.1038/s41560-018-0191-3 zhang, h., wang, r., lin, d., zeng, y., & lu, x. (2018). ni-based nanostructures as high-performance cathodes for rechargeable ni−zn battery. chemnanomat, 4(6), 525–536; https://doi.org/10.1002/cnma.201800078 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1016/j.jtice.2012.08.003 https://doi.org/10.1016/j.hydromet.2010.03.012 https://doi.org/10.1016/j.hydromet.2019.105122 https://doi.org/10.14710/ijred.2021.34484 https://doi.org/10.1002/ejic.200500348 https://doi.org/10.5277/ppmp170137 https://doi.org/10.1007/s10665-020-10061-8 https://doi.org/10.3390/min10070613 https://doi.org/10.1088/1757-899x/742/1/012025 https://doi.org/10.14203/metalurgi.v36i2.591 https://doi.org/10.14716/ijtech.v13i2.4641 https://doi.org/10.1021/acs.chemmater.8b03900 https://doi.org/10.1038/s41560-018-0191-3 https://doi.org/10.1002/cnma.201800078 utilization of the spent catalyst as a raw material for rechargeable battery production: the effect of leaching time, type, and concentration of organic acids tabita kristina mora ayu panggabeana, ratna frida susantia, widi astutib, himawan tri bayu murti petrusc, anastasia prima kristijartia, kevin cleary wantaa,0f 1. introduction international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (2), 366-374 | 366 https://doi.org/10.14710/ijred.2023.49910 issn: 2252-4940/© 2022.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id design of optical collimator system for vehicle speed gun using non-imaging optics le minh nhuta , thanh-tuan phama* , tien-dung tranb , vu dinh huana, seoyong shinc† a faculty of vehicle and energy engineering, ho chi minh city university of technology and education – hcmute, ho chi minh city, vietnam b energy management department, chez bong co. ltd, ho chi minh city, vietnam c department of information and communication engineering, myongji university, yongin, korea abstract. vehicle speed guns are usually used in normal sunlight conditions (daytime). if we want to use vehicle speed guns in low light conditions (nighttime), the illuminator is needed to provide sufficient light for the vehicle speed gun to take photos. the illuminator must fulfill two requirements: (i) using the infrared wavelength to ensure that the driver is not startled by dazzling eyes by the illuminator of the proposed speed gun system and (ii) high energy efficiency to make the illuminator compact leading to the use a small battery system to improve the portable of the proposed vehicle speed gun. in this study, an illuminator using a collimator system designed by using non-imaging optics is introduced. leds with infrared wavelength are chosen from the library of lighttoolstm, the structure of collimated is designed to transfer the illumination from the leds array to a square area of 3x3 m2 to cover the vehicle to detect the vehicle number plate. the design process is built based on the conservation of optical path length in the matlab program. after that, the designed collimator is simulated in lighttoolstm software. the promising results of the simulation in lighttoolstm show that the collimator can efficiently transfer light from the led array to the target area with a uniformity of about 70 % and optical efficiency of about 80 %. keywords: illumination, collimator, vehicles speed gun, lighttools, non-imaging optics, matlab programming, non-imaging optics. @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 15th oct 2022; revised: 7th january 2023; accepted: 2nd feb 2023; available online: 10th feb 2023 1. introduction a vehicle speed gun operates based on the principle of the doppler effect; thus, it can work in any light condition (hamelmann et al 2019). however, a vehicle speed gun must be used with a camera to take photos serving as evidence. therefore, a vehicle speed gun is usually used in the daytime. if it is used at night, a flashlight is required for the camera. the flashlight for the speed gun must use infrared illumination with wavelengths longer than 750 nm to guarantee the safe movement of drivers. in addition, a flashlight should have high power to enable the camera to take photos from a long distance, around 70 m to 100 m. this requirement leads to the need of using a large battery for the flashlight. a battery system that is not compact will hinder the applicability of the speed gun system as a portable mobility device. on the other hand, a small battery with high mobility does not have enough energy for a high-power illumination instrument of a speed gun. there is an effective solution for this problem which is to use led (light-emitting diode) because of its great light intensity with small energy consumption. as a result, a compact battery system working for a long time is available, thus improving the mobility and performance of vehicle speed guns. leds typically emit light with a beam view angle of about 120-degree angle if * corresponding author email: tuanpt@hcmute.edu.vn (t.t. pham) † corresponding author email: sshin@mju.ac.kr (s. shin) no optical lens or reflector is used (liu et al 2019). therefore, a led array for illumination is just effective only if it is designed and used with accordant optical lenses with special application. in this case, if optical lenses are not suitable for leds array, light from leds will not spread and distribute into the target area, meaning a waste of energy. to design a collimator for a led array, non-imaging optics (winston et al 2005, chaves et al 2008) is a suitable method compared to imaging optics thanks to its simplicity and flexibility. non-imaging optics are completely suitable for applications such as illumination (koshel et al 2012) and aiming for transferring radiation energy (0’gallagher et al 2008) to the target area. especially, non-imaging is usually used for designs that need a high tolerance to ignore the effects of error in the assembly and production processes of optical elements (winston et al 2018). non-imaging optics was first invented by winston et al (2005) and then developed by minano et al (2015), and benitez (2004) for many applications in illumination and solar energy (dross et al 2004, gutierrez et al 1996, chong et al 2009). in non-imaging optics, there are two main methods for optical design such as flow line, and simultaneously multiple surfaces (sms) (chaves et al 2008, benitez et al 2004). the flow line is typically used to design compound parabolic concentrators (cpcs) while the sms method, which is modern research article https://doi.org/10.14710/ijred.2023.49910 https://doi.org/10.14710/ijred.2023.49910 mailto:tuanpt@hcmute.edu.vn mailto:sshin@mju.ac.kr https://orcid.org/0000-0002-0603-8689 https://orcid.org/0000-0002-2682-1427 https://orcid.org/0000-0003-3168-6941 https://orcid.org/0000-0003-3746-6835 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.49910&domain=pdf l.m. nhut et al int. j. renew. energy dev 2023, 12(2), 366-374 |367 issn: 2252-4940/© 2023. the author(s). published by cbiore and flexible, has many applications in designing optical lenses in 2d and 3d such as concentrators, distributors, collimators for concentrator photovoltaic systems, daylighting, and illumination systems (buljan et al 2014, gul et al 2016, martin et al 2002, tsangrassoulis et al 2008). in the sms method, a lens is built consisting of two surfaces, which are determined by positioning all points over these surfaces. each point is determined by using the conservation of optical path length. non-imaging optics is used widely in recent years in the design of optical concentrators such as studies by pham et al. (2018), pham et al (2019), mohedano et al. (mohedano 2016), and ryu et al. (2006), etc. in addition, contributors and distributors in daylighting systems are designed by using non-imaging optics in published papers by pham et al. (2017), irfan et al. (2014, irfan et al. (2012), etc. these aspects show that non-imaging optics is a suitable method to design an optical system for a vehicle speed gun. furthermore, focusing on the improvement of the performance of the vehicle speed gun, there are some studies on the design of light sources for replacing ultrasound signals such as a study from erik kreifeldt et al. (1996), a study from kumar et al. (2019, kumar et al (2014), and a study from muzal et al. (2016). in those studies, the authors have tried to use laser sources to detect the velocity of the vehicle. although the speed of light is higher than sound in the doppler effect, however, the accuracy is still limited and that makes the controller more complicated. on the other hand, some studies focus on the design of a control system to improve the performance of vehicle speed guns by handling the reflected signal. there are some studies in this field such as studies from adnan et al. (2013), mandava et al. (2018), and nguyen et al. (2014). all these mentioned analyses illustrate that many methods can be used to improve the performance of the vehicle speed gun depending on the specific condition of the application. however, almost vehicle speed gun systems in commercial can work well in daytime conditions. therefore, in this work, we limit the project to applying the vehicle speed gun in the nighttime condition to detect the speed of cars without modifying much of the main system. the extended application for the nighttime condition is used by adding the illuminator which can use efficient energy in a compact battery. in the field of estimating the speed of cars in nighttime condition, there is few studies come from setiyono et al. (2021), hassan et al. (2016), dharhir et al. (2019), and abdelwahed et al. (2022), etc. although there are few studies on the estimation of speed guns at nighttime, however, almost focus on data analysis and image processing to detect the speed of the vehicle. there is no literature for design a collimator for the illuminator of the vehicle speed gun. thus, the design of a collimator of illuminator with high energy efficiency is promising to apply the normal vehicle speed gun in nighttime conditions without increasing cost much in real conditions. in this paper, the study focuses on the design of an optical collimator system for a led array with infrared wavelength (from 750 nm to 850 nm) applied to a vehicle speed gun system. the collimator is designed using non-imaging optics based on the conservation of optical wavelength to guide the light emitting from the led array to a target 3x3 m2 with a distance of about 100 m. the structure of each collimator is divided into two parts for the construction of each: one part uses total internal reflection while the other part uses the refraction phenomenon to design a cartesian oval surface to collimate the light from led. each collimator is simulated in matlab, then the data is imported to lighttoolstm to do simulation and raytracing to evaluate the performance of the collimator. 2. design of the collimator for vehicle speed gun 2.1 design of collimator using refraction for each lens to design a collimator for led, the dimension of led is needed. led is chosen from the library of lighttoolstm with specifications as shown in fig. 1. from the dimension of each led, a led array consisting of 8x8 leds is proposed with dimensions of 180x180 mm with 64 high-power infrared leds. the collimator is an array consisting of 64 single lenses in which each lens acts the role of a collimator for each led. all lenses are combined by positioning on the transparent substrate. the lens is constructed by two surfaces: the entry surface is flat while the exit surface is a cartesian oval surface. the light emitting from led is refracted at the entry surface, then the refracted light will be refracted again at the exit surface to be a parallel ray outside of the lens to reach the target area with a distance of about 100 m. in actual conditions, led is not a point light source. however, for the sake of design simplicity, led is assumed as a point light source. the position of the point light source is determined as the intersection of two edge rays of the light emitted from led with a 1200 angle as shown in fig. 2. based on the position of the focal point, the shape of the collimator is designed which is shown in fig. 3. in this design, the conservation of optical path length is the most important. the edge rays emitting from unreal focal point f with angle 1200 are determined. in addition, the entry flat surface of the collimated lens is placed in front of the led. from this information, the position of the edge ray going into the lens is determined. the ray entering the lens will be refracted at the entry surface following snell’s theorem: sin 𝛼𝛼 = 𝑛𝑛 × sin 𝛽𝛽 (1) where α and β are incident and refracted angles compared to the normal of the entry surface while n is the refractive index of poly methyl methacrylate (pmma) for the wavelength of 750 nm. fig. 1 the shape and dimension of led that is chosen from the library of lighttoolstm l.m. nhut et al int. j. renew. energy dev 2023, 12(2), 366-374 |368 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 2 the position of the unreal focal point as a point source fig. 3 the point light source from focal point f illuminates the lens in the flat entry surface and the bundle ray exiting the lens is an array of parallel rays by using snell’s theorem, the direction of refracted ray can be determined. depending on the thickness of the lens’s base, the position of point b1 is calculated. the refracted lights are guided by the cartesian oval surface to exit the lens as parallel rays. the cartesian oval surface is evaluated by using the conservation of optical path length leading to every point of exit surface is determined as shown in fig. 3. a1+n×b1+c1=a2+n×b2+c2=a3+n×b3=a4+n×b4+c4=⋯=opl (2) where a1, a2, a3, a4, b1, b2, b3, b4, c1, c2, c4 are partial optical path lengths of the rays emitting from led; opl is the sum of optical path lengths of each ray. all of them are length with the unit in mm. in this design, a1, a2, a3, and a4 are the positions of the light entering the lens in the entry surface where snell’s law is used to determine the directions of refracted rays. b1, b2, b3, and b4 are points to evaluate the cartesian oval surface. c1, c2, b3, and c4 are points that are in the same wavefront. the conservation is always right for one wavefront. in conclusion, the position of focal point f for the point source is firstly assumed to evaluate the pan of light coming to the entry surface. snell’s law is secondarily used to determine the directions of any refracted light. the thickness substrate’s lens and directions are used to estimate the two extreme points of the cartesian oval surface. finally, the conservation of optical path length and numerical method is used to determine representing points of the cartesian oval surface. combining every point calculated to estimate the cartesian oval surface (freeform exit surface). as a result, the collimator is designed completely consisting of a flat entry surface and the freeform exit surface to collimate the light emitting from led to go to the target area. after that, the data can be used to revolve around the center axis to create a 3d collimator in lighttools as shown in fig. 4. fig. 4 the single led and collimator and array of leds and collimators to apply this design of led and collimator lenses to vehicle speed guns, the lenses are combined by using only one substrate for all lenses. each lens will receive a light emitting from led to transfer the light to exit the lens as a bundle of parallel rays. however, a drawback of this design is that it is highly sensitive to alignment. if the array of collimators is not assembled accurately, all lenses will have errors in making parallel rays leading to the bundle of rays, thus cannot transfer correctly to the target. therefore, this design requires a highaccuracy assembly process, which is hard to achieve in practical conditions. to overcome this drawback, another design of collimator is proposed in which the collimator lens is designed by using simultaneously total internal reflection and refraction. the design of this collimator lens is introduced in detail in the next part. 2.2 design of collimator using total internal reflection and refraction in each lens in general, the collimator lens can be divided into two parts: the first part is designed using refraction while the second part is designed using the total internal reflection. dimensions of the first part and second part can be chosen depending on the size of the led’s lens. the collimator lens has a structure with a hole in the center to hold the led, the exit surface is flat as shown in fig. 5. in this design, the led collimator lens comprises a refractive center section (front center entry), a refractive center section (side center entry), and a side reflective surface where the reflection is due to total internal reflection (tir). the sections of the front center entry and side center entry capture a different angular portion of the light source as shown in fig. 6. fig. 5 led with collimator lens with a hole in the center to hold led’s lens l.m. nhut et al int. j. renew. energy dev 2023, 12(2), 366-374 |369 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 6 the collimator lens with two portions for led using tir and refraction to design there are some requirements for this design which are (i) the hole center diameter is bigger than that of the led’s lens; (ii) the inner height of the hole center is higher than that of the led’s lens; (iii) diameter of exit surface is equal to the outer circle of led’s base. fig. 5 shows the components of the led and collimator lens. the process for the design of the led collimator is described in several steps as follows. step 1. the designer chooses the total height of the collimator lens and the angular portion of light spread for the front center entry and side center entry. step 2. using the diameter of the center hole, the surface where a portion of the bundle of rays comes into the lens is estimated. the side center entry surface can be freeform or cylindrical. however, to make the design simple, the cylindrical surface is chosen to receive a portion of illumination emitting from led. step 3. the bundle of rays coming to the side center entry is refracted at the interface to go to the side reflective surface following snell’s law. the shape of the side reflective surface is estimated so that the rays are reflected as a total internal reflection. to calculate the side reflective surface, the position of unreal focal point f_side is chosen and the conservation of optical path length is used in eq. 3. 𝑛𝑛 × 𝑎𝑎1 + 𝑛𝑛 × 𝑏𝑏1 = 𝑛𝑛 × 𝑎𝑎2 + 𝑛𝑛 × 𝑏𝑏2 = 𝑛𝑛 × 𝑎𝑎3 + 𝑛𝑛 × 𝑏𝑏3 = 𝑂𝑂𝑂𝑂𝑂𝑂 (3) where n is the refractive index of the lens medium for the light with a wavelength of 750 nm. the side reflective surface adapts eq. 3 should be a portion of the parabolic surface with the focal point being unreal focal point f_side in fig. 6. in this step, the incident angle 𝛾𝛾 of a ray in fig. 6 should be greater than the critical angle 𝛾𝛾critical which is used to calculate the total internal reflection in eq. 4. n × sin γcritical = 𝑛𝑛𝑎𝑎𝑎𝑎𝑎𝑎 × sin 900 (4) where nair is the refractive index of air. if the incident angle 𝛾𝛾 is smaller the 𝛾𝛾critical incident light will refract at the side reflective surface to go outside of the lens. when the incident angle 𝛾𝛾 increases, the refractive angle will increase immediately until reach a value 900 at which the total internal reflection happens leading to the refracted beam will be changed to the reflected beam inside of the lens. therefore, if the value of n of the lens is 1.486 and the value of nair is 1, the critical angle 𝛾𝛾critical should be about 42.40. so that if that requirement for incident angle 𝛾𝛾 is not adapted, the position of unreal focal point f_side should be chosen again until the requirement adapting. step 4. based on the inner height and diameter of the center hole, the positions of ac1 and ac3 are estimated completely. the conservation of optical path length is calculated for every ray coming to the front center entry in eq. 5. 𝑎𝑎1 + 𝑛𝑛 × 𝑏𝑏1 = 𝑎𝑎2 + 𝑛𝑛 × 𝑏𝑏2 = 𝑎𝑎3 + 𝑛𝑛 × 𝑏𝑏3 = 𝑂𝑂𝑂𝑂𝑂𝑂 (5) any ray coming to the front center entry adapts the conservation of optical path length leading to the freeform of the front center entry surface is estimated. step 5. when the side center entry and the front center entry are estimated already, the data of the crossectional view revolves around the center axis along the led collimator to get a collimator lens as fig. 5. after that, an array of leds and collimators can be created to be a flashlight for a vehicle speed gun as shown in fig. 7. in this structure of collimator lens, the requirement for the high accuracy assembly process is reduced significantly because every led and collimator is independent of others. all these steps to design the collimator lens are carried out in the matlab program. the data of the cross-sectional view of the collimator is calculated using a numerical method. after that, the discrete data from matlab is inserted into lighttoolstm software to draw the collimator in 3d (3 dimensions). the flowchart of the design procedure is shown in fig. 8. in general, the design is firstly carried out in matlab based on some initial parameters relating to the characteristics of led. in programming, the conditions to stop the process are always important. in this process, stopping conditions are total internal reflection and size of the collimator lens. tir has to happen at the side reflective surface while the size of the collimator lens should be equal to the size of the outer circle of the led’s base. if the designed lens is not meet these requirements, the unreal focal point f_side should be chosen again to obtain all requirements. the data collected in matlab is just a 2d design. after that, the 2d data from matlab is inserted into lighttools to draw the object in 3d by revolving the shape of the lens in 2d. the 3d lens draw in lighttoolstm is used to simulate for estimation of the performance of the proposed system. fig. 7 the array of leds and collimator lenses applied to the vehicle speed gun l.m. nhut et al int. j. renew. energy dev 2023, 12(2), 366-374 |370 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 8 the flow chart to design the array of leds and collimators applied to a vehicle speed gun in addition, the uniformity of light distribution over the target is also an important parameter to estimate the performance of the proposed system which is estimated in lighttoolstm and is calculated in eq. 6. 𝑈𝑈 = 100 − 𝑚𝑚𝑎𝑎𝑚𝑚−𝑚𝑚𝑎𝑎𝑚𝑚 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 × 100% (6) where average, min, and max are the mean value, lowest value, and highest value of the intensity of light over the target area, respectively. 3. results and discussion this study aims to apply the designed collimator lens to a flashlight system for a vehicle speed gun performing in the night condition with a compact and effective battery. the main requirement of the illuminator of the proposed vehicle speed gun is the wavelength belonging to the range of the invisible spectrum for safe driving. it depends on the properties of the charge coupled device (ccd) camera used for the vehicle speed gun system. generally, the illuminator should have a wavelength in a range from 750 nm to 950 nm. for convenience, in this study, the characteristics of led with a wavelength of 750 nm are chosen from the lighttoolstm library as shown in fig. 1. pmma material is a good selection for collimator lenses because of its high transmission coefficient of about 93 % (ali 2015) as shown in fig. 9. in this study, the design of the collimator lens is applied to the vehicle speed gun system so that the collimator lens will have a small spread angular distribution of exiting light to distribute the light to the target area. that small spread angle will be obtained by modifying the side reflective surface or modifying the flat exit surface be a curve surface. the modification of the side reflective surface is chosen to make the design process simple. the spread angle is calculated depending on the distance from the light source to the target area. there are two ways to illuminate the vehicle a long distance of around 100 m or a shorter distance of about 30 m. fig. 9 the refractive index and transmission of pmma material following the wavelength for the designed collimator lens if the light source is placed at the same place as speed gun systems, the distance from the light source is long so that the spreading angle should be small even equal to zero. for long distances, a random defect of emission light from led is enough for the spread of light at the target area and it is difficult to reach the requirement of lux on the target area in outside illumination for a ccd (charge coupled device) camera taking photos (vu 2017). therefore, the flashlight source can be remotely controlled by using wireless communication to put the light source near the target area. wireless communication is a mature technology and it can be applied easily to the vehicle speed gun system. the distance from the light source to the target area (car’s position) is chosen at about 30 m as shown in fig. 10. table 1 illustrates the parameters of simulation and ray-tracing of the optical system in lighttoolstm software to estimate the performance of the designed collimator lenses in the flashlight source for the vehicle speed gun system. the discrete data from matlab is inserted into lighttoolstm to build the flashlight source. fig. 11 a) shows the ray-tracing of a single led and single collimator. the results show that the rays emitting from led are divided into two sections. the outer section entering the collimator is refracted to go to the side surface where the rays are total internal reflected to exit at the collimator lens at a flat exit surface. the center section of bundle rays enters the collimator at the freeform surface where the directions of rays are changed to go to the exit surface as parallel rays. however, led is not a point source in reality and simulation leading to the lightly spread of the bundle of rays exiting the collimator. the ray-tracing is carried out for an array of leds and collimators which is illustrated in fig. 11 b). fig. 10 the way to set up the flashlight source for a vehicle speed gun using wireless communication to reduce the distance from the light source to the target vehicle l.m. nhut et al int. j. renew. energy dev 2023, 12(2), 366-374 |371 issn: 2252-4940/© 2023. the author(s). published by cbiore table 1 parameters for simulation and ray-tracing in lighttoolstm items value wavelength 750 nm the outer circle of the led’s base 21.5 mm distance from the light source to the target area 30 m spreading angle < 2.90 inner center heigh 6.5 mm diamter of inner center hole 6.0 mm total height of collimator lens 15 mm diameter of the flat exit surface 21.5 mm lens material pmma refractive index of pmma 1.486 number of rays for simulation and ray-tracing 3.000.000 rays power of single led 1 w number of lm per wat 500 lm number of lenses and leds 8 x 8 = 64 elements the dimension of the flashlight source consists of 64 elements 180 x 180 mm2 target area the average luminance on target are 3 x 3 m2 150 lx fig. 11 the ray-tracing of a) a single led and collimator and b) an array of leds and collimators the simulation of the flashlight source of the vehicle speed gun is conducted in lighttoolstm with wavelength 750 nm and 3.000.000 rays. the results of the simulation process show that the distribution of light over the target area is quite uniform as illustrated in fig. 12. the uniformity u can be changed as a function of some parameters such as the distance of the light source and target area, the wavelength of the led, and the interview beam angle of the led. when the characteristics of led in real conditions are different from the design parameters or the assembly of components is not perfect, all those aspects affect to uniformity and optical efficiency of the designed flashlight source for the vehicle speed gun. therefore, the effect of those aspects was analyzed in lighttoolstm to evaluate the ability to perform in real conditions. fig. 12 ray-tracing of the flashlight source in lighttoolstm and illumination distribution over the target area fig. 13 the optical efficiency and uniformity are functions of a) the beam view angle of led and b) the wavelength emission of led fig. 13 a) shows that the uniformity and optical efficiency of the designed collimator are independent of the change of beam view angle of led from 1050 to 1350. this trend relates to the structure of each collimator lens. the structure of the collimator lens is divided into two parts the outer angle and the inner angle. in which the inner refraction part has a hole center that can cover all rays emitting from led with a beam view angle smaller than 1800. in this design, the rays with emission angles from leds smaller than 300 are guided by refraction at the front center entry to be a horizontal ray that will go forward to the target area. in addition, the rays with emission angle from led greater than 300 are refracted and total internal reflected (tir) to be parallel rays exiting the lens. although the simulation results show that the beam view angle of led does not affect much the uniformity and the optical efficiency of the collimator, l.m. nhut et al int. j. renew. energy dev 2023, 12(2), 366-374 |372 issn: 2252-4940/© 2023. the author(s). published by cbiore the size of the light source will affect significantly the performance of the collimator lenses. however, the size of the emission area source is kept constant in this design because the led characteristics are inserted from the library of lighttoolstm. this mention illustrates that the designed collimator lenses can be used with different leds with different beam view angles of emission beam of leds. fig. 13 b shows that the change in the wavelength of led will affect the optical efficiency and uniformity of the collimator system. when wavelength changes leading to the refractive index of the medium lens (pmma) will be changed, thus, the refractive angle of incident rays coming to the interface will also change. depending on the wavelengths which are shorter or longer the exiting bundle of rays will be diverged or converged leading to uniformity and the optical efficiency will be changed. the optical efficiency changes slightly as a function of wavelength and reaches the maximum value at wavelength 750 nm which is chosen to design the collimator lens. the optical efficiency still reaches the value of about 80 % when the wavelength is 810 nm (infrared wavelength) which is suitable for the application of a vehicle speed gun. the optical efficiency of the collimator lenses can be changed following the change of wavelength; however, the change is slight with the sensitivity which is calculated by the ratio of change in optical efficiency and the change of wavelength that has the value of about 10 %. this analysis illustrates that the designed collimator lenses can be used with quite freely infrared led sources. on the other hand, the uniformity tends to increase slightly because of the refractive index decreasing leading to the collimated beam diverging a little. the value of uniformity is always higher than 70 % in the range of analyzed infrared wavelength. this analysis illustrates that the designed collimator is suitable for different infrared leds. in another word, this designed collimator is not only suitable for pmma materials but also it can be made with different materials for lenses. all aspects mentioned above show that the designed collimator can be carried out in actual conditions with an acceptable range of tolerance. there are some studies about the determination of a car’s speed in a nighttime as mentioned above. however, the approach of using infrared wavelength as a flashlight to take photos which are used as evidences of the car’s speed over the limit velocity is a novel idea. the advantage of this approach is the structure of the conventional vehicle speed gun system does not need to be modified. a portable vehicle speed gun using the dopler effect can be used in the nighttime, however, the photos for evidence issue should be solved. therefore, using infrared led as a flashlight for the aim of taking photos in a vehicle speed gun is a suitable method with minimum extended cost. depending on the application, the unique design of the lens for the led chip is necessary to guide the illumination energy to the target area to meet the requirements of the particular application. therefore, to use effectively the infrared led with a compact battery, the design of a collimator for the infrared led in the lighttoolstm library is proposed in this study. the design of the collimator is performed and simulated in lighttoolstm to estimate the rationality and the performance when it is applied to a flashlight of a vehicle speed gun system. promising simulation results show the high potential to perform the proposed flash light assembled with a designed collimator to the vehicle speed gun to extend the ability of the traditional vehicle speed gun. the simulation results of the proposed collimator have been compared to other studies published before. the comparison illustrates the designed collimator is acceptable. in the design of collimators from haobo et al. (haobo 2015) and from vidal et al. (vidal 2014), the optical efficiency is about 80 % which is quite similar to that of the proposed collimator. however, both collimators have a distance from led to target area short of about 5 m which is not suitable for the application of flash light of a vehicle speed gun. moreover, the structure of these collimators is more complicated and bigger compared to the proposed system. these properties will negatively affect the portability of the vehicle speed gun. furthermore, the simulation results of the proposed collimator have been compared to that of studies from the research groups chen et al. (2014) and vu et al. (2017). in the research of chen et al., although the optical efficiency and uniformity are higher than that of the proposed collimator, the size and the divergent angle of the collimator lens are large which can make the flashlight bulky and have a negative effect on the utility of energy of the battery. in terms of the study from vu et al. the optical efficiency is quite similar to that of the proposed collimator. however, the divergence angle is quite large of about 7.50 which can not apply to the flashlight of a vehicle speed gun with a long distance from the light source to the target area. the designed collimator in vu et al. research is just suitable for illumination distribution with the collimator placed in front of the distributor lens. based on these analyses, the comparison between collimators that are designed for different applications is difficult. however, all those analyzed aspects show the necessity of designing a particular collimator for infrared led applied to a vehicle speed gun. consequently, the simulation results obtained from the proposed collimator illustrate the conformity to apply to a flashlight of a vehicle speed gun. the simulation results also show a high potential to make a prototype to investigate the system in real conditions. all simulation results used for comparison are shown in table 2. additionally, fig. 14 shows the shapes and structures of some collimators published already to have an overview of the comparison. table 2 the properties of the proposed system and other studies. properties\studies proposed study (chen et al 2014) (haobo et al 2015) (vidal et al 2014) (vu et al 2017) led type ≥ 750 nm cree xr-e cree xp-e2 525 nm ±15 nm white led distance of target 30 m not mentioned 5 m 5 m infront of distributor diameter of collimator 21.5 mm 32.36 mm 17.12 mm 60 mm 10 mm type of collimator refraction and tir refraction and tir refraction and tir refraction and reflection refraction optical efficiency ~ 80 % 89 % 79.2 % 80 % ~ 80 % uniformity ~ 71 % 78 % not mentioned not mentioned not mentioned for collimator divergence angle < 20 20 and 40 30 130 < 20 7.50 l.m. nhut et al int. j. renew. energy dev 2023, 12(2), 366-374 |373 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 14 the structure of some collimators compared to the proposed collimator 6. conclusion in this study, the design of the collimator in detail is introduced using non-imaging optics. the collimator system is a combination of many single collimators which is constructed in the matlab program with some initial parameters for the design process such as the wavelength of 750 nm, materials for the lens of pmma, etc. two different structures of single collimator lenses are proposed, and after that, the characteristics of both are analyzed to choose the better structure. the chosen single collimator lens is structured consisting of two parts: the outer part uses refraction and total internal reflection to guide the ray entering the collimator while the inner part uses refraction to guide the rays coming to the front center entry interface. the size of each part is chosen freely using the beam view angles of emission rays from led. in this design, the outer part and the inner part are separated at the angle of 300 beam view angle. a single lens is designed in the matlab program, and after that, the data from matlab is transferred to lighttoolstm to draw three dimensions of the collimator. the 3d object in lighttoolstm is added the optical properties to do the ray-tracing and simulation to evaluate the performance of the single collimator lens and the collimator system. the simulation results show that the designed collimator array can be carried out with different infrared leds. furthermore, the collimator lens can be made from different materials without changing much of the performance of the collimator which is applied to a vehicle speed gun system. all analyzed aspects in this study illustrate the high potential of the collimator system combined with the infrared led to apply a vehicle speed gun to increase the performance of the speed gun at nighttime performance. although the study stops at the simulation in lighttoolstm, good properties of the designed collimator system motivate a plan to create a prototype to evaluate the performance of the designed collimator system in actual conditions soon. based on simulation results in this study, a prototype of an illuminator to carry out a vehicle speed gun system in nighttime conditions is a promise. the simulation results are used to compare to other collimators that are designed by other research groups. acknowledgments this study is highly supported by the faculty of vehicles and energy engineering – ho chi minh university of technology and education (hcmute) and the non-imaging optics lab – myongji university korea. author contributions: thanh-tuan pham and seoyon shin conceived the presented idea. thanh-tuan pham and le minh nhut developed the theory and performed the computations. tien-dung tran and vu dinh huan verified the analytical methods. seoyong shin encouraged the group to investigate a specific aspect and supervised the findings of this work. all authors discussed the results and contributed to the final manuscript. all authors have read and agreed to the published version of the manuscript. conflicts of interest: the authors declare no conflict of interest. references abdel-wahed, t., & el esawey, m. (2022). a comparison of daytime and nighttime operating speed on rural multi-lane highway sections in egypt. sohag engineering journal, 2(2), 66-78. https://doi.org/10.21608/sej.2022.135873.1012 adnan, m. a., sulaiman, n., zainuddin, n. i., & besar, t. b. h. t. (2013, april). vehicle speed measurement technique using various speed detection instrumentation. in 2013 ieee business engineering and industrial applications colloquium (beiac) (pp. 668-672). ieee. https://doi.org/10.1109/beiac.2013.6560214 ali, u., karim, k. j. b. a., & buang, n. a. (2015). a review of the properties and applications of poly (methyl methacrylate)(pmma). polymer reviews, 55(4), 678-705. https://doi.org/10.1080/15583724.2015.1031377 bassani, m., catani, l., cirillo, c., & mutani, g. (2016). night-time and daytime operating speed distribution in urban arterials. transportation research part f: traffic psychology and behaviour, 42, 5669. https://doi.org/10.1016/j.trf.2016.06.020 benı´tez, p., min˜ano, j. c., blen, j., mohedano, r. n., chaves, j. l., dross, o., ... & falicoff, w. (2004). simultaneous multiple surface optical design method in three dimensions. optical engineering, 43(7), 1489-1502. https://doi.org/10.1117/1.1752918 benitez, p., minano, j. c., blen, j., mohedano, r., chaves, j., dross, o., ... & falicoff, w. (2004, january). sms design method in 3d geometry: examples and applications. in nonimaging optics: maximum efficiency light transfer vii (vol. 5185, pp. 18-29). spie. https://doi.org/10.1117/12.506857 buljan, m., mendes-lopes, j., benítez, p., & miñano, j. c. (2014). recent trends in concentrated photovoltaics concentrators’ architecture. journal of photonics for energy, 4(1), 040995-040995. https://doi.org/10.1117/1.jpe.4.040995 chaves, j. (2008). introduction to nonimaging optics. crc press. doi: https://doi.org/10.1201/9781420054323 chen, c., & zhang, x. (2014). design of optical system for collimating the light of an led uniformly. josa a, 31(5), 1118-1125. https://doi.org/10.1364/josaa.31.001118 cheng, h., xu, c., jing, x., & tam, h. y. (2015). design of compact led free-form optical system for aeronautical illumination. applied optics, 54(25), 7632-7639. https://doi.org/10.1364/ao.54.007632 chong, k. k., siaw, f. l., wong, c. w., & wong, g. s. (2009). design and construction of non-imaging planar concentrator for concentrator photovoltaic system. renewable energy, 34(5), 13641370. https://doi.org/10.1016/j.renene.2008.09.001 dhahir, b., & hassan, y. (2019). using horizontal curve speed reduction extracted from the naturalistic driving study to predict curve collision frequency. accident analysis & prevention, 123, 190-199. https://doi.org/10.1016/j.aap.2018.11.020 dross, o., mohedano, r., benitez, p., minano, j. c., chaves, j., blen, j., ... & munoz, f. (2004, september). review of sms design methods and real world applications. in nonimaging optics and efficient illumination systems (vol. 5529, pp. 35-47). spie. https://doi.org/10.1117/12.561336 gul, m., kotak, y., & muneer, t. (2016). review on recent trend of solar photovoltaic technology. energy exploration & exploitation, 34(4), 485-526.martin, k.l. an overview of daylighting systems. sol. energy 2002, 73, 77–82. https://doi.org/10.1177/0144598716650552 gutiérrez, m., miñano, j. c., vega, c., & benítez, p. (1996). application of lorentz geometry to nonimaging optics: new three-dimensional ideal concentrators. josa a, 13(3), 532-540. https://doi.org/10.1364/josaa.13.000532 hamelmann, p., vullings, r., kolen, a. f., bergmans, j. w., van laar, j. o., tortoli, p., & mischi, m. (2019). doppler ultrasound technology https://doi.org/10.21608/sej.2022.135873.1012 https://doi.org/10.1109/beiac.2013.6560214 https://doi.org/10.1080/15583724.2015.1031377 https://doi.org/10.1016/j.trf.2016.06.020 https://doi.org/10.1117/1.1752918 https://doi.org/10.1117/12.506857 https://doi.org/10.1117/1.jpe.4.040995 https://doi.org/10.1201/9781420054323 https://doi.org/10.1364/josaa.31.001118 https://doi.org/10.1364/ao.54.007632 https://doi.org/10.1016/j.renene.2008.09.001 https://doi.org/10.1016/j.aap.2018.11.020 https://doi.org/10.1117/12.561336 https://doi.org/10.1177/0144598716650552 https://doi.org/10.1364/josaa.13.000532 l.m. nhut et al int. j. renew. energy dev 2023, 12(2), 366-374 |374 issn: 2252-4940/© 2023. the author(s). published by cbiore for fetal heart rate monitoring: a review. ieee transactions on ultrasonics, ferroelectrics, and frequency control, 67(2), 226-238. https://doi.org/10.1109/tuffc.2019.2943626 hassan, s. a., wong, l., mashros, n., alhassan, h. m., sukor, n. s. a., rohani, m., & minhans, a. (2016). operating speed of vehicles during rainfall at night: case study in pontian, johor. jurnal teknologi, 78(7-2). koshel, r. j. (ed.). (2012). illumination engineering: design with nonimaging optics. john wiley & sons. https://doi.org/10.1002/9781118462539 kreifeldt, erik. "new laser speed gun detector thwarts speed measurement." optics and photonics news 7.10 (1996): 6-6. https://doi.org/10.1364/opn.7.10.000006 kumar, a., jaiswal, a., jaiswal, n., & sharma, r. (2014). vehicles anticollision system. international journal of computer applications, 99(19), 7-9. https://research.ijcaonline.org/volume99/number19/pxc389763 5.pdf kumar, vimal, shankar c. subramanian, and rajesh rajamani. "vehicle tracking for heavy road vehicle collision avoidance with an inexpensive solid state laser sensor." 2019 ieee intelligent transportation systems conference (itsc). ieee, 2019. https://doi.org/10.1109/itsc.2019.8917420 liu, b., sang, x., yu, x., gao, x., liu, l., gao, c., ... & du, j. (2019). timemultiplexed light field display with 120-degree wide viewing angle. optics express, 27(24), 35728-35739. https://doi.org/10.1364/oe.27.035728 miñano, j. c., mohedano, r., &amp; benítez, p. (2015). nonimaging optics. the optics encyclopedia. mohedano, rubén, and ralf leutz. "cpv optics." handbook of concentrator photovoltaic technology (2016): 187-238. muzal, m., mierczyk, z., zygmunt, m., wojtanowski, j., & piotrowski, w. (2016, december). measurement of vehicles speed with full waveform lidar. in laser technology 2016: progress and applications of lasers (vol. 10159, pp. 351-362). spie. https://doi.org/10.1117/12.2262839 mandava, m., gammenthaler, r. s., & hocker, s. f. (2018, august). vehicle speed enforcement using absolute speed handheld lidar. in 2018 ieee 88th vehicular technology conference (vtc-fall) (pp. 1-5). ieee. https://doi.org/10.1109/vtcfall.2018.8690893 nguyen, v. c., & dinh, d. k. (2014, october). length and speed detection using microwave motion sensor. in 2014 international conference on advanced technologies for communications (atc 2014) (pp. 371-376). ieee. https://doi.org/10.1109/atc.2014.7043414 o'gallagher, j. (2008). nonimaging optics in solar energy (vol. 2). morgan & claypool publishers.. pham, t. t., vu, n. h., & shin, s. (2017). daylighting system based on novel design of linear fresnel lens. buildings, 7(4), 92. https://doi.org/10.3390/buildings7040092 pham, t. t., vu, n. h., & shin, s. (2018). design of curved fresnel lens with high performance creating competitive price concentrator photovoltaic. energy procedia, 144, 16-32. https://doi.org/10.1016/j.egypro.2018.06.004 pham, t. t., vu, n. h., & shin, s. (2019). novel design of primary optical elements based on a linear fresnel lens for concentrator photovoltaic technology. energies, 12(7), 1209. https://doi.org/10.3390/en12071209 ryu, k., rhee, j. g., park, k. m., & kim, j. (2006). concept and design of modular fresnel lenses for concentration solar pv system. solar energy, 80(12), 1580-1587. https://doi.org/10.1016/j.solener.2005.12.006 setiyono, b., ratna, d., ilmi, m. f., & usadha, i. g. n. r. (2021, february). estimation of vehicle speed at night based on rear lights. in journal of physics: conference series (vol. 1752, no. 1, p. 012087). iop publishing. https://doi.org/10.1088/17426596/1752/1/012087 tsangrassoulis, a. a review of innovative daylighting systems. adv. build. energy res. 2008, 2, 33–56. https://doi.org/10.3763/aber.2008.0202 ullah, i., & shin, s. (2014). highly concentrated optical fiber-based daylighting systems for multi-floor office buildings. energy and buildings, 72, 246-261. https://doi.org/10.1016/j.enbuild.2013.12.031 ullah, i., & shin, s. y. (2012). development of optical fiber-based daylighting system with uniform illumination. journal of the optical society of korea, 16(3), 247-255. https://doi.org/10.1109/oecc.2012.6276589 vidal, e., otaduy, d., gonzález, f., saiz, j. m., & moreno, f. (2009, august). design and optimization of a collimating optical system for high divergence led light sources. in current developments in lens design and optical engineering x (vol. 7428, pp. 115-124). spie. https://doi.org/10.1117/12.825287 vu, n. h., pham, t. t., & shin, s. (2017). led uniform illumination using double linear fresnel lenses for energy saving. energies, 10(12), 2091. https://doi.org/10.3390/en10122091 winston, r., jiang, l., & ricketts, m. (2018). nonimaging optics: a tutorial. advances in optics and photonics, 10(2), 484-511. https://doi.org/10.1364/aop.10.000484 winston, r., miñano, j. c., & benitez, p. g. (2005). nonimaging optics. elsevier. isbn: 978-0-12-759751-5 https://doi.org/10.1016/b978-0-12-759751-5.x5000-3 wu, j., liu, z., li, j., gu, c., si, m., & tan, f. (2009, july). an algorithm for automatic vehicle speed detection using video camera. in 2009 4th international conference on computer science & education (pp. 193-196). ieee. https://doi.org/10.1109/iccse.2009.5228496 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1109/tuffc.2019.2943626 https://doi.org/10.1002/9781118462539 https://doi.org/10.1364/opn.7.10.000006 https://research.ijcaonline.org/volume99/number19/pxc3897635.pdf https://research.ijcaonline.org/volume99/number19/pxc3897635.pdf https://doi.org/10.1109/itsc.2019.8917420 https://doi.org/10.1364/oe.27.035728 https://doi.org/10.1117/12.2262839 https://doi.org/10.1109/vtcfall.2018.8690893 https://doi.org/10.1109/atc.2014.7043414 https://doi.org/10.3390/buildings7040092 https://doi.org/10.1016/j.egypro.2018.06.004 https://doi.org/10.3390/en12071209 https://doi.org/10.1016/j.solener.2005.12.006 https://doi.org/10.1088/1742-6596/1752/1/012087 https://doi.org/10.1088/1742-6596/1752/1/012087 https://doi.org/10.3763/aber.2008.0202 https://doi.org/10.1016/j.enbuild.2013.12.031 https://doi.org/10.1109/oecc.2012.6276589 https://doi.org/10.1117/12.825287 https://doi.org/10.3390/en10122091 https://doi.org/10.1364/aop.10.000484 https://doi.org/10.1016/b978-0-12-759751-5.x5000-3 https://doi.org/10.1109/iccse.2009.5228496 design of optical collimator system for vehicle speed gun using non-imaging optics le minh nhuta, thanh-tuan phama0f , tien-dung tranb, vu dinh huana, seoyong shinc1f 1. introduction international journal of renewable energy development int. j. renew. energy dev. 2023, 12(2), 300-312 | 300 https://doi.org/10.14710/ijred.2023.50065 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id biodiesel production from a naturally grown green algae spirogyra using heterogeneous catalyst: an approach to rsm optimization technique teku kalyania* , lankapalli sathya vara prasada , aditya kolakotib adepartment of mechanical engineering, andhra university college of engineering, visakhapatnam, 530003, india bdepartment of mechanical engineering, raghu engineering college(a), visakhapatnam, 531162, india abstract. the present study focuses on oil extraction and biodiesel production from naturally grown green spirogyra algae. solvent oil extraction and oil expeller techniques were used to extract the spirogyra algae oil (salo), and the oil yields were compared to identify the most productive method. using chicken eggshell waste (cesw) heterogeneous catalyst (hc) was prepared for the production of spirogyra algae oil biodiesel (salobd). furthermore, box–behnken (bb) assisted response surface method (rsm), an optimisation technique, was used in this study to achieve maximum algae biodiesel yield. from the 29 experimental trails, 96.18 % salobd was achieved at molar ratio (10:1), heterogeneous catalyst (0.6 wt.%), temperature (48 oc), and time (180 minutes). the predicted values of r2 (97.51%) and adj. r2 (95.02 %) are found to be encouraging and fits well with the experimental values. the output results showed that hc was identified as the significant process constraint followed by the time. the fatty acid composition (fac) analysis by gas chromatography (gcms) reveals the presence of 29.3% unsaturated composition and 68.39 wt. % of the saturated composition. finally, the important fuel properties of salobd were identified in accordance with astm d6751. the results obtained using chicken eggshell waste (cesw) for the production of biodiesel were recommended as a diesel fuel replacement to resist energy and environmental calamities. keywords: dried algae powder, dried algae flakes, heterogeneous catalyst, response surface method, spirogyra. @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 17th july 2022; revised: 22nd dec 2022; accepted: 18th january 2023; available online: 25th january 2023 1. introduction globally, the utilisation of non-renewable fuels like petrodieselderived fossil fuels stood as one of the primary sources of fuel supply for transportation, agriculture and industries (höök & tang, 2013). during the earlier days of its discovery, fossil fuels were identified as the most popularly used fuels because of their high efficiency, accessibility and low cost. however, with the increasing demand and high consumption, the availability of petrodiesel-derived fuels is on the verge of extinction. also, the exhaust emissions from petro-diesel-fuelled engines contribute to the reduction of air quality and increase environmental air pollution around the globe. consequently, the search for renewable and clean-burning fuels is attracting a lot of interest. to overcome the challenges, research on various feedstock oils revealed that biodiesels derived from different naturally existing oils might replace the current petrodiesel-derived fuel. therefore, biodiesels from various edible and non-edible oils have become viable alternatives to traditional hydrocarbon fuels like petro-diesel. biodiesels are clean burning renewable fuels, popular for their biodegradable capacity, reliability, less toxic nature and improved engine performance. biodiesel oils are divided into three (3) generations listed as first (1st) generation, second (2nd) generation and third (3rd) generation. the 1st generation of * corresponding author email: kalyanithermal@gmail.com (t. kalyani) biodiesel oils is related to edible oils. for the production of biodiesel, these oils raised the debate about food vs fuel issues. this concern led to 2nd generation of biodiesels. these are linked with non-edible oils and are proven to be a better source of producing biodiesel oils compared to the first generation because these biodiesels are readily available and less expensive. however, crop cultivation, production and oil yield are considered significant tasks in the production of secondgeneration biodiesel oils. these drawbacks made researchers focus on third-generation fuels termed algae biofuels. the algae biodiesels are emerging worldwide since they are available naturally, easy to grow, have high oxygen profiles, capable of reducing emissions etc. algae species, known as a clean, renewable fuel, are identified as one of the fast-growing biomasses to convert fuel into biodiesel. around the globe, more than 60,000 algae species are growing, and nearly 35,000 algae species have been identified (mata et al., 2010). algae groups are classified into seven types listed as red, green, blue-green (cyanobacteria), brown, phytoplankton, seaweeds, and other algae species (nur et al 2015). green microalgae spirogyra division comes under chlorophyta with lengths ranging from 10 to 100 µm. irradiation of light and temperature are the two important factors considered in achieving maximum growth of spirogyra algae. research article https://doi.org/10.14710/ijred.2023.50065 https://doi.org/10.14710/ijred.2023.50065 mailto:kalyanithermal@gmail.com https://orcid.org/0000-0002-3844-2483 https://orcid.org/0000-0001-6411-3299 https://orcid.org/0000-0002-7515-8318 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.50065&domain=pdf t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 301 issn: 2252-4940/© 2023. the author(s). published by cbiore table 1 presence of oil content in some of the microalgae species microalgae species oil content (dry wt. %) chlorella 27 -32 schizochytrium 51-77 dunaliella tertiolecta 35-42 spirogyra 22-35 natashia 44-47 monallanthus salina 22 nannochloris 20-24 nannochloropsis 30-68 neochloris oleoabundans 45-48 botryococcus braunii 26-75 tetraselmis sueica 14 33 isochrysis 733 source: (bhateria & dhaka, 2014); (chisti, 2007); (konga et al., 2017) the range of light intensity varies from 32 to 38 µmol m-2s-2, and temperatures from 12 to 29 oc are considered optimum. altogether, 400 spirogyra free-float algae were identified and commonly found in freshwater near pond surfaces and saline waters (ananthi et al., 2021). species that possess unicellular growth in rivers, open ponds, and freshwater resources are called microalgae. it is evident from research reports that microalgae species produce high oil yields ranging from 20 to 80%, as shown in table 1 (bhateria & dhaka, 2014). the oil yield of microalgae with 70% oil by wt.% in biomass generates 136900 l/ha, and with 30 % oil by wt.% in biomass is 58700 l/ha (chisti, 2007). biodiesel production and characterisation research from various edible and non-edible oils is an ongoing process. oils that are inexpensive and easily accessible have a minute impact on food vs fuel conflict and are recommended as the better source for biodiesel production. however, due to the inadequate atomisation problem in diesel engines, it is suggested to use low-viscosity biodiesels (less than 40 cst.) (kolakoti & appa rao, 2020). hence, the transesterification process is recognised as one of the best processes for reducing kinematic viscosity (hariyanto et al., 2021); this process also gains high efficiency while converting it into biodiesel. the molar ratio (mr), time, temperature, and catalyst concentration (cc) are four significant process parameters used during transesterification. among these, cc plays a beneficial role in achieving the maximum amount of biodiesel. based on the availability and reaction process, two (2) types of catalysts exist, homogenous and heterogeneous catalysts. homogeneous catalyst is most regularly used in the transesterification process as they are highly catalytic in nature, readily available and available at a low price (supriyadi et al., 2022). the most commonly used solutions are homogeneous catalysts: potassium hydroxide (koh) and sodium hydroxide (naoh). due to its corrosive nature, the purification process requires more water and remains expensive. furthermore, the solution mixed in water is tough to collect and reuse. research was being carried out to mitigate these issues using the heterogeneous catalyst (hc). the key benefits of using a heterogeneous catalyst to prepare biodiesel include its environmental friendliness, ease of accessibility, separability, excellent stability and ability to be reused for up to five trials (kolakoti & satish, 2020; hadiyanto et al 2016). kolakoti et al., (2022) prepared a green heterogeneous catalyst calcinated at 700oc for 3 hours using moringa oleifera leaves that are available naturally for biodiesel production using palm oil. the maximum biodiesel achieved was 92.82 % at optimum conditions. fan et al. (2019) carried out an investigation using waste banana peel as a heterogeneous catalyst for the production of biodiesel. the results showed the presence of strong alkalinity, recyclable nature, and catalytic performance observed in the unique alkaline catalyst derived from the waste banana peel. moreover, after the transesterification process, heterogeneous catalyst uses less water for purification, which is inexpensive. as a result, heterogeneous catalyst attracts a lot of interest and suggests this procedure as a practical and cost-effective transesterification method. with the improvement in the latest technologies, the oil and biodiesel yields were estimated using various tools like analysis of variance (anova) and artificial neural network (ann) (kolakoti et al., 2020), which show high accuracy in yield predictions (kolakoti et al., 2020). another researcher (kumar et al., 2020) performed an experimental investigation on algae – jatropha to produce biodiesel yield using the response surface method (rsm) and anova analysis. similarly, (chozhavendhan et al., 2020) identified significant changes by changing process parameters such as catalyst concentration, oil-to-alcohol ratio, reaction temperature and reaction time parameters and concluded that catalyst plays a beneficial role in processing biodiesel. kolakoti et al., (2021) used waste chicken egg shells as a heterogeneous catalyst to produce biodiesel from cooking oil. also, the biodiesel yield was compared using two optimisation techniques rsm and ann. the results concluded that 91% of biodiesel yield is achieved from the two techniques. the discussion shows that there is more potential for algal biofuel as a sustainable green fuel. in the current study, naturally grown green spirogyra algae are collected directly from two open ponds from january to may 2022. a four-step technique (collection – harvesting – oil extraction – transesterification) is used to produce the spirogyra algae biodiesel. this study mainly focuses on collecting algae from open ponds without any kind of cultivation process, nutrients and chemicals. furthermore, the spirogyra algae grow naturally using sunlight, carbon dioxide and rainwater (reddy & majumder, 2014). the collected algae were cleaned and processed in two forms to extract oil, and oil yields via two different techniques were compared. one is the solvent extraction method (i) using dried algae powder, and the second is the oil expeller method (ii) using dry algae flakes. furthermore, the obtained spirogyra algae oil (salo) is converted into spirogyra algae oil biodiesel (salobd) using a heterogeneous catalyst (hc) prepared from chicken eggshell waste (cesw). very limited studies in this combination (salo + cesw) are available to convert raw algae oil to biodiesel with chicken eggshell waste as a heterogeneous catalyst mooted this study. furthermore, an optimisation technique was used in this study to obtain maximum algae biodiesel yield. for this, box– behnken (bb) supported response surface method (rsm) was applied. a number of 29 experimental trials were conducted randomly by varying the significant process constraints such as heterogeneous catalyst, molar ratio, time and temperature. hence the main objective of the present study is to extract algae oil and compare oil yields using two oil extraction techniques and to produce biodiesel from a naturally grown green algae spirogyra using chicken egg shell waste as a heterogeneous catalyst. finally, using rsm optimisation technique, the experimental biodiesel yield prediction was attained by varying the molar ratio, heterogeneous catalyst, time and temperature, conducting 29 experimental runs. 2. methods and materials 2.1 materials chemicals such as methanol (ch3oh) with (99.8 %) purity, sulphuric acid (h2so4) with (99.99 %) purity and n-hexane with (99 %) purity of different grades were purchased from sigma aldrich (merck) visakhapatnam, india. the chicken eggshell wastes are collected from a local restaurant in visakhapatnam. t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 302 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 1 locations of spirogyra algae collection the condenser apparatus required for oil extraction and transesterification processes are soxhlet apparatus, reflux, steam distillation unit, round bottom flask, hot plate magnetic stirrer, thermometers, filter papers, conical flask, separatory flasks, distilled water and glassware of borosil make. the apparatus required for experimentation are utilised from the research fuel laboratory. 2.2 four-step process for preparation of spirogyra algae biodiesel 2.2.1. collection the naturally grown green spirogyra algae were collected from two open ponds for 120 days (every 15 days) at a recorded temperature ranging from 22 to 29 oc. as shown in figure 1, the algae species were identified in two different locations, the first location is near thathavarikittali reservoirs, garividi mandal, at 18°17'03.0"n latitude and 83°28'57.9"e longitude, and the second location are identified at 18°13′40.1′′n latitude and 83°28′58.1′′e longitude near meesalapeta, nellimarla, vizianagaram, andhra pradesh, india. 2.2.2 harvesting and processing dried algae powder and dried algae flakes the wet algae blooms collected from two locations weighed 158 kg and were cleaned with distilled water to remove unwanted plants, snails, mud etc., as shown in fig. 2. after processing, a net weight of 96 kg of wet algae is obtained and divided into two equal parts (48 kg each) to prepare into two forms: dried algae powder and dried algae flakes. the dried algae powder is processed as shown in figure 3. initially, spirogyra wet algae were separated layer by layer and dried for five (5) consecutive days at 8:1 sun hour during the temperature range from 22 to 29oc and later dried in a hot air oven for one hour by maintaining 90oc to remove excess moisture content. later, the dried algae layers were converted into dried algae powder (fig. 3) using a mechanical grinding machine available in the research lab. finally, the weight of the dried algae powder obtained was 8 kg. in a similar approach, the processing of dried algae flakes is shown in figure 4. firstly, the obtained wet algae were ground into a smooth paste. the smooth paste was transferred into a lab tray and exposed to sunlight for five (5) consecutive days. about 14 kg of dried algae flakes were obtained out of 48 kg of wet algae paste. furthermore, the spirogyra algae oil extraction was carried out using the solvent extraction method and oil expeller press method. finally, the algae oil was extracted using chemical and mechanical methods, and the oil yields were compared to identify the most productive method. fig. 2 algae cleaning process fig. 3 processing of dried algae powder fig. 4 processing of dried algae flakes 2.2.3 oil extraction in third-generation algae biodiesels, the oil extraction process plays a beneficial role in which the solid form of dried algae powder/flakes is transferred into a liquid (algae oil) form known as the extraction process (topare et al., 2011). the oil extraction process uses various chemical techniques like solvent extraction and supercritical (sc) co2 methods. similarly, physical and mechanical techniques like microwave heating, ball milling, ultrasonication and oil expeller methods etc (mubarak et al., 2015). according to the research that is currently accessible, oil expeller techniques, supercritical co2, and solvent extraction procedures are the most efficient ways to recover oil between 50 and 80 percent (konga et al., 2017). each oil extraction technique has its own benefits and drawbacks, as mentioned in t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 303 issn: 2252-4940/© 2023. the author(s). published by cbiore the literature that is currently available (bhargavi et al., 2018). in the present investigation, two techniques were proposed to extract algae oil and to compare the oil yield, viz., solvent extraction process (the chemical method i) and oil expeller press process (mechanical method ii). 2.2.3.1 solvent extraction process (method i) in the present study, spirogyra algae oil was extracted by a solvent extraction process using the soxhlet apparatus (konga et al., 2017). the solvent extraction process is the most popular method, renowned for its high extraction efficiency and ease of use. as shown in fig. 5a, the setup includes a heater control, round bottom flask, soxhlet apparatus, reflux condenser and glass beakers necessary for extraction purposes. in order to obtain the maximum benefit from the solvent oil extraction process, optimum conditions must be maintained throughout the process. the oil extraction process was initiated by mixing the dried algae powder and n-hexane in the proportions of 1:2. the prepared algae powder with a weight of 300 grams was filled in the soxhlet apparatus, and 600 ml of n-hexane solvent was transferred into the flask (kalyani et al., 2023) and the mixture is heated until the temperature reaches 65oc. upon heating, vapours flow up from the side tube through the reflux condenser unit to reach the soxhlet housing chamber, where vapours are condensed and interact with solid dried algae powder. as a result of the interaction, some of the desirable chemicals in algae dissolve in the hot solvent. after the reaction, the soxhlet apparatus was filled with the hot algae oil mixture. furthermore, the solvent and the algae oil mixture were transferred into the round bottom flask through a siphon tube. the process was repeated until the dried algae powder turned into pale colour for an extraction time of 18 hours. the hot algae oil and the n-hexane solvent were separated using a steam distillation process. finally, 64 ml of salo was obtained, and 90% of the n-hexane was recovered. finally, the percentage of salo yield was calculated using equation 1. % yield of algae oil (wt%) = mass of oil extracted (grams) the total mass of dried algae powder or flakes × 100 (1) 2.2.3.2 oil expeller process (method ii) the second method used to extract salo is a mechanical screw-type oil expeller press technique. in this process, the algae oil was extracted using dried algae flakes (see fig. 4) without using solvents. to extract algae oil from the oil expeller press method, initially, 48 kg of wet algae paste was processed and finally, 14 kg of dried algae flakes were obtained. these flakes were positioned directly into the oil expeller and crushed in the local oil crushing machine unit, as shown in fig. 5b. finally, from the oil expeller process, 1680 ml of salo was extracted and 12 % of the oil yield was obtained. though this method does not require any chemicals to extract oil, it requires a huge quantity of algae biomass and it is tough to recover the traces of oil stored under the expeller. finally, the oil extraction rate of 70 to 75% is noted (mubarak et al., 2015), which is less than the solvent extraction process (topare et al., 2011). hence, it is evident that the solvent extraction method is more effective than the oil expeller press. the spirogyra algae oil extracted using methods i and ii signify that the solvent extraction method is more effective than the oil expeller method. therefore, the solvent extraction process was used to extract spirogyra algae oil (salo), and the spirogyra algae oil biodiesel (salobd) was prepared using the transesterification process. 2.2.4 catalyst preparation the consumption of eggs has increased vigorously in recent years after the outbreak of the alarming national situation. worldwide, 76.7 to 85.2 mmt (million metric tons) of eggs are consumed, with india, the us, the eu, and china ranking as the top four producers (kolakoti & satish, 2020). these chicken eggshell wastes (cesw) were relinquished to surroundings causing a rapid increase in environmental pollution. proper usage of the best out of this waste helps to reduce ecological defilement. in this context, the hc prepared from the cesw (piker et al., 2016) is one of the significant steps for preparing neat salobd. a sample of 11 nos. of the whole chicken egg shell waste was collected near the neighbourhood restaurant. further, cleaned with deionised water and gently wiped with filter paper. further, the egg shells covered inside with shrill layers were removed and dried in a hot air oven for 24 hours. the dried cesw are crushed into tiny parts and calcinated in a muffle furnace by maintaining 600oc for 8 hours, as shown in fig 6. pandit & fulekar, (2017) prepared catalyst by maintaining it at 900oc for about 8 to 9 hours. similarly, kolakoti & satish, (2020) performed an experimental investigation to prepare lowgrade cooking oil using heterogeneous catalysts from waste egg shells. after several experimental trials, the hc was found desirable at 600oc. finally, the product cao was stored for its expedited usage. fig. 5 (a) stages of the solvent extraction process (method i), (b) stages of oil expeller press process (method ii) t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 304 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 6 preparation of chicken eggshell waste as heterogeneous catalyst. 2.2.5 design of experiments by response surface method (rsm) the response surface method is one of the most accepted optimisation techniques for experimental conditions. it is one of the beneficial approaches to show the actual and predicted values by varying the significant parameters. this method fits well in various applications such as biodiesel yield prediction, composites, nanocomposite designs, etc. sohail et al., (2020) performed experimental investigations for the production of spirogyra biodiesel and determined their physicochemical properties. the biodiesel yield was optimised using response surface methodology. chemical and enzymatic behaviour for spirogyra biodiesel was determined by considering four significant parameters, viz., molar ratio, catalyst concentration, time, and temperature. the optimised yield was reported as (77.3 ± 1.27%) and in enzymatic transesterification, the yield attained was (93.2 ± 1.27%). in another area of study, tran et al., (2020) worked on a central composite design based on response surface methodology to reach the optimal dyes removal conditions, using graphene oxide@cofe2o4 nanocomposite for congo red (cr), methyl red (mr), and crystal violet (cv). the results obtained from confirmation tests exhibited very low errors and outstanding removal efficiencies attained between 93.0% and 99.7%. a similar approach is carried out by (tran et al., 2019) using an efficient optimisation technique and response surface methodology to optimise the experimental conditions for removing chloramphenicol pharmaceutical from wastewater using cu3(btc)2-derived porous carbon as an efficient adsorbent. also, the quadratic model was used to assess compatibility and suitability using anova analysis. the results reported that a low magnitude p-value (<0.0001) was obtained with a coefficient of determination r2 = 0.9457, and adequate precision (ap) ratio was observed to be close to 0.1. nguyen et al., (2022) examined the construction of zeolitic-imidazolate framework (zif-8) and zif-8-derived porous carbon using response surface methods. the findings showed that the quadratic regression models are statistically significant. hence based on the literature studies, an rsm optimisation technique is used to estimate the actual and predicted value of the biodiesel yield by varying the significant parameters such as hc, mr, temperature and time with 29 design of experiments using minitab -19 software. the four significant process constraints are shown in table 2. table 2 four process constraints with ranges constraints units low medium high heterogeneous catalyst (hc) (wt.%) 0.6 1.2 1.8 molar ratio (mr) ---8:1 10:1 12:1 temperature oc. 38 48 58 time mins 120 150 180 t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 305 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 7 stages of spirogyra algae biodiesel oil production with heterogeneous catalyst table 3 design experimental runs with experimental and rsm biodiesel yield percentages experimental runs hc temperature time mr exp. yield % rsm yield % 1 1.8 48 150 12 53.65 54.82 2 1.2 58 180 10 93.54 89.84 3 1.2 58 150 12 92.21 89.80 4 0.6 48 150 8 91.04 88.23 5 1.2 38 150 12 77.69 77.98 6 1.2 58 120 10 76.44 76.35 7 1.2 48 150 10 75.8 75.71 8 0.6 48 120 10 86.98 86.63 9 1.2 38 120 10 79.21 82.21 10 1.2 58 150 8 74.85 73.97 11 1.2 48 180 12 92.63 94.87 12 1.2 48 150 10 75.47 75.71 13 1.2 48 150 10 74.58 75.71 14 1.8 48 180 10 62.01 62.71 15 1.2 38 150 8 89.84 91.65 16 1.8 38 150 10 71.24 66.98 17 1.8 48 150 8 55.89 57.18 18 1.2 48 120 12 64.69 66.34 19 1.8 58 150 10 52.47 54.00 20 1.2 48 150 10 75.54 75.71 21 0.6 48 150 12 95.69 92.75 22 0.6 38 150 10 91.65 91.42 23 0.6 58 150 10 92.99 95.54 24 1.2 48 150 10 75.25 75.71 25 1.2 48 120 10 76.99 73.21 26 1.2 38 180 10 90.45 89.84 27 1.2 48 180 8 75.24 75.83 28 1.8 48 120 10 52.14 51.71 29 0.6 48 180 10 95.97 96.76 t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 |306 issn: 2252-4940/© 2023. the author(s). published by cbiore 2.2.6 transesterification process the spirogyra algae oil (salo) obtained from the steam distillation process was heated up to 110oc, to remove the presence of water (h2o) content in the salo. furthermore, the salo fuel parameters were tested and it was reported that the kinematic viscosity was 19.20 mm2/s, density of 898.21 kg/m3, and ffa of 1.68 % (yadav et al., 2019). it is evident that the algae oil is reported with high kinematic viscosity, which is not recommended to use in the existing diesel engine. however, to reduce the kinematic viscosity of salo, a widely used transesterification process was performed to convert it into lowviscous salobd. initially, the raw salo oil was collected in a conical flask and positioned on a magnetic heater with a hot plate. further, heated until it reached the desired temperature and then a heterogeneous catalyst (cao) with methanol (ch3oh) was added. the mixture was stirred continuously using a magnetic stirrer, as shown in fig. 7. the catalyst (cao) was built up with an oxygen anion, which initiated to form a strong methoxide anion, thereby assisting the presence of triglycerides in the oil to transform diglycerides to monoglycerides. finally, esters and glycerine were formed. as represented in fig. 7, the hot mixture of biodiesel and heavy glycerine was transferred into a separatory flask. later, the glycerine was collected into a beaker, and the desired spirogyra algae oil biodiesel salobd was obtained. further, the salobd was washed with distilled water and shaken to remove the soap solution. finally, neat biodiesel was produced by heating the washed biodiesel to a temperature of 100 oc using a magnetic heater. the procedure was repeated by diversifying the four process constraints, as shown in table 3. the experimental approach of biodiesel yield (%) was calculated using equation 2. as shown in table 3, the rsm optimisation technique was repeated randomly with 29 experimental runs to obtain the predicted rsm biodiesel yield (%), which is presented in fig. 9 𝑆𝐴𝐿𝑂𝐵𝐷 𝑌𝑖𝑒𝑙𝑑(%) = 𝑊𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑠𝑝𝑖𝑟𝑜𝑔𝑦𝑟𝑎. 𝑎𝑙𝑔𝑎𝑒 𝑏𝑖𝑜𝑑𝑖𝑒𝑠𝑒𝑙 𝑊𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑠𝑝𝑖𝑟𝑜𝑔𝑦𝑟𝑎 𝑎𝑙𝑔𝑎𝑒 𝑜𝑖𝑙 × 100 (2) 3. result and discussions 3.1 comparison of salo yield using solvent extraction and oil expeller methods the spirogyra algae bloom was collected from two open ponds for 120 days (every 15 days) from january to may 2022, with a total weight of 158 kg. a net weight of 96 kg of fresh wet algae was obtained after cleaning and divided into two equal shares (48 kg each) to prepare dried algae powder and dried algae flakes. the oil is extracted using solvent extraction (method i) and an oil expeller press (method ii), and the oil yield percentages were estimated and compared. in solvent extraction, 48 kg of wet algae is separated layer by layer, from which 8 kg of dried algae powder was obtained. a sample of 300 grams of dried algae powder and 600 ml of n-hexane solvent was prepared in the ratio of 1:2 (dried algae powder: solvent), and 64 ml of salo was obtained after extraction. finally, from 8 kg of dried algae powder, 1707 ml of salo with an oil yield of 21.33% was obtained. whereas, in the oil expeller process, 48 kg of wet algae blooms were ground into a fine paste, transferred into a lab tray, and exposed to sunlight. 14 kg of dried algae flakes were obtained and pressed in the oil expeller. finally, 1680 ml of salo with 12 % of oil yield was obtained, as represented in table 4. therefore, based on the oil yield percentage, the solvent extraction method is more effective than the oil expeller method. it is evident from table 4 that the solvent extraction method i, with 21.33 % of algae oil yield, is effective compared with the oil expeller method ii, with 12 % of algae oil yield. 3.2 comparison of oil yield with different percentages of dried algae powder samples – method i three varied percentages (100 %, 75 %, and 50 %) of dried algae powder samples were used to estimate and compare the extracted oil yield with the effective solvent extraction process with the same extraction time and temperature. the salo yield obtained at various percentages of dried algae raw samples was presented in table 5. it is evident from table 5 that 21.33 % of algae oil yield was obtained from 100 % dried algae powder. similarly, 19.33 % of the algae oil yield was obtained from 75 % dried algae powder, and 18 % of the algae oil yield was obtained from 50 % dried algae powder. finally, the results confirmed that with 100% dried algae powder, the oil yield attained was higher and contained low moisture content compared with 75 % and 50 % dried algae powder. furthermore, when compared with the present study and reference study (konga et al., 2017), an increase of 4.55 % in algae oil yield was observed at 100% dried algae powder. similarly, 15.05 % improvement was observed near 75 % dried algae powder and 18.42 % improvement was observed near 50 % dry dried algae powder. the improvement in algae oil yield for three dryness samples, compared with the present and reference study, were indicated in fig. 8 table 4 algae oil extraction using chemical and mechanical methods description method i method ii method of extraction solvent oil expeller press apparatus soxhlet screw type treatment form spirogyra dried algae powder spirogyra dried algae flakes solvent mixing nhexane --- weight of the dried algae (kg) 8 14 algae oil obtained (litres) 1.707 1.680 % yield of algae oil (%) 21.33 12 t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 |307 issn: 2252-4940/© 2023. the author(s). published by cbiore table 5 effect of percentage dried algae sample and comparison of spirogyra algae oil yield obtained using a solvent extraction process no. description method i 1 dryness (%) of algae samples 100 75 50 2 extraction method solvent solvent solvent 3 form dried algae powder dried algae powder dried algae powder 4 time (hrs.) 18 18 18 5 temperature (oc) 65 65 65 6 sample wt. (kg) 0.3 0.3 0.3 7 total wt. of dried algae powder (kg) 8 8 8 8 total algae oil yield (litres) 1.707 1.624 1.512 9 % yield of algae oil (present study) 21.33 19.33 18 10 % yield of algae oil (konga et al., 2017) 20.4 16.80 15.20 11 improvement (%) 4.55 15.05 18.42 fig. 8 improvement in algae oil yield for three dryness samples compared with present and reference study 3.3 model fitting using the rsm optimisation technique, the four process constraints indicated in table 2 were varied for 29 design of experiments (doe), as shown in table 3. the maximum amount of experimental biodiesel yield of 95.97 % was achieved at mr (10:1), hc (0.6 %-wt), temperature (48oc), and time (180 minutes). however, for the same process constraints, the rsm biodiesel yield attained was 96.76 %, as shown in fig. 9. finally, three confirmation tests were conducted under the same conditions and revealed that 96.18% of the average salobd yield was achieved when compared with the experimental biodiesel yield. equation 3 represents a quadratic equation that was generated to calculate the relation between the salo biodiesel yield and the four process constraints for the 29 doe. the model summary of accuracy was estimated by the coefficient of determination. from fig. 10, the r2 value representing actual and predicted biodiesel yield for 29 doe attained was 97.51%, confirming ≥ 97% of the data was consistent with the achieved values. salo biodiesel yield % = 694 + 50.9 hc 9.39 temperature 2.498 time 46.06 mr 11.22 hc*hc + 0.0607 temperature*temperature + 0.00309 time*time + 0.394 mr*mr 0.838 hc*temperature + 0.0122 hc*time 1.44 hc*mr + 0.00488 temperature*time + 0.3689 temperature*mr + 0.1497 time*mr (3) furthermore, the adjusted r2 attained was 95.02 %, which affirms that the quality and fitness of the model picked were quite encouraging and concluded that the values predicted fit well with experimental values. hence the optimisation technique proved that the model is satisfactory in estimating the biodiesel yield from salo. 3.4 the analysis of variance (anova) the influence of four process constraints (hc, mr, temperature, and time) is beneficial in attaining maximum t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 308 issn: 2252-4940/© 2023. the author(s). published by cbiore biodiesel yield. the significant changes in process parameters during the transesterification process until biodiesel preparation were analysed using anova analysis. it was apparent in table 6 that the hc was observed as an influencing parameter, followed by the time. 3.5 free fatty acid analysis the free fatty acid (ffa) composition significantly impacted the fuel properties, specifically density, viscosity, cetane number, and calorific value. the fatty acid composition (fac) profile was tested using gas chromatography (gcms), and the results of the present study with the saturated and unsaturated composition were reported in table 7. it is evident from table 7 that the total saturated composition was reported as 68.39 wt.% and the unsaturated composition was 29.3 wt.%. as shown in fig.11 and fig. 12, the significant contribution of saturated fac was palmitic acid with 24.84 wt.%, and unsaturated fac was oleic acid with 26.12 wt.%. therefore, from table 7, it was apparent that the composition of the ffa profile in the present study was observed with very close compositions compared with the available literature studies. finally, the fac of the spirogyra biodiesel falls in good agreement compared with the literature studies. fig. 9 experimental and rsm biodiesel yield fig. 10 actual and predicted biodiesel yield interpretation t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 309 issn: 2252-4940/© 2023. the author(s). published by cbiore table 6 the analysis of variance (anova) source dof sum of squares mean square f-value p-value model 14 5060.84 361.49 39.14 <0.0001 linear 4 3905.85 976.46 105.73 <0.0001 hc 1 3567.99 3567.99 386.33 <0.0001 temperature 1 25.75 25.75 2.79 0.117 time 1 311.36 311.36 33.71 <0.0001 mr 1 2.93 2.93 0.32 0.582 square 4 462.46 115.62 12.52 <0.0001 hc*hc 1 107.01 107.01 11.59 0.004 temperature*temperature 1 241.52 241.52 26.15 <0.0001 time*time 1 47.00 47.00 5.09 0.041 mr*mr 1 14.65 14.65 1.59 0.228 2-way interaction 6 545.12 90.85 9.84 <0.0001 hc*temperature 1 101.10 101.10 10.95 0.005 hc*time 1 0.19 0.19 0.02 0.887 hc*mr 1 11.87 11.87 1.29 0.276 temperature*time 1 8.58 8.58 0.93 0.351 temperature*mr 1 217.71 217.71 23.57 <0.0001 time*mr 1 205.66 205.66 22.27 <0.0001 error 14 129.30 9.24 lack-of-fit 10 128.45 12.84 60.20 0.001 pure error 4 0.85 0.21 total 28 5190.14 table 7 comparison of fatty acid composition profiles composition type wt.% spirogyra algae (kumar et al., 2020) present study reference (chozhavendhan et al., 2020) reference (richmond, 2004) arachidic acid saturated 2.1 0.3 1.28 2.24 behenic acid saturated 1.55 0.2 1.39 0.33 capric acid saturated 0.41 0.3 0.3 caprylic acid saturated 0.55 0.3 0.3 lauric acid saturated 22.68 29.9 21.90 linoleic acid unsaturated 3.18 2.5 2.5 4.88 myristic acid saturated 13.05 17.0 15.29 oleic acid unsaturated 26.12 30.5 21.62 54.89 palmitic acid saturated 24.84 14.2 28.63 15.64 stearic acid saturated 3.21 3.0 4.86 2.10 other 2.31 1.8 1.28 3.6 fuel property analysis biodiesel fuel plays a crucial role in commercialisation and have the ability to compete with diesel fuels. the salobd produced from the transesterification process is characterised by vital fuel properties: density, kinematic viscosity, cetane number, calorific value, pour point, cloud point, flash point, and fire point, as shown in table 8. the earlier studies on biodiesel fuels stated that the vital properties of the fuel, like flash point, fire point, and cetane number, are higher than diesel fuel. this signifies a positive indicator to depot fuel and the retard time during combustion is lower for biodiesels (kalyani et al., 2023). also, biodiesel fuels with high viscosity, density, and low calorific value, report an adverse effect on the engine operating characteristics. one reason for these variations is the appearance of unsaturated and saturated fac. t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 310 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 11 gcms spectrum for palmitic acid fig. 12 gcms spectrum for oleic acid . table 8 key fuel properties of spirogyra algae oil biodiesel properties units astm d6751 (kalyani et al., 2023) diesel (kalyani et al., 2023)(saeed et al., 2021) salobd (present study) microalgae (kalyani et al., 2023); (kalyani et al., 2023)(pilotorodríguez et al., 2017) kinematic viscosity mm2/sec 3.5 to 5.0 2.75 4.24 2 to 5.2 density kg/m3 860 to 890 831 888 850 to 870 cetane number min 51 48 53.86 37 to 72 calorific value mj/kg 44 42.1 37 to 41 pour point oc -15 to -16 -13 5.8 6 cloud point oc -3 to -12 12.4 flash point oc min 120 86 145 115 fire point oc min 132 96 153 it is evident from table 8 that the kinematic viscosity and density of salobd are 54.18 % and 6.85 % higher compared to mineral diesel fuel. this change may result in an increase in fuel consumption and a decrease in brake thermal efficiency. the increase may be due to the presence of long-chain fatty acids, i.e. palmitic acid (fig. 11). the flash point and fire point temperatures were observed to be high in salobd. this is due to the rise in carbon number from saturated fatty acids like palmitic acids, a beneficial sign for storage and safe handling. furthermore, the cetane number (cn) of salobd was reported with an increment of 12.20 % than the diesel fuel due to the high composition of saturated fac. for better engine performance, the increment in cetane number is always recommended since it restricts the delay period during combustion. a low calorific value was identified in salobd compared to mineral diesel fuel due to the existence of zero sulphur and low hydrocarbons (kolakoti et al., 2022). lastly, the vital fuel properties of salobd, associated with the literature (piloto-rodríguez et al., 2017) observed to be in a close relationship, make biodiesel fuels suitable for the smooth operation of the existing diesel engine. 4. conclusions chicken eggshell waste (cesw) prepared as a heterogeneous catalyst (hc) was effectively tested on the spirogyra algae oil (salo) for the production of biodiesel. with the emergence of the rsm optimisation technique, a better yield was attained. based on the experimental investigations, the following conclusions were drawn. the spirogyra used in this study is naturally grown in an open pond, and it was collected every 15 days a month for 120 days without using any cultivation process, nutrients, or chemicals. the maximum salo yield obtained was 21.33% for the solvent extraction process and 12% salo yield for the oil expeller press process, which confirms that the solvent extraction process was more effective than the oil expeller method. it was observed that the oil yield percentage of salo achieved an increment of 4.55 % over the reference study. about 96.18 % of the salo biodiesel yield was attained during the transesterification process. the analysis of variance (anova) confirmed that the heterogeneous catalyst concentration followed by time is the most influencing constraint. the key fuel properties of the present study compared with the literature studies are observed within the t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 311 issn: 2252-4940/© 2023. the author(s). published by cbiore range of astm d6751. therefore, using chicken eggshell waste (cesw) for biodiesel production, especially for spirogyra algae oil, was recommended as a diesel fuel replacement to resist the energy and environmental calamities. author contributions: t.k.: conceptualization, data curation, investigation, methodology, formal analysis, writing—original draft, writing – review & editing., l.s.v.; formal analysis, project administration, resources., a.k.; supervision, methodology, writing— lay outing and editing, data visualization. all authors have read and agreed to the published version of the manuscript. funding: the authors declare that no funds, grants, or other support were received during the research. the authors have no relevant financial or non-financial interests to disclose. conflicts of interest: the authors declare no conflict of interest. references ananthi, v., balaji, p., sindhu, r., kim, s. h., pugazhendhi, a., arun, a. (2021). a critical review on different harvesting techniques for algal based biodiesel production. science of the total environment, 780, 146467. https://doi.org/10.1016/j.scitotenv.2021.146467 bhargavi, g., nageswara rao, p., & renganathan, s. (2018). review on the extraction methods of crude oil from all generation biofuels in last few decades. iop conference series: materials science and engineering, 330(1). https://doi.org/10.1088/1757899x/330/1/012024 bhateria, r., & dhaka, r. (2014). algae as biofuel. biofuels, 5(6), 607–631. https://doi.org/10.1080/17597269.2014.1003701 chisti, y. (2007). biodiesel from microalgae. biotechnology advances, 25(3), 294–306. https://doi.org/10.1016/j.biotechadv.2007.02.001 chozhavendhan, s., vijay pradhap singh, m., fransila, b., praveen kumar, r., & karthiga devi, g. (2020). a review on influencing parameters of biodiesel production and purification processes. current research in green and sustainable chemistry, 1–2(april), 1– 6. https://doi.org/10.1016/j.crgsc.2020.04.002 fan, m., wu, h., shi, m., zhang, p., & jiang, p. (2019). well-dispersive k2o–kcl alkaline catalyst derived from waste banana peel for biodiesel synthesis. green energy and environment, 4(3), 322–327. https://doi.org/10.1016/j.gee.2018.09.004 hadiyanto h., lestari s.p., widayat w. (2016). preparation and characterization of anadara granosa shells and caco3 as heterogeneous catalyst for biodiesel production. bulletin of chemical reaction engineering & catalysis, 11 (1), 21 26; https://doi.org/10.9767/bcrec.11.1.402.21-26 hariyanto, r. a. b., firmansyah, r. a., burhan, r. y. p., & zetra, y. (2021). synthesis of bio-additive for low sulphur diesel: transesterification of soybean oil and ethylene glycol using k2co3 catalyst. automotive experiences, 4(1), 44–50. https://doi.org/10.31603/ae.4694 höök, m., & tang, x. (2013). depletion of fossil fuels and anthropogenic climate change-a review. energy policy, 52, 797–809. https://doi.org/10.1016/j.enpol.2012.10.046 kalyani, t., prasad, l. s. v., & kolakoti, a. (2023). preparation and physicochemical properties of naturally grown green spirogyra algae biodiesel. chemical industry and chemical engineering quarterly, 29(1), 75–85. https://doi.org/10.2298/ciceq220215015k kalyani, t., sathya, l., prasad, v., & kolakoti, a. (2023). effect of triacetin as an oxygenated additive in algae biodiesel fuelled ci engine combustion, performance, and exhaust emission analysis. fuel, 338, 127366. https://doi.org/10.1016/j.fuel.2022.127366 kolakoti, a., & appa rao, b. v. (2020). relative testing of neat jatropha methyl ester by preheating to viscosity saturation in idi engine -an optimisation approach. international journal of automotive and mechanical engineering, 17(2), 8052–8066. https://doi.org/10.15282/ijame.17.2.2020.23.0604 kolakoti, a., jha, p., mosa, p. r., mahapatro, m., & kotaru, t. g. (2020). optimisation and modelling of mahua oil biodiesel using rsm and genetic algorithm techniques. mathematical models in engineering, 6(2), 134–146. https://doi.org/10.21595/mme.2020.21357 kolakoti, a., mosa, p. r., kotaru, t. g., & mahapatro, m. (2020). optimisation of biodiesel production from waste cooking sunflower oil by taguchi and ann techniques. journal of thermal engineering, 6(5), 712–723. https://doi.org/10.18186/thermal.796761 kolakoti, a., prasadarao, b., satyanarayana, k., setiyo, m., köten, h., & raghu, m. (2022). elemental, thermal and physicochemical investigation of novel biodiesel from wodyetia bifurcata and its properties optimization using artificial neural network (ann). automotive experiences, 5(1), 3–15. https://doi.org/10.31603/ae.6171 kolakoti, a., & satish, g. (2020). biodiesel production from low-grade oil using heterogeneous catalyst: an optimisation and ann modelling. australian journal of mechanical engineering, 00(00), 1– 13. https://doi.org/10.1080/14484846.2020.1842298 kolakoti, a., setiyo, m., & rochman, m. l. (2022). a green heterogeneous catalyst production and characterization for biodiesel production using rsm and ann approach. international journal of renewable energy development, 11(3), 703– 712. https://doi.org/10.14710/ijred.2022.43627 kolakoti, a., setiyo, m., & waluyo, b. (2021). biodiesel production from waste cooking oil: characterisation, modeling and optimisation. mechanical engineering for society and industry, 1(1), 22–30. https://doi.org/10.31603/mesi.5320 konga, a. k., muchandi, a. s., & ponnaiah, g. p. (2017). soxhlet extraction of spirogyra sp. algae: an alternative fuel. biofuels, 8(1), 29–35. https://doi.org/10.1080/17597269.2016.1196328 kumar, s., jain, s., & kumar, h. (2020). experimental study on biodiesel production parameter optimization of jatropha-algae oil mixtures and performance and emission analysis of a diesel engine coupled with a generator fueled with diesel/biodiesel blends. acs omega, 5(28), 17033–17041. https://doi.org/10.1021/acsomega.9b04372 mata, t. m., martins, a. a., & caetano, n. s. (2010). microalgae for biodiesel production and other applications: a review. renewable and sustainable energy reviews, 14(1), 217–232. https://doi.org/10.1016/j.rser.2009.07.020 mubarak, m., shaija, a., & suchithra, t. v. (2015). a review on the extraction of lipid from microalgae for biodiesel production. algal research, 7, 117–123. https://doi.org/10.1016/j.algal.2014.10.008 nguyen, d.t.c, vo, d.n., nguyen, t.t., nguyen, t.t.t., nguyen, l.t.t., tran, t.v. (2022). optimisation of tetracycline adsorption onto zeolitic–imidazolate framework-based carbon using response surface methodology. surfaces and interfaces, 28, 101549. https://doi.org/10.1016/j.surfin.2021.101549 nur m.m.a. and hadiyanto h.(2015) enhancement of chlorella vulgaris biomass cultivated in pome medium as biofuel feedstock under mixotrophic conditions. journal of engineering and technological sciences, 47(5), 487 497; https://doi.org/10.5614/j.eng.technol.sci.2015.47.5.2 pandit, p. r., & fulekar, m. h. (2017). egg shell waste as heterogeneous nanocatalyst for biodiesel production: optimised by response surface methodology. journal of environmental management, 198, 319–329. https://doi.org/10.1016/j.jenvman.2017.04.100 piker, a., tabah, b., perkas, n., & gedanken, a. (2016). a green and low-cost room temperature biodiesel production method from waste oil using egg shells as catalyst. fuel, 182, 34–41. https://doi.org/10.1016/j.fuel.2016.05.078 piloto-rodríguez, r., sánchez-borroto, y., melo-espinosa, e. a., & verhelst, s. (2017). assessment of diesel engine performance when fueled with biodiesel from algae and microalgae: an overview. renewable and sustainable energy reviews, 69(january 2016), 833–842. https://doi.org/10.1016/j.rser.2016.11.015 reddy, a., & majumder, a. b. (2014). use of a combined technology of ultrasonication, three-phase partitioning, and aqueous enzymatic oil extraction for the extraction of oil from spirogyra sp. journal of engineering; article id 740631. https://doi.org/10.1155/2014/740631 https://doi.org/10.1016/j.scitotenv.2021.146467 https://doi.org/10.1088/1757-899x/330/1/012024 https://doi.org/10.1088/1757-899x/330/1/012024 https://doi.org/10.1080/17597269.2014.1003701 https://doi.org/10.1080/17597269.2014.1003701 https://doi.org/10.1016/j.biotechadv.2007.02.001 https://doi.org/10.1016/j.crgsc.2020.04.002 https://doi.org/10.1016/j.gee.2018.09.004 https://doi.org/10.9767/bcrec.11.1.402.21-26 https://doi.org/10.31603/ae.4694 https://doi.org/10.1016/j.enpol.2012.10.046 https://doi.org/10.2298/ciceq220215015k https://doi.org/10.1016/j.fuel.2022.127366 https://doi.org/10.15282/ijame.17.2.2020.23.0604 https://doi.org/10.21595/mme.2020.21357 https://doi.org/10.18186/thermal.796761 https://doi.org/10.31603/ae.6171 https://doi.org/10.1080/14484846.2020.1842298 https://doi.org/10.14710/ijred.2022.43627 https://doi.org/10.31603/mesi.5320 https://doi.org/10.1080/17597269.2016.1196328 https://doi.org/10.1021/acsomega.9b04372 https://doi.org/10.1016/j.rser.2009.07.020 https://doi.org/10.1016/j.algal.2014.10.008 https://doi.org/10.1016/j.surfin.2021.101549 https://doi.org/10.1016/j.jenvman.2017.04.100 https://doi.org/10.1016/j.fuel.2016.05.078 https://doi.org/10.1016/j.rser.2016.11.015 https://doi.org/10.1155/2014/740631 t. kalyani et al int. j. renew. energy dev 2023, 12(2), 300-312 | 312 issn: 2252-4940/© 2023. the author(s). published by cbiore richmond, a. (2004). handbook of microalgal culture: biotechnology and applied phycology. blackwell science ltda blackwell publishing company. https://doi.org/10.1002/9780470995280 saeed, a., hanif, m. a., hanif, a., rashid, u., iqbal, j., majeed, m. i., moser, b. r., & alsalme, a. (2021). production of biodiesel from spirogyra elongata, a common freshwater green algae with high oil content. sustainability, 13(22), 1–10. https://doi.org/10.3390/su132212737 sohail, s., mumtaz, m. w., mukhtar, h., touqeer, t., anjum, m. k., rashid, u., ghani, w. a. w. a. k., & choong, t. s. y. (2020). spirogyra oil-based biodiesel: response surface optimisation of chemical and enzymatic transesterification and exhaust emission behavior. catalysts, 10(10), 1–12. https://doi.org/10.3390/catal10101214 supriyadi, s., purwanto, p., anggoro, d. d., & hermawan, h. (2022). the effects of sodium hydroxide (naoh) concentration and reaction temperature on the properties of biodiesel from philippine tung (reutealis trisperma) seeds. automotive experiences, 5(1), 57–67. https://doi.org/10.31603/ae.5986 topare, n. s., raut, s. j., renge, v. c., khedkar, s. v., chavan, y. p., & bhagat, s. l. (2011). extraction of oil from algae by solvent extraction and oil expeller method. international journal of chemical sciences, 9(4), 1746–1750. https://www.tsijournals.com/articles/extraction-of-oil-fromalgae-by-solvent-extraction-and-oil-expeller-method.pdf tran, t. van, nguyen, d. t. c., le, h. t. n., ho, h. l., nguyen, t. t., doan, v. d., nguyen, t. d., & bach, l. g. (2019). response surface methodology-optimised removal of chloramphenicol pharmaceutical from wastewater using cu3(btc)2-derived porous carbon as an efficient adsorbent. comptes rendus chimie, 22(11–12), 794–803. https://doi.org/10.1016/j.crci.2019.09.004 tran, t. van, nguyen, h. t. t., dang, h. h., nguyen, d. t. c., nguyen, d. h., pham, t. van, & tan, l. van. (2020). central composite design for optimising the organic dyes remediation utilising novel graphene oxide@cofe2o4 nanocomposite. surfaces and interfaces, 21, 100687. https://doi.org/10.1016/j.surfin.2020.100687 yadav, m., chavan, s. b., singh, r., bux, f., & sharma, y. c. (2019). experimental study on emissions of algal biodiesel and its blends on a diesel engine. journal of the taiwan institute of chemical engineers, 96, 160–168. https://doi.org/10.1016/j.jtice.2018.10.022 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1002/9780470995280 https://doi.org/10.3390/su132212737 https://doi.org/10.3390/catal10101214 https://doi.org/10.31603/ae.5986 https://www.tsijournals.com/articles/extraction-of-oil-from-algae-by-solvent-extraction-and-oil-expeller-method.pdf https://www.tsijournals.com/articles/extraction-of-oil-from-algae-by-solvent-extraction-and-oil-expeller-method.pdf https://doi.org/10.1016/j.crci.2019.09.004 https://doi.org/10.1016/j.surfin.2020.100687 https://doi.org/10.1016/j.jtice.2018.10.022 http://creativecommons.org/licenses/by-sa/4.0/ international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (3), 590-600 | 590 https://doi.org/10.14710/ijred.2023.52835 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id the application of equilibrium optimizer for solving modern economic load dispatch problem considering renewable energies and multiple-fuel thermal units hung duc nguyena,b* , khoa hoang truonga,b , nhuan an lea,b adepartment of power delivery, ho chi minh city university of technology (hcmut), 268 ly thuong kiet street, district 10, ho chi minh city, vietnam bvietnam national university ho chi minh city, linh trung ward, thu duc city, ho chi minh city, vietnam abstract. this study presents a modern version of the economic load dispatch (meld) problem with the contribution of renewable energies and conventional energy, including wind, solar and thermal power plants. in the study, reduction of electricity generation cost is the first priority, while the use of multiple fuels in the thermal power plant is considered in addition to the consideration of all constraints of power plants. two meta-heuristic algorithms, one conventional and one recently published, including particle swarm optimization (pso) and equilibrium optimizer (eo), are applied to determine the optimal solutions for meld. a power system with ten thermal power plants using multiple fossil fuels, one wind power plant, and three solar power plants is utilized to evaluate the performance of both pso and eo. unlike other previous studies, this paper considers the meld problem with the change of load demands over one day with 24 periods as a real power system. in addition, the power generated by both wind and solar power plants varies at each period. the results obtained by applying the two algorithms indicate that eo is completely superior to pso, and the solutions found by eo can satisfy all constraints. particularly in case 1 with different load demand values, eo achieves better total electricity production cost (tegc) than pso by 0.75%, 0.87%, 0.13%, and 0.45% for the loads of 2400 mw, 2500 mw, 2600 mw and 2700 mw. moreover, eo also provides a faster response capability over pso through the four subcases although eo and pso are run by the same selection of control parameters. in case 2, the high efficiency provided by eo is still maintained, though the scale of the considered problem has been substantially enlarged. specifically, eo can save $51.2 compared to pso for the minimum tegc. the savings cost is equal to 0.33% for the whole schedule of 24 hours. with these results, eo is acknowledged as a favourable search method for dealing with the meld problem. besides, this study also points out the difference in performance between a modern meta-heuristic algorithm (eo) and the classical one (pso). the modern metaheuristic algorithm with special structure is highly valuable for complicated problem as meld. keywords: economic load dispatch; particle swarm optimization; equilibrium optimizer; multiple fuels; thermal generator; renewable energies. @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 25th feb 2023; revised: 16th april 2023; accepted: 26th april 2023 ; available online: 1st may 2023 1. introduction economic load dispatch (eld) is considered one of the most crucial problems in power system operation. the determination of an optimal solution to eld allows the operators to save more operation cost and reduce environmental damage (xiang et al., 2021). nowadays, the concept of the conventional economic load dispatch (celd) is obsoleted and does not fit modern power systems anymore because of its static nature. in addition, celd only considers thermal generators as the sole generating source. however, the modern economic load dispatch (meld) was updated once multiple objective functions and renewable energy sources (res) (zhang et al., 2021; shen et al., 2019; li et al., 2020) were taken into account. specifically, several objective functions can be listed, such as reducing the total electricity generating cost (tegc), reducing the entire emissions (ree), etc. these objectives can be considered simultaneously or separately, depending on the different targets such as financial factor or/and technical factor. besides, both celd and meld have a lot in common, such as the set of related constraints and * corresponding author email: hungnd@hcmut.edu.vn (h.d.nguyen) the type of variable needed to be found. in terms of related constraints, several typical constraints can be named, such as the power balance constraint, the generator operational constraints, the multiple fuel constraint, etc. about the types of variables when dealing with both celd and meld problems, there are two of them, including the control and the dependent variables. these variables must be defined prior to any kind of computation. particularly with these mentioned problems, control variables are the power generated by all the existent generators in the system except for the first generator, which is considered the dependent variable. in the whole process of solving such celd and meld problems, an optimal solution is acknowledged if only both control variables and dependent variables satisfy all the related constraints with the minimum value of the fitness function. the types of res integrated with power systems are mainly solar and wind energies. therefore, the thermal, wind and solar power plants supply enough power to loads in meld, while only thermal power plants are in charge of the role in celd (hlalele et al., 2021; kim and kim, 2020). in research article https://doi.org/10.14710/ijred.2023.52835 https://doi.org/10.14710/ijred.2023.52835 mailto:hungnd@hcmut.edu.vn https://orcid.org/0000-0001-7831-3814 https://orcid.org/0000-0002-1237-0071 https://orcid.org/0000-0003-3880-471x http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.52835&domain=pdf h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |591 issn: 2252-4940/© 2023. the author(s). published by cbiore this study, meld problem with the integration of res into thermal power plants is solved. at the beginning of the foundation for the celd problem, the solution for the problem is commonly given by the old fashion computing methods such as the gradient method, the quadratic method, and the conventional iterative method. the real efficiency of these methods seems to be good and acceptable as the theorem study. however, at the time, meld is widely studied and the scale of the problem is enlarged so much over celd. besides, many complicate constraints are also taken into account. as a result, the whole complicated degree of meld is increased substantially. the application of old fashion computing methods begins showing more drawbacks such as a low response, a poor accuracy. luckily, in two past decades, computing methods have witnessed a huge leap forward in meta-heuristic methods to cope with highdegree complicated problems. there have been a huge number of meta-heuristic applied to meld problem, such as genetic algorithm (ga) (chen and chang, 1995), particle swarm optimization (pso) (park et al., 2005), artificial algae algorithm (aaa) (kumar and dhillon, 2018), modified differential evolution (mde) (nguyen et al., 2018), grey wolf optimal (pradhan et al., 2016), tunicate swarm optimizer (tso) in (hien et al., 2021), differential evolution (de) in (parouha and das, 2018), improved firework algorithm (ifa) (zare et al., 2021), quantum pso (qpso) (xin-gang et al., 2020), improved manta ray optimization (imro) (hassan et al., 2021), simplex search-based pso (ssm-pso) (chopra et al., 2021), improved bird swarm optimization (ibso) (fu et al., 2020), distributed roust optimization (dro) (chang et al., 2021), hybrid grey wolf optimizer (al-betar et al., 2020), pattern search and sequential quadratic programing-based genetic algorithm (ps-sqp-ga) (alsumait et al., 2010), double weight-based pso (dw-pso) (kheshti et al., 2018), improved bacterial foraging algorithm (ibfa) (pandit et al., 2012), acceleration coefficients-based pso (ac-pso) (ghasemi et al., 2019), nondominated sorting-based genetic algorithm (nsga) (basu, 2008), chaotic differential evolution (cde) (coelho and mariani, 2006), biogeography optimization (bo) (xiong and shi, 2018), and ameliorated dragonfly algorithm (ada) (suresh et al., 2019). these algorithms have shown a good performance as compared to deterministic algorithms based on lagrange relaxation, newton, gradient search (vaisakh and reddy, 2013; nguyen et al., 2019); however, these algorithms have not been applied to deal with the complex fuel cost functions of thermal units with three fuel options (chen et al., 2020; pham et al., 2022). multiple fuel options can be used for generating units in thermal power plants and the different options can bring more benefits to thermal power plants. but the units with several fuel options can bring more challenges to optimization tools since the generation of unit is not continuous within allowable ranges (dieu et al., 2013; jeyakumar et al., 2006). almost applied algorithms for the multiple fuel units are strong and not much dependent on lagrange function, excluding augmented lagrange hopfield network (alhn) (dieu et al., 2013). in fact, these applied methods are comprised of modified pso (jeyakumar et al., 2006), differential evolution (noman and iba, 2008), selfadaptive differential evolution (sde) (balamurugan and subramanian, 2007), adaptive real coded genetic algorithm (arc-ga) (amjady and nasiri-rad, 2010), and improved evolutionary programming (iep) (park et al., 1998). these algorithms have reported good results with minimum fuel cost and high stable search ability; however, the search speed was almost neglected once the comparison of setting parameters was not implemented. in this study, the original version of pso (kennedy and eberhart, 1995) and one modern meta-heuristic algorithm called equilibrium optimizer (eo) (faramarzi et al., 2020) are applied to determine the optimal solution for meld with multiple fuel options and renewable energies. the multiple fuel options are examined in the operation cost function belonging to thermal generators. the renewable generating sources, including one wind farm and three solar power plants, and the variation of power output values from these sources within 24 hours are included. the first priority of this study is to reduce the total operation cost as much as possible. the application of both eo and pso is considered to be a good example for evaluating the performance of a modern meta-heuristic algorithm and a classical algorithm. briefly, the novelties of this study can be seen on different aspects, including proposing the meld problem where varied load demands are employed with a day and the 24-single periods, the variation of power generated from both wind and solar power plants is examined throughout the 24 periods, and the difference in raw performance between the modern metaheuristic algorithm and the classical one is clarified and pointed out based on the results. the key contributions of the study are stated in the four claims. firstly, the proposed meld is successfully solved considering the presence of both solar and wind power plants. secondly, the optimal solution for meld is determined under the consideration of the multiple fuel constraints. thirdly, eo is proved to be the best applied method for the proposed problem through the comparisons of the results reached by pso and other previous studies. lastly, the superiority of a modern meta-heuristic algorithm (eo) over the classical one (pso) is proved and demonstrated by results and figures. in this study, two power systems with different complicated levels are considered for reaching the objective function of reducing the total electric generation cost over optimal schedule horizon, one hour for the first system and 24 hours for the second system. the two systems use the same ten thermal generating units in which each thermal generating unit can use two or three fuel types for electric production. the total cost values from the two systems are comparison criteria to evaluating the performance of eo and other algorithms. 2. problem descriptions. 2.1 main objective function as mentioned earlier, there are three types of generating sources in this study, including thermal power plants, wind power plants, and solar power plants. however, only the operation process of thermal generators consumes fossil fuels. so, the major target of the problem is to reduce the total electricity generation cost (tegc) from thermal power plants. the target can be formulated as follows. 𝑅𝑒𝑑𝑢𝑐𝑖𝑛𝑔 𝑇𝐸𝐺𝐶 = ∑ 𝐸𝐺𝐶𝑔 𝐺 𝑔=1 (1) where egcg ($/h) is the fuel cost of the gth thermal generator and formulated by (park et al., 2005): 𝐸𝐺𝐶𝑔 = 𝑔 + 𝛿𝑔𝑃𝐺𝑔 + 𝛾𝑔𝑃𝐺𝑔 2 𝑤𝑖𝑡ℎ 𝑔 = 1,…,𝐺 (2) tegc is the total fuel cost of all g generators working in one hour meanwhile 𝐸𝐺𝐶𝑔 is the fuel cost of one generator only. so, the unit of tegc is $ but that is $/h for egcg. the cost function of the thermal generator, 𝐸𝐺𝐶𝑔described in equation (2) is depicted at fig.1a. h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |592 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 1 models of fuel cost function for thermal units: a) one fuel option; b) multiple fuel options in addition, the multiple fuels aspect is also considered in this study, as shown in fig. 1b. according to (dieu et al., 2013), the mathematical expression of 𝐸𝐺𝐶𝑔 is represented by: 𝐸𝐺𝐶𝑔 = { 𝑔1 + 𝛿𝑔1𝑃𝐺𝑔 + 𝛾𝑔1𝑃𝐺𝑔 2 , 𝑓𝑢𝑒𝑙 1,𝑃𝐺𝑔1,𝑙𝑠 ≤ 𝑃𝐺𝑔 ≤ 𝑃𝐺𝑔1,ℎ𝑠 𝑔2 + 𝛿𝑔2𝑃𝐺𝑔 + 𝛾𝑔2𝑃𝐺𝑔 2 , 𝑓𝑢𝑒𝑙 2,𝑃𝐺𝑔2,𝑙𝑠 ≤ 𝑃𝐺𝑔 ≤ 𝑃𝐺𝑔2,ℎ𝑠 … 𝑔𝑤 + 𝛿𝑔𝑤𝑃𝐺𝑔 + 𝛾𝑔𝑤𝑃𝐺𝑔 2, 𝑓𝑢𝑒𝑙 𝑤,𝑃𝐺𝑔𝑤,𝑙𝑠 ≤ 𝑃𝐺𝑔 ≤ 𝑃𝐺𝑔𝑤,ℎ𝑠 (3) where 𝑔1, 𝛿𝑔1, and 𝛾𝑔1 are fuel consumption factors while using fuel 1 of generator g; 𝑃𝐺𝑔1,𝑙𝑠 and 𝑃𝐺𝑔1,ℎ𝑠 are the lowest and highest power generated by generator g whilst using fuel 1. similarly, 𝑔2, 𝛿𝑔2 and 𝛾𝑔2 are fuel consumption factors of generator g whilst using fuel 2. 𝑃𝐺𝑔2,𝑙𝑠 and 𝑃𝐺𝑔2,ℎ𝑠 are the lowest and highest power generated by generator g whilst using the fuel 2. finally, 𝑔𝑤, 𝛿𝑔𝑤, and 𝛾𝑔𝑤 are fuel consumption factors while using the fuel 𝑤 of generator g; and 𝑃𝐺𝑔𝑤,𝑙𝑠 and 𝑃𝐺𝑔𝑤,ℎ𝑠 are the lowest and highest power generated by generator g whilst using the fuel 𝑤. 2.2 constraints there are important constraints that must be respected while solving both celd and meld. they are described one by one as follows: • power balance constraint: this constraint is mainly about the relationship between the generating side and the demand side. according to (duong et al., 2021), the mathematical expression of the constraint is presented by eq 4: 𝑃𝐷 + 𝑃𝐿 = ∑ 𝑃𝐺𝑔 𝐺 𝑔=1 + 𝑃𝑊 + 𝑃𝑆 (4) in equation (4), pl is calculated by using the equation (5) below (dieu et al., 2013): 𝑃𝐿 = ∑ ∑ 𝑃𝐺𝑔𝐵𝑔ℎ𝑃𝐺ℎ 𝐺 ℎ=1 𝐺 𝑔=1 + ∑ 𝐵0𝑔𝑃𝐺𝑔 𝐺 𝑔=1 + 𝐵00 (5) • generator operational constraints: this constraint is about the working limit belonging to thermal generators. that means power output must allocate in the allowed range as depicted in equation (6). besides, the working limitation of a typical thermal generator is illustrated in fig. 2. according to (dieu et al., 2013), this constraint is formulated as follow: 𝑃𝐺𝑔,𝑚𝑖𝑛 ≤ 𝑃𝐺𝑔 ≤ 𝑃𝐺𝑔,𝑚𝑎𝑥 (6) • the electricity producing constraint of solar power plant: according to (phan et al., 2021) the power generated by solar power plant is limited as follows: ∑𝑃𝑆𝐺,𝑞 𝑁𝑆 𝑞 ≤ 80% × 𝑃𝐷 (7) 𝑃𝑆𝐺,𝑞 𝑚𝑖𝑛 ≤ |𝑃𝑆𝐺,𝑞| ≤ 𝑃𝑆𝐺,𝑞 𝑚𝑎𝑥 (8) 3. the applied methods to determine the optimal solution for meld in this study, two meta-heuristic methods, particle swarm optimization (pso) (kennedy and eberhart, 1995) and equilibrium optimizer (eo) (faramarzi et al., 2020) are applied to solve the given problem. pso is inspired by the foraging practice of animal swarms in real life such as fish, bird, ox, etc., while eo is inspired by physical law named equilibrium state of mass. the key difference between pso and eo is their update process for new solutions, which will be described in the following subsections: fig. 2 the working limitation of a thermal generator h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |593 issn: 2252-4940/© 2023. the author(s). published by cbiore 3.1 the original version of pso the update process for new solutions belonging to pso includes two main steps: the velocity update and the new position update. these steps are clarified by mathematical equations (9) and (10) as follows: 𝑉𝑖 𝑛𝑒𝑤 = 𝑉𝑖 + 𝑐𝑞1 × 𝛾1 × (𝑃𝐵𝑒𝑠𝑡,𝑖 − 𝑃𝑖) + 𝑐𝑞2 × 𝛾2 × (𝑃𝐺𝑏𝑒𝑠𝑡,𝑖 − 𝑃𝑖) (9) 𝑃𝑖 𝑛𝑒𝑤 = 𝑃𝑖 + 𝑉𝑖 𝑛𝑒𝑤 (10) 3.2 the equilibrium optimizer (eo) the update process of eo is conducted based on the references around the four best solutions at each iteration. the key elements of the whole process are described as follows: 𝑆𝑖 𝑛𝑒𝑤 = 𝑆𝑠 + (𝑆𝑖 − 𝑆𝑠)𝐸𝑥 + 𝐺𝑒 𝑣𝑟 (1 − 𝐸𝑥) 𝑤𝑖𝑡ℎ 𝑖 = 1, . . . ,𝑁𝑃 (11) in equation (11), the exponential term (ex) and the generation rate (ge) are calculated by using the equations below: 𝐸𝑥 = 𝜃1𝑠𝑖𝑔𝑛(𝜔 − 0.5)(𝑒 −𝑡𝑓.𝑣𝑟 − 1) (12) where 𝜃1 is set by 2, and the integer number (𝜔) is a random value between 0 and 1. besides, the time length factor (𝑡𝑓) is a value that changes with each iteration, and its value depends entirely on the maximum iteration number (𝐼𝑡𝑀𝑎𝑥) and the current iteration number (𝐼𝑡). 𝑡𝑓 can be determined by using the equation (13) 𝑡𝑓 = (1 − 𝐻 𝐻𝑀𝑎𝑥 ) 𝜃2( 𝐻 𝐻𝑀𝑎𝑥 ) (13) where 𝛽2 is set by 1 (faramarzi et al., 2020), 𝐻 is the current iteration, and 𝐻𝑀𝑎𝑥 is the maximum quantity of iteration. about the generation rate (ge), this term can be found by using the equation (14) below: 𝐺𝑒 = 𝐸𝑥.𝛿.(𝑆𝑠 − 𝑣𝑟.𝑆𝑖) (14) where, 𝛿 = { 𝑟𝑑3 2 𝑖𝑓 𝑟𝑑4 ≥ 𝑐𝑓 0 𝑒𝑙𝑠𝑒 (15) where 𝒄𝒇 is set by 0.5 (faramarzi et al., 2020). 4. results and discussions the study is conducted on a personal computer with a central processing unit (cpu) of 2.0 ghz and 8 gb of random-access memory (ram). the coding is implemented in matlab software with version 2018b. the two study cases are implemented as follows: study case 1: the system with ten thermal units using multiple fuels for four load demand cases, including 2400, 2500, 2600, and 2700 mw. data of the system is taken from (park et al., 1998). study case 2: one wind and three solar photovoltaic power plants are integrated into the above ten-unit system. the power output supplied by wind farms and solar power plants over one day are taken from the studies (zhang et al., 2017) and (augusteen et al., 2016), respectively. fig. 3 the results obtained by pso and eo after 50 independent runs corresponding to different load demand values: a) load of 2400mw, b) load of 2500mw, c) load of 2600mw, and d) load of 2700mw h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |594 issn: 2252-4940/© 2023. the author(s). published by cbiore 4.1 the determination of population and maximum iteration initially, the determination of optimal settings for the initial control parameters, including population and maximum iteration, is one of the first hurdles that must be overcome. to deal with the first hurdle, different simulations have been conducted before the certain values for the population and the maximum number of iterations are established. due to the high complexity posted by the set of constraints, the large scale of search space explicitly seen in case 2, and the non-convex characteristic of both celd and meld, eo and pso are executed with different settings of population and maximum iteration. for study case 1, the population is set to 30, 40, 50, and 60, respectively. the maximum iteration is also varied from 80, 90, 100, and 150, respectively. the results obtained by all experiments revealed that the value of the population strongly influences the quality of the solutions and the execution time of each iteration, while the iteration number highly involves the quality of the solutions and the execution time of an independent run. as a result, when the population is set to a high value, the quality of the solutions will be improved; however, each iteration will require more time to complete. if maximum iteration is fixed at a high value, the quality of the solution at the end of each independent run could be enhanced, but more time would be consumed to finish a run. more importantly, if both population and maximum iteration are set to large values, the applied methods will result in the same optimal solutions after a long execution time. in that circumstance, the performance of the two applied methods cannot be evaluated precisely and reliably. by analysing the results obtained by the mentioned experiments, 50 and 100 are considered the optimal setting for population and maximum iteration. these settings also perfectly serve the initial purpose of evaluating the performance of the eo and pso. study case 2 is more complicated than study case 1 by considering 24 periods instead of one period. the setting of case 1 is applied for each hour of case 2. so, case 2 takes more simulation time than case 1 as a result. 4.2 results of case 1 fig. 3 presents the results obtained by pso and eo for 50 trial runs. the four subfigures have the same characteristic that eo can reach approximately the same solution for over 50 runs but those from pso highly fluctuate. in addition, pso cannot reach the same best solution as eo. fig. 4 shows the best run of pso and eo for the four cases. specifically, for the first load demand case with 2400 mw, eo only needs over 50 iterations to reach the best value of the considered fitness function, while pso cannot perform the same even if the last iteration is used on this best run. for the last three cases of load demand, eo still maintains its fast-response capability over pso by reaching the fitness value with fewer iterations. particularly, with load demands of 2500, 2600, and 2700 (mw), eo also requires approximately 60 iterations to obtain the best value, while pso cannot achieve any similar value on the three comparisons. clearly, eo is much faster than pso in finding the most optimal generation solutions. fig. 5 shows the best cost for the four cases obtained by pso, eo and other methods. in the figure, pso is the worst method while eo can reach the same or slightly smaller cost than others. it should be emphasized here that eo only search solution by using 50 and 100 for population and iterations, whereas others must use higher than 100 for iterations excluding alhn (dieu et al., 2013), which is a deterministic algorithm. because of this evidence, the superiority of eo over other previous methods is not deniable, so this study focuses on analyzing the efficiency of eo and pso for each subfigure corresponding to each level of load demand. particularly, in the first case of load demand in fig. 5 with 2400 mw, the tegc achieved by eo is only $481.723, while the similar value obtained by pso is up to $485.317. the savings cost saved by eo over pso in this case is approximately $3.6, corresponding to 0.75%. on the three remaining load demand levels, the tgec values achieved by eo are, respectively, $526.239, $574.381, and $623.809. the similar values resulted by pso are, fig. 4 the best convergences obtained by pso and eo among 50 independent runs corresponding to different load demand values: a) load of 2400mw, b) load of 2500mw, c) load of 2600mw, and d) load of 2700mw h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |595 issn: 2252-4940/© 2023. the author(s). published by cbiore respectively, $530.826, $575.128, and $626.259. the savings costs saved by eo over pso in these cases are $4.587, $0.747, and $2.784, corresponding to 0.87%, 0.13%, and 0.45%, respectively. as stated in the study (nguyen et al., 2021), a method with better minimum, mean and maximum objective function is more effective than others. so, eo is more suitable than pso for the system. 4.3 the results of case 2 in the study case, the integrated system supplies electricity to loads over 24 hours as shown in fig. 6. in addition, the generation of the wind and solar photovoltaic power plants are also given in fig. 6. both pso and eo are applied to determine the optimal generation of the ten thermal generators and the summary of 50 runs is given in fig. 7. in the figure, three comparison criteria are employed to analyse the efficiency of the two applied methods while dealing with the larger scale of the meld problem, including the minimum tegc within 24 hours (minimum), average tegc within 24 hours (average), and maximum tegc within 24 hours (maximum). the green bars stand for the results obtained by pso, while the blue ones represent the similar values achieved by eo. for the first criterion, eo achieves $15384.5 of the tegc, while that of pso is $15435.7. it is easy to figure out that eo has saved $51.2, or 0.33%, over pso on this criterion in an operation day. next, in fig. 5 the comparison between two applied methods and other methods on different values of load demand: : a) load of 2400mw, b) load of 2500mw, c) load of 2600mw, and d) load of 2700mw fig. 6 load demand and generation of wind and solar plants over 24 hours 2 0 0 0 2 0 0 0 2 0 0 0 2 0 0 0 2 4 0 0 2 4 0 0 2 4 0 0 3 3 0 0 3 3 0 0 3 3 0 0 3 3 0 0 3 3 0 0 3 4 0 0 3 4 0 0 3 3 0 0 3 3 0 0 3 0 0 0 3 0 0 0 3 0 0 0 3 0 0 0 3 0 0 0 3 0 0 0 2 0 0 0 2 0 0 0 0 100 200 300 400 500 600 700 800 900 1000 0 500 1000 1500 2000 2500 3000 3500 4000 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 p o w e r fr o m s o la r a n d w in d p o w e r p la n ts ( m w ) p o w e r d e m a n d ( m w ) hour power demand (mw) wind power (mw) solar power (mw) h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |596 issn: 2252-4940/© 2023. the author(s). published by cbiore the average and maximum criteria, the savings costs saved by eo over pso are even larger. specifically, the savings costs in these criteria are $325.7 and $1132.1, corresponding to 2.12% and 7.31%, respectively. clearly, eo completely outperforms pso while dealing with the large-scale meld. fig. 8 shows the fuel cost at each hour for the best, mean, and worst solutions of over 50 solutions obtained by pso and eo, corresponding with the subfigures a, b, and c. in terms of the best solution presented in fig. 8a, eo always offers a better value of tegc per hour. particularly, regardless of the variation in load demand within 24 hours, the tegc values found by eo at each hour are lower than the ones obtained by pso at the same time. however, the differences between the tegc values resulting from eo compared to the similar ones belonging to fig. 7 summary of 50 runs found by pso and eo 15435.7 15720.3 16624.9 15384.5 15394.6 15492.8 14500 15000 15500 16000 16500 17000 minimum average maximum t e g c ( $ ) pso eo fig. 8 the tegc of each hour obtained by pso and eo: a) tegc of the best run, b) tegc of the mean run, c) tegc of the worst run h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |597 issn: 2252-4940/© 2023. the author(s). published by cbiore pso are not much in fig. 8a. hence, the outstanding characteristics of eo over pso are not clearly shown. in fig. 8b, the efficiency of a modern meta-heuristic algorithm such as eo starts to reveal itself. in the figure, the contrast in tegc values obtained by eo and pso starts to depart from each other in all hours and is easy to observe. eo always results in lower values of tegc over pso throughout the 24 hours. noticeably, the tegcs found by eo are always better than pso even in the high load demand hours. finally, the observations on fig. 8c indicate that the high efficiency of eo is enhanced when compared to pso. particularly, the better degree of tegc values resulted by eo can be seen vividly for 24 hours, regardless of high demand hours or low demand ones. in addition to that, the tegc values from 8 to 17 hours found by eo compared pso are substantially better than the same period as mentioned in fig. 8b. the most optimal generations for the ten thermal units corresponding to the best solution are reported in fig. 9. in general, the generation rate of the units is dependent on the load level at each hour and units with higher generation is more effective than units with lower generation. 4.3 discussion on performance of eo and on renewable energies eo is a metaheuristic algorithm mainly based on randomization, exploitation, and exploration. randomization is a general characteristic for approximately all algorithms belonging to the metaheuristic family. so, eo as well as pso are sensitive to settings of population, iterations and run number, leading to different results for the number of trials. because of the unexpected characteristic, eo has been implemented for 50 trials for each study case to summarize the best, mean and worst results for comparisons. fig. 4 above indicates the very stable ability of eo in reaching the best solution for four study cases. eo found the same solution quality for the 50 runs, which can be seen via the line of fuel cost, meanwhile pso must suffer very high fluctuation among the 50 runs, especially the deviation between the best and the worst runs. the two algorithms were tested by using the same population of 50 and the same iteration of 100. about the structure, eo only use one main equation (11) to update solutions, here they are generation of thermal units. pso has updated velocity and location in which velocity is equivalent to an increased interval and location is solution. the algorithms have the same characteristic of using randomization, but eo almost do not have sensitivity to the randomization. here, the setting of population and iteration are 50 and 100, not having enough impact on the change of eo but they highly influence the fluctuations of pso. as compared to other previous algorithms shown in previous studies, the best performance of eo cannot be shown in terms of reaching less fuel costs than these algorithms. eo only reached smaller cost than several algorithms such as sde, iep, and ardga. other algorithms did not show all simulated results as the study, so the full comparison is impossible to carry out. on the other hand, presentation of settings of iterations and population was not done in the studies too. however, the use of 50 and 100 for population and iteration is not high setting for the algorithm. and the comparison with pso is the evidence for this statement. pso could not reach the best solution although it was run 50 trials. so, eo is a very effective algorithm for this research. in the second case of testing eo performance, the supplied power to load over 24 hours with an additional supply from solar and wind power plants are both employed. the second case is much more complicated than the first case in terms of multiple hours and the presence of renewable energies-based generating units. the uncertainty of wind and solar have not been considered, and this is the major shortcoming of the study. however, the power sources have a high contribution to form a load curve for one day within 24 hours and reduce the fuel cost from thermal units. the fixed power of the renewable energiesbased units can guarantee energy security, i.e., load demand can be satisfied all hours. however, there are other cases that the real power of renewable power plants is smaller than predetermined values. this case is serious for the power system and frequency can be reduced to a smaller value than rated frequency. for this case, thermal units or battery energy storage system (bess) as considered by (kheiter et al., 2022) can supply more power to compensate for the lack of the renewable energies-based units. to reach the purpose, bess must save enough energy for discharge meanwhile the thermal units copes with the challenge of increasing power or starting up shut down units. this unexpected issue should be solved in the future work and the power system will become more effective in the future. besides, the uncertainties and the mathematical models of renewable energy sources, including solar and wind power plants, as considered in (kaur et al., 2021; khamharnphol et al., fig. 9 the most optimal generations of each thermal unit over 24 hours 0 500 1000 1500 2000 2500 3000 3500 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 p o w e r o u tp u t (m w ) hour unit 1 unit 2 unit 3 unit 4 unit 5 unit 6 unit 7 unit 8 unit 9 unit 10 h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |598 issn: 2252-4940/© 2023. the author(s). published by cbiore 2022), should be employed while solving the meld problem as another complex constraint for assessing the efficiency of a modern meta-heuristic algorithm such as eo. this implementation is also a great method to improve the overall quality of the study and make it closer to practice, where the penetration rate of these sources is growing. 5. conclusion in this study, one conventional ten-thermal unit power system and one integrated ten-thermal unit, three-solar photovoltaic plant and one wind-plant power systems are successfully solved by conventional meta-heuristic algorithm, named particle swarm optimization and a modern one, named equilibrium optimizer. the evaluation of results about the minimum and maximum total electricity generating cost values indicated that eo is completely superior to pso. specifically, eo provides a quick response capability, lower fluctuations of fitness values for fifty independent runs, and fast convergence to optimal results. therefore, eo has been acknowledged as a highly effective method for solving meld problems. in the future, the meld should be expanded with the consideration of higher-degree complicated constraints such as prohibited operation zone, valve point effects and uncertainties of renewable. in addition to highly valuable contributions above, this study also copes with the following shortcomings: ➢ other constraints of thermal generating units are still not evaluated on this study such as valve point effects, ramp-rate limits, prohibited zones, etc. ➢ the study only considers the case that both wind and solar power plants generate enough power to meet the demand as predicted. for other cases with a mismatch between the power generated by wind and solar power plants and the load demand, the evaluation in these cases is not taken place. as a result, no actions are proposed to deal with these scenarios while power from wind and solar power plants is lower or higher than load demand. ➢ the uncertainties of the wind and solar power plants is not clearly discussed and evaluated strictly. by fully acknowledging the shortcomings, there are many improvements that must be conducted in the future. ➢ more high-complex constraints related to meld must be taken into account. ➢ the uncertainties of power generated by wind and solar power plants must be fully evaluated while solving meld. ➢ more scenarios of power system operation must be employed and analyzed. especially, as there is a mismatch between the forecast and the real power production from wind and solar power plants. consequences as well as solutions for the cases have to be calculated and proposed to reach the least impact. nomenclature g the number of thermal generators 𝑔, 𝛿𝑔, and 𝛾𝑔 the fuel coefficients pgg the power output of thermal generator g (mw) 𝑃𝐷 the power demand required by the load (mw) 𝑃𝐿 the total power loss of the transmission process (mw) ∑ 𝑃𝑔 𝐺 𝑔=1 the total power generated by all thermal generators (mw) 𝑃𝑊 and 𝑃𝑆 the power supplied by wind and solar generators (mw) bgh, b0g, and b00 the loss factors pgg and pgh the power injected by the generators g and h (mw) 𝑃𝐺𝑔,𝑚𝑖𝑛 and 𝑃𝐺𝑔,𝑚𝑎𝑥 the working limitations of generator g (mw) ∑ 𝑃𝑆𝐺,𝑞 𝑁𝑆 𝑞 the total power generated by all solar power plants (mw) 𝑃𝑆𝐺,𝑞(mw) active power generated by solar power plant q (mw) 𝑃𝑆𝐺,𝑞 𝑚𝑖𝑛 and 𝑃𝑆𝐺,𝑞 𝑚𝑎𝑥 the minimum and maximum power generated by solar generator q (mw) np the population number 𝑉𝑖 𝑛𝑒𝑤 and 𝑃𝑖 𝑛𝑒𝑤 the new velocity and new position of the individual i. 𝑉𝑖 and 𝑃𝑖 the current velocity and current position of individual i qc1 and cq2 the accelerating factors 𝛾1 and 𝛾2 the random numbers in the interval between 0 and 1 𝑃𝐵𝑒𝑠𝑡,𝑖 and 𝑃𝐺𝑏𝑒𝑠𝑡,𝑖 the best position at the time considered and the best position at all times of the individual i. vr a random value in the range of (0, 1) 𝑟𝑑3 and 𝑟𝑑4 the random values in the interval between 0 and 1 𝑐𝑓 the comparative factor acknowledgments we acknowledge ho chi minh city university of technology (hcmut), vnu-hcm for supporting this study. author contributions: h.d.n.: methodology, formal analysis, writing—original draft, k.h.t.; supervision, resources, project administration, n.a.l.; conceptualization, writing—review and editing, project administration, validation. all authors have read and agreed to the published version of the manuscript. funding: this research was funded by ho chi minh city university of technology (hcmut), vnu-hcm. conflicts of interest: authors declare no conflict of interest. references al-betar, m.a., awadallah, m.a. & krishan, m.m. (2020). a non-convex economic load dispatch problem with valve loading effect using a hybrid grey wolf optimizer. neural computing and applications, 32 (16), 12127–12154. https://doi.org/10.1007/s00521-019-04284-9. alsumait, j.s., sykulski, j.k. & al-othman, a.k. (2010). a hybrid ga– ps–sqp method to solve power system valve-point economic dispatch problems. applied energy, 87 (5), 1773–1781. https://doi.org/10.1016/j.apenergy.2009.10.007. amjady, n. & nasiri-rad, h. (2010). solution of nonconvex and nonsmooth economic dispatch by a new adaptive real coded genetic algorithm. expert systems with applications, 37 (7), 5239– 5245. https://doi.org/10.1016/j.eswa.2009.12.084. augusteen, w.a., geetha, s. and rengaraj, r. (2016). economic dispatch incorporation solar energy using particle swarm optimization. paper presented at 2016 3rd international conference on electrical energy systems (icees). 2016. pp. 67–73. https://doi.org/10.1109/icees.2016.7510618. balamurugan, r. & subramanian, s. (2007). self-adaptive differential evolution-based power economic dispatch of generators with valvepoint effects and multiple fuel options. international journal of electrical and computer engineering, 1 (3), 543–550. https://doi.org/10.1007/s00521-019-04284-9 https://doi.org/10.1016/j.apenergy.2009.10.007 https://doi.org/10.1016/j.eswa.2009.12.084 https://doi.org/10.1109/icees.2016.7510618 h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |599 issn: 2252-4940/© 2023. the author(s). published by cbiore basu, m. (2008). dynamic economic emission dispatch using nondominated sorting genetic algorithm-ii. international journal of electrical power & energy systems, 30(2), 140–149. https://doi.org/10.1016/j.ijepes.2007.06.009. chang, x., xu, y., sun, h., khan, i. (2021). a distributed robust optimization approach for the economic dispatch of flexible resources. international journal of electrical power & energy systems, 124, 106360. https://doi.org/10.1016/j.ijepes.2020.106360. chen, c., zou, d. & li, c. (2020). improved jaya algorithm for economic dispatch considering valve-point effect and multi-fuel options. ieee access, 8: 84981–84995. https://doi.org/10.1109/access.2020.2992616 . chen, p.-h. & chang, h.-c. (1995) large-scale economic dispatch by genetic algorithm. ieee transactions on power systems, 10 (4), 1919– 1926. https://doi.org/10.1109/59.476058. chopra, n., brar, y.s. & dhillon, j.s. (2021). an improved particle swarm optimization using simplex-based deterministic approach for economic-emission power dispatch problem. electrical engineering, 103(3), 1347–1365. https://doi.org/10.1007/s00202-020-01164-7 . coelho, ld.s. & mariani, v.c. (2006) combining of chaotic differential evolution and quadratic programming for economic dispatch optimization with valve-point effect. ieee transactions on power systems, 21(2), 989–996. https://doi.org/10.1109/tpwrs.2006.873410. dieu, v.n., ongsakul, w. & polprasert, j. (2013) the augmented lagrange hopfield network for economic dispatch with multiple fuel options. mathematical and computer modelling, 57 (1), 30–39. https://doi.org/10.1016/j.mcm.2011.03.041. duong, m.q., nguyen, t.t. & nguyen, t.t. (2021) optimal placement of wind power plants in transmission power networks by applying an effectively proposed metaheuristic algorithm li, y. (ed.). mathematical problems in engineering, 2021, 1015367. https://doi.org/10.1155/2021/1015367. faramarzi, a., heidarinejad, m., stephens, b., & mirjalili, s. (2020). equilibrium optimizer: a novel optimization algorithm. knowledgebased systems, 191, 105190. https://doi.org/10.1016/j.knosys.2019.105190. fu, c., zhang, s. & chao, k.-h. (2020). energy management of a power system for economic load dispatch using the artificial intelligent algorithm. electronics, 9(1). https://doi.org/10.3390/electronics9010108. ghasemi, m., akbari, e., zand, m., hadipour, m., ghavidel, s., & li, l. (2019) an efficient modified hpso-tvac-based dynamic economic dispatch of generating units. electric power components and systems, 47 (19–20), 1826–1840. https://doi.org/10.1080/15325008.2020.1731876. hassan, m.h., houssein, e. h., mahdy, m.a., & kamel, s. (2021). an improved manta ray foraging optimizer for cost-effective emission dispatch problems. engineering applications of artificial intelligence, 100, 104155. https://doi.org/10.1016/j.engappai.2021.104155. hien, c.t., ha, p.t., phan-van, t. h., & pham, t. m. (2021). multiperiod economic load dispatch with wind power using a novel metaheuristic. gmsarn international journal, 16 (2022), 165–173. hlalele, t.g., zhang, j., naidoo, r. m., & bansal, r. c. (2021). multiobjective economic dispatch with residential demand response programme under renewable obligation. energy, 218, 119473. https://doi.org/10.1016/j.energy.2020.119473. jeyakumar, d.n., jayabarathi, t. & raghunathan, t. (2006). particle swarm optimization for various types of economic dispatch problems. international journal of electrical power & energy systems, 28 (1), 36–42. https://doi.org/10.1016/j.ijepes.2005.09.004. kennedy, j. & eberhart, r. (1995) “particle swarm optimization.” in proceedings of icnn’95 international conference on neural networks. 1995. pp. 1942–1948 vol.4. https://doi.org/10.1109/icnn.1995.488968. kheshti, m., ding, l., ma, s., & zhao, b. (2018). double weighted particle swarm optimization to non-convex wind penetrated emission/economic dispatch and multiple fuel option systems. renewable energy, 125, 1021–1037. https://doi.org/10.1016/j.renene.2018.03.024. kim, j. & kim, k.-k.k. (2020). dynamic programming for scalable justin-time economic dispatch with non-convex constraints and anytime participation. international journal of electrical power & energy systems, 123, 106217. https://doi.org/10.1016/j.ijepes.2020.106217. kumar, m. and dhillon, j.s. (2018). hybrid artificial algae algorithm for economic load dispatch. applied soft computing, 71: 89–109. https://doi.org/10.1016/j.asoc.2018.06.035. kheiter, a., souag, s., chaouch, a., boukortt, a., bekkouche, b., & guezgouz, m. (2022). energy management strategy based on marine predators algorithm for grid-connected microgrid. international journal of renewable energy development, 11(3), 751– 765. https://doi.org/10.14710/ijred.2022.42797. kaur, s., brar, y.s. and dhillon, j.s. (2021). short-term hydro-thermalwind-solar power scheduling: a case study of kanyakumari region of india. international journal of renewable energy development, 10(3), 635–651. https://doi.org/10.14710/ijred.2021.35558. khamharnphol, r. kamdar, i., waewsak, j., chaichan, w., khunpetch, s., chiwamongkhonkarn, s., & gagnon, y. (2022). microgrid hybrid solar/wind/diesel and battery energy storage power generation system: application to koh samui, southern thailand. international journal of renewable energy development, 12(2), 216–226. https://doi.org/10.14710/ijred.2023.47761. li, x., wang, w., wang, h., wu, j., fan, x., & xu, q. (2020). dynamic environmental economic dispatch of hybrid renewable energy systems based on tradable green certificates. energy, 193, 116699. https://doi.org/10.1016/j.energy.2019.116699. nguyen, t. t., vu, q. n., duong, m. q., & le, v. d. (2018). modified differential evolution algorithm: a novel approach to optimize the operation of hydrothermal power systems while considering the different constraints and valve point loading effects. energies, 11(3), 540. https://doi.org/10.3390/en11030540. nguyen, t. t., vo, d. n., tran, h. v., & le, v. d. (2019). optimal dispatch of reactive power using modified stochastic fractal search algorithm. complexity, 2019, article id 4670820, 28 pages. https://doi.org/10.1155/2019/4670820 . nguyen, t. t., nguyen, t. t., duong, l. t., & truong, v. a. (2021). an effective method to solve the problem of electric distribution network reconfiguration considering distributed generations for energy loss reduction. neural computing and applications, 33, 16251641. https://doi.org/10.1007/s00521-020-05092-2. noman, n. & iba, h. (2008). differential evolution for economic load dispatch problems. electric power systems research, 78 (8), 1322– 1331. https://doi.org/10.1016/j.epsr.2007.11.007. pandit, n., tripathi, a., tapaswi, s., & pandit, m. (2012). an improved bacterial foraging algorithm for combined static/dynamic environmental economic dispatch. applied soft computing, 12 (11), 3500–3513. https://doi.org/10.1016/j.asoc.2012.06.011. park, j.b., lee, k. s., shin, j. r., & lee, k. y. (2005). a particle swarm optimization for economic dispatch with nonsmooth cost functions. ieee transactions on power systems, 20 (1), 34–42. https://doi.org/10.1109/tpwrs.2004.831275. park, y.m., won, j.r. & park, j.b. (1998). a new approach to economic load dispatch based on improved evolutionary programming. engineering intelligent systems for electrical engineering and communications, 6 (2), 103–110. parouha, r.p. & das, k.n. (2018). economic load dispatch using memory based differential evolution. international journal of bioinspired computation, 11 (3), 159–170. https://doi.org/10.1504/ijbic.2018.091700. pham, l.h., dinh, b.h. & nguyen, t.t. (2022). optimal power flow for an integrated wind-solar-hydro-thermal power system considering uncertainty of wind speed and solar radiation. neural computing and applications, 34 (13), 10655–10689. https://doi.org/10.1007/s00521-022-07000-2 . phan, v.-d., duong, m.q., doan, m. m., & nguyen, t. t. (2021). optimal distributed photovoltaic units placement in radial distribution system considering harmonic distortion limitation. international journal on electrical engineering & informatics, 13 (2). https://doi.org/10.15676/ijeei.2020.13.2.7 pradhan, m., roy, p.k. & pal, t. (2016). grey wolf optimization applied to economic load dispatch problems. international journal of electrical power & energy systems, 83, 325–334. https://doi.org/10.1016/j.ijepes.2016.04.034. shen, x., zou, d., duan, n., & zhang, q. (2019). an efficient fitnessbased differential evolution algorithm and a constraint handling technique for dynamic economic emission dispatch. energy, 186, 115801. https://doi.org/10.1016/j.energy.2019.07.131. suresh, v., sreejith, s., sudabattula, s.k., & kamboj, v. k. (2019). demand response-integrated economic dispatch incorporating https://doi.org/10.1016/j.ijepes.2007.06.009 https://doi.org/10.1016/j.ijepes.2020.106360 https://doi.org/10.1109/access.2020.2992616 https://doi.org/10.1109/59.476058 https://doi.org/10.1007/s00202-020-01164-7 https://doi.org/10.1109/tpwrs.2006.873410 https://doi.org/10.1016/j.mcm.2011.03.041 https://doi.org/10.1155/2021/1015367 https://doi.org/10.1016/j.knosys.2019.105190 https://doi.org/10.3390/electronics9010108 https://doi.org/10.1080/15325008.2020.1731876 https://doi.org/10.1016/j.engappai.2021.104155 https://doi.org/10.1016/j.energy.2020.119473 https://doi.org/10.1016/j.ijepes.2005.09.004 https://doi.org/10.1109/icnn.1995.488968 https://doi.org/10.1016/j.renene.2018.03.024 https://doi.org/10.1016/j.ijepes.2020.106217 https://doi.org/10.1016/j.asoc.2018.06.035 https://doi.org/10.14710/ijred.2022.42797 https://doi.org/10.1016/j.energy.2019.116699 https://doi.org/10.3390/en11030540 https://doi.org/10.1155/2019/4670820 https://doi.org/10.1007/s00521-020-05092-2 https://doi.org/10.1016/j.epsr.2007.11.007 https://doi.org/10.1016/j.asoc.2012.06.011 https://doi.org/10.1109/tpwrs.2004.831275 https://doi.org/10.1504/ijbic.2018.091700 https://doi.org/10.1007/s00521-022-07000-2 https://doi.org/10.15676/ijeei.2020.13.2.7 https://doi.org/10.1016/j.ijepes.2016.04.034 https://doi.org/10.1016/j.energy.2019.07.131 h.d.nguyen et al int. j. renew. energy dev 2023, 12(3), 590-600 |600 issn: 2252-4940/© 2023. the author(s). published by cbiore renewable energy sources using ameliorated dragonfly algorithm. electrical engineering, 101 (2), 421–442. https://doi.org/10.1007/s00202-019-00792-y. vaisakh, k. & reddy, a.s. (2013). msfla/ghs/sfla-ghs/sde algorithms for economic dispatch problem considering multiple fuels and valve point loadings. applied soft computing, 13 (11), 4281–4291. https://doi.org/10.1016/j.asoc.2013.07.001. xiang, y., wu, g., shen, x., ma, y., gou, j., xu, w., & liu, j. (2021). low-carbon economic dispatch of electricity-gas systems. energy, 226, 120267. https://doi.org/10.1016/j.energy.2021.120267 . xin-gang, z., ze-qi, z., yi-min, x., & jin, m. (2020). economicenvironmental dispatch of microgrid based on improved quantum particle swarm optimization. energy, 195, 117014. https://doi.org/10.1016/j.energy.2020.117014. xiong, g. & shi, d. (2018). hybrid biogeography-based optimization with brain storm optimization for non-convex dynamic economic dispatch with valve-point effects. energy, 157, 424–435. https://doi.org/10.1016/j.energy.2018.05.180 . zare, m., narimani, m. r., malekpour, m., azizipanah-abarghooee, r., & terzija, v. (2021). reserve constrained dynamic economic dispatch in multi-area power systems: an improved fireworks algorithm. international journal of electrical power & energy systems, 126, 106579. https://doi.org/10.1016/j.ijepes.2020.106579. zhang, h., liang, s., ou, m., & wei, m. (2021). an asynchronous distributed gradient algorithm for economic dispatch over stochastic networks. international journal of electrical power & energy systems, 124, 106240. https://doi.org/10.1016/j.ijepes.2020.106240 . zhang, h., yue, d., xie, x., dou, c., & sun, f. (2017). gradient decent based multi-objective cultural differential evolution for short-term hydrothermal optimal scheduling of economic emission with integrating wind power and photovoltaic power. energy, 122, 748– 766. https://doi.org/10.1016/j.energy.2017.01.083. © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1007/s00202-019-00792-y https://doi.org/10.1016/j.asoc.2013.07.001 https://doi.org/10.1016/j.energy.2021.120267 https://doi.org/10.1016/j.energy.2020.117014 https://doi.org/10.1016/j.energy.2018.05.180 https://doi.org/10.1016/j.ijepes.2020.106579 https://doi.org/10.1016/j.ijepes.2020.106240 https://doi.org/10.1016/j.energy.2017.01.083 international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (3), 74-759 | 749 https://doi.org/10.14710/ijred.2023.49165 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id demand response based microgrid's economic dispatch muhammad hammad saeeda* , md sohel ranaa, md kausaraahmedb, claude ziad el-bayehc , wang fangzonga acollege of electrical engineering and new energy (ceene), china three gorges university (ctgu), yichang, china bcollege of electrical engineering and information, southwest petroleum university (swpu), chengdu, china cdepartment of electrical engineering, bayeh institute, amchit, lebanon abstract. the development of energy management tools for next-generation distributed energy resources (der) based power plants, such as photovoltaic, energy storage units, and wind, helps power systems be more flexible. microgrids are entities that coordinate ders in a persistently more decentralized fashion, hence decreasing the operational burden on the main grid and permitting them to give their full benefits. a new power framework has emerged due to the integration of ders-based microgrids into the conventional power system. with the rapid advancement of microgrid technology, more emphasis has been placed on maintaining the microgrids' long-term economic feasibility while ensuring security and stability. the objective of this research is to provide a multi-objective economic operation technique for microgrids containing air-conditioning clusters (acc) taking demand response into account. a dynamic price mechanism is proposed, accurately reflecting the system's actual operational status. for economic dispatch, flexible loads and air conditioners are considered demand response resources. then, a consumer-profit model and an ac operating cost model are developed, with a set of pragmatic constraints of consumer comfort. the generation model is then designed to reduce the generation cost. finally, a microgrid simulation platform is developed in matlab/simulink, and a case is designed to evaluate the proposed method's performance. the findings show that consumer profit increases by 69.2% while acc operational costs decrease by 18.2%. moreover, generation costs are reduced without sacrificing customer satisfaction. keywords: air-conditioning cluster (acc), demand response (dr), distributed generation (dg), economic operation, microgas turbine (mgt), microgrid (mg) @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 24th sept 2022; revised: 16th april 2023; accepted: 20th june 2023; available online: 30th june 2023 1. introduction microgrids are an integral part of today's smart grid. microgrids serve as an effective interface for distributed generators (dg), loads, energy storage systems (ess), and control units while sustaining dg's cost-effectiveness, environmental protection, and flexibility. they supply high-quality power to ensure that consumers' power usage is safe and reliable. the microgrid operation mode, on the other hand, is extremely adaptable (saeed et al., 2021a). for operation, mgs can be incorporated into a large power grid. they can also operate independently of the electric grid in autonomous/islanded mode. the remote islanded mgs operate in the event of a power grid breakdown to ensure the power supply of important loads (mhankale & thorat, 2018; liu et al., 2021; dashtdar et al. 2022; saeed et al, 2022). economy, reliability, and energy-saving are three characteristics that must be considered in the operation of islanded-operated microgrid systems (saeed et al., 2021b; recalde et al., 2020; xu et al., 2020; pothireddy et al., 2022). with the continuous development of microgrid control technologies, mgs have achieved safe and stable operation, and their reliability has been greatly guaranteed (saeed et al., 2022; salkuti 2022). therefore, researchers are paying more attention to the economics of mg operation (battula et al. 2021; wang et al. 2010). * corresponding author email: m.saeed003@outlook.com (m.h.saeed) the two types of economic power system dispatch are static dispatch and dynamic dispatch (ross et al., 1980; han, 2001; attaviriyanupap, 2002; basu, 2008). based on the system's operational conditions during each independent period, static economic dispatch determines the equipment's priority and mode of operation. the dynamic economic dispatch takes into account the scheduling cycle's lowest cost and coordinates across the various distribution generations (dgs) throughout several periods , so it is better adapted to the needs of a system in actual operation. researching the dynamic economic dispatch is therefore crucial. it is challenging to resolve the dynamic economic dispatch problem with renewable energy sources because they are prone to randomness and disruptions (calderaro et al., 2014). hsiao et al. (2021) developed an mg energy optimization approach that combined reliability and economy to realize the dynamic economic operation. salkuti et al. (2015), and salkuti (2017) presented a novel multi-objective day-ahead market clearing (damc) process that incorporates demand response offers while taking into account an extensive voltage-dependent load model. salkuti et al. (2016) also suggested a dynamic reserve activation strategy that is most effective in case of a line outage, a load rise, or both. he considered the spinning reserves provided by traditional thermal generators, hydropower units, research article https://doi.org/10.14710/ijred.2023.49165 https://doi.org/10.14710/ijred.2023.49165 https://doi.org/10.14710/ijred.2023.49165 mailto:m.saeed003@outlook.com https://orcid.org/0000-0002-2547-5016 https://orcid.org/0000-0002-8268-8878 https://orcid.org/0000-0002-9441-5148 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.49165&domain=pdf m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |750 issn: 2252-4940/©2023. the author(s). published by cbiore and load demands. salkuti (2018) developed an optimum emergency strategy utilizing the coordinated activity of slow and rapid reserves for the safe functioning of the power system at the least expensive and proved that the dynamic emergency reserve activation strategy has an advantage over the sequential reserve activation approach. murty & kumar (2020) developed a mathematical model for the economics of an mg consisting of a combined heat and power (chp) system that took environmental conditions into account. the simulation results show that scheduling batteries optimally can lower the system's overall running cost. however, it does not consider the reliability of the islanded operation of the system. as multi-objective optimization targets, ma et al. (2015) examines three indicators of active power loss, pollutant gas emissions, and system voltage stability. through modeling and analysis, it provides ideas for the monitoring and operation of smart grids. as a relatively cheap economic dispatch resource, demand response not only expands the mg's economic dispatch capabilities but also benefits the demand response users (vardakas et al., 2015), (rajkumar et al., 2011), (chen et al., 2011), (wang et al., 2020), and (pourbabak et al., 2018). jindal et al. (2018) proposed a novel analysis and management scheme that considers equipment adjustment variables, electrical priority indicators, and electrical product priority, as well as several algorithms for dr decision-making. ali et al. (2022) established price-based demand response models, including wind power, diesel engines, and energy storage systems to solve the optimal operating strategy through multi-objective wind driven optimization (mowdo), and verified the method's effectiveness through real-time simulations. to maximize the microgrid’s total profit, nguyen et al. (2018) proposed a novel smart grid pricing plan based on a demand response model including time-varying loads and finally used the alternating direction method of multipliers to solve the optimization problem analytically. hao et al. (2021) proposed a new chaotic binary gravitational search algorithm (ibgsa) for an islanded microgrid. the results show that implementing the demand response strategy in the optimal allocation process can effectively alleviate the systems’ investment cost, maintain the power supply’s reliability, and aggravate the renewable energy consumption. john et al. (2008) designed a framework for incorporating wind energy systems in the economic dispatch problem considering the overestimation and underestimation of the available power and numerically solved the optimization problem using the stochastic wind speed characterization based on the weibull probability density function. wu et al. (2014) built a chp microgrid system with various distributed energy resources and used the maximum comprehensive benefits as the objective function for dynamic economic dispatch. the objective function is solved using an improved particle swarm optimization (pso) technique paired with monte carlo simulation. zachar & daoutidis (2016) employed economic model predictive control to keep costs down while still achieving the constraints. the suggested dispatch approach can successfully reject forecasting errors and satisfy the defined energy exchange and storage level goals. imtiaz et al. (2021) implemented an incentive-based demand response model in a grid-connected microgrid and solved the economic dispatch problem with the dragonfly algorithm (da). the objective of the proposed method is to achieve the lowest fuel cost, the lowest transferable power cost, and the highest demand response benefit for the microgrid operator. the findings of da are compared to those of other modern algorithms such as the crow-search algorithm, the ant-lion optimizer, particle swarm optimization, and the genetic algorithm. in practice, a microgrid contains a variety of dg types. the dgs will exhibit various aspects in the dynamic economic dispatch when operating in various modes and according to various scheduling schemes. unpredictability and discontinuities will make the economic dispatch more challenging (kumar et al., 2019). this research suggests a dynamic electricity price mechanism based on the real-time electricity price mechanism that can more accurately reflect the operating status of the system for mgs with significant renewable energy penetration. 2. dynamic electricity price mechanism supply and demand must always be matched in power systems (ulbig et al., 2014). dynamic electricity tariffs are commonly mentioned as instruments for demand-side management to ensure this grid stability (dutta and mitra, 2017). the adoption of real-time pricing tariffs is attributed to a rise in overall economic efficiency because dynamic prices per unit of electricity reflect their short-run societal marginal costs of provision better than constant prices per unit of electricity do (borenstein, 2005). usually, the time-sharing tariff strategy is exploited to adjust the consumer's electricity consumption behavior according to the pre-set price. however, this strategy does not reflect the changes in the system's operating environment for real-time operation, particularly in the autonomous mgs having a high penetration rate of renewable energy resources. the presence of a large number of ders makes the operating environment of the autonomous mg greatly affected by the ambient and complex. the island's changing environment also reduces the accuracy of the time-sharing power price mechanism period. therefore, a dynamic electricity price mechanism which reflects the real-time operational status of the system can address the user's electricity consumption behavior more effectively. the real-time update of dynamic electricity price affects the access volume of user-side adjustable load, and the plausible dynamic tariff mechanism can promote demand-side response and maximize user-side revenue. electricity price is determined by the average marginal cost of system operation, while renewable energy resources and micro-gas turbines mainly provide the power in a microgrid. so, the dynamic electricity price mechanism consists of base electricity price, renewable energy power price adjustment, and micro gas-turbine power price adjustment. the base price is fixed and is the basis for forming the dynamic electricity price. the electricity price adjustment of renewable energy generation units is related to the actual power production by rers. the adjustment of the electricity price of the micro-gas turbine is related to the power produced by the micro-gas turbine. when the amount of renewable energy generation increases, the dynamic electricity price should be reduced, and the user side should be guided to increase the access volume of adjustable load to promote electricity consumption. therefore, dynamic electricity prices are negatively correlated with renewable energy power. the cost of generation of the micro-gas turbine increases with the increase of its generated power, which eventually increases the average operating cost of the microgrid. at this stage, to alleviate the burden on the generation-side of the microgrid, dynamic electricity prices should be increased to guide users to reduce the access volume of adjustable load. therefore, m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |751 issn: 2252-4940/©2023. the author(s). published by cbiore dynamic electricity price is positively correlated with micro-gas turbine force. to sum up, defining dynamic tariffs is as follows: 𝑐𝑑 (𝑡) = 𝑐𝑏 − 𝑐𝑟𝑒𝑔 (𝑡) + 𝑐𝑚𝑡 (𝑡) (1) where, 𝑐𝑑 (𝑡) represents the dynamic electricity price for the tperiod, 𝑐𝑏 represents the base price of electricity, 𝑐𝑟𝑒𝑔 (𝑡) represents the amount of electricity price adjustment for renewable energy generation in the t-period, 𝑐𝑚𝑡 (𝑡) represents the amount of electricity price adjustment for micro-gas turbine power generation during the t-period. 𝑐𝑟𝑒𝑔 (𝑡) is determined by the total amount of renewable energy power generation collected by the system during t−1 period, and the total amount of power generation by the microgrid is given by the following formula: 𝑐𝑟𝑒𝑔 (𝑡) = 𝑘𝑟𝑒𝑔 . ∑𝑝 𝑖 𝑟𝑒𝑔 (𝑡−1) ∑𝑝 𝑖 𝑟𝑒𝑔 (𝑡−1)+𝑝𝑖 𝑚𝑡(𝑡−1) (2) here, 𝑘𝑟𝑒𝑔 represents the electricity price adjustment factor for renewable energy generation, 𝑝𝑖 𝑟𝑒𝑔 (𝑡 − 1) represents the power output of the ith rer in the time t−1. 𝑝𝑖 𝑚𝑡 (𝑡 − 1) denotes the power output of the ith micro gas turbine during period t−1. it can be seen from equations (1) and (2) that when the mgt’s output remains unchanged with an increase of renewable energy generation, the amount of electricity price adjustment for renewable energy generation increases. and the dynamic price reduction encourages users to connect more adjustable loads to promote electric energy consumption. from equation (12), it can be obtained that the cost of generation of the mgt becomes a convex quadratic function of its output. the increase in the cost of generation from mgts will increase dynamic electricity prices. so, the exact definition of the electricity price adjustment coefficient of the mgt power generation as the second convex function of its output is shown in the following formula: 𝑐𝑚𝑡 (𝑡) = 𝑎. (∑pi 𝑚𝑡 (𝑡 − 1))2 + 𝑏. (∑pmt 𝑖 (𝑡 − 1)) (3) where a and b represent the electricity price adjustment coefficient of the mgt respectively. when the renewable energy power generation remains unchanged with the increase in output of the mgt, the electricity price adjustment amount of the mgt’s generated power increases and can be calculated by equation (2). the amount of electricity price adjustment for renewable energy power generation will reduce, so the dynamic electricity price will rise, limiting adjustable load access. based on the renewable energy generation and micro-gas turbine power generation in the previous period, the system determines the dynamic tariff to be executed in the next period and updates the dynamic electricity price before the following time period. 3. the multi-objective demand response model the conventional microgrid economic optimization model focuses on the microgrid power generation side or demand side and neglects the user's comfort. in the future, microgrid development needs to optimize the power generation side and demand side as a whole to improve the economy and reliability of microgrid operation. therefore, based on the proposed dynamic electricity price mechanism, this section establishes a multi-objective economic operation model that integrates the generationside and demand-side of the mg. for optimal demand side scheduling, the communities for adjustable loads and air conditioning are used as demand response resources. to maximize user benefits and reduce acc operating costs while maintaining user comfort, the airconditioning cluster's (acc) operating cost and adjustable user benefit model were developed. to ensure that the cost of power generation is kept to a minimum while the microgrid is in operation, a cost model for power generation has been devised. 3.1. user-side revenue model the revenue function on the user side consists of a utility function and a cost function. the utility function represents the relationship between the size of the power consumption of the load and the benefits generated. the parameters σi and ωi are introduced to describe the utility of different adjustable loads. the electricity utility of load i can be expressed as follows: 𝑈𝑖 (𝑡) = { 𝜎𝑖 𝑙𝑖 𝑓𝑙 (𝑡) + 𝜔𝑖 2 (𝑙𝑖 𝑓𝑙 (𝑡)) 2 , 0 < 𝑙𝑖 𝑓𝑙 (𝑡) < − 𝜎𝑖 𝜔𝑖 𝜎𝑖 2 2𝜔𝑖 , 𝑙𝑖 𝑓𝑙 (𝑡) ≥ − 𝜎𝑖 𝜔𝑖 (4) thereinto σi，ω𝑖 represent the parameters of adjustable load i, satisfying ω𝑖 > 0，σi > 0, determined by the type of load. 𝑙𝑖 𝑓𝑙 represents the adjustable load i during the time-period t, 𝑈𝑖 (𝑡) represents the electrical utility of adjustable load k, at 𝑙𝑖 < σi < ω𝑖 time. 𝑈𝑖 (𝑡) is directly proportional to the access amount 𝑙𝑖 𝑓𝑙 of the adjustable load i. when 𝑙𝑖 𝑓𝑙 ≥ σiω𝑖 , even if the adjustable load connection amount 𝑙𝑖 𝑓𝑙 continues to increase, the electricity utility 𝑈𝑖 (𝑡) of the adjustable load remains unchanged. the cost function represents the user's electricity bill, following the user's specific usage trends. the cost of electricity for adjustable loads can be expressed as follows: 𝐹𝑖 (𝑡) = 𝑐𝑑 (𝑡) − 𝑙𝑖 𝑓𝑙 (𝑡) (5) here, 𝐹𝑖 (𝑡) represents the cost of electricity for adjustable load i. the combined utility function and the cost function can be represented by the user-side revenue function as follows: 𝑃 = ∑ui(𝑡) − ∑fi(𝑡) (6) when the dynamic tariff increases, the utility of adjustable load will remain unchanged, and the cost of electricity will increase. this results in lower revenue on the user side, so the adjustable load will be reduced to ensure maximum benefit on the user side. when the dynamic tariff is reduced, access to adjustable load will increase due to the reduction of electricity charges to obtain more significant revenue on the user side. 3.2. air conditioning operating cost model airconditioning has periodic operating characteristics. the indoor temperature fluctuates up and down between user-set temperature ranges. when the air conditioning in cooling mode is on and running, the lower-temperature limit tdown is reached in the room. when the air conditioner stops working, the room temperature continues to rise. the air conditioner starts running again with the room temperature reaching the uppertemperature limit tup. the thermal power process of the air conditioner can be simulated by the equivalent thermal parameter model (etp), and the expression for the room temperature is as follows: m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |752 issn: 2252-4940/©2023. the author(s). published by cbiore { 𝑇𝑖 𝑡+1 = 𝑇0 𝑡 − (𝑇0 𝑡 − 𝑇𝑖 𝑡 )𝑒 − δt rc , 𝑆 = 0 𝑇𝑖 𝑡+1 = 𝑇0 𝑡 − ηpr − (𝑇0 𝑡 − ηpr − 𝑇𝑖 𝑡 )𝑒 − δt rc , 𝑆 = 1 (7) in the above expression, 𝑇𝑖 𝑡+1 and 𝑇𝑖 𝑡 represent t+1 and tmoment room temperature, respectively. 𝑇0 𝑡 represents the ambient temperature at the t-moment. p indicates the cooling/ thermal power (kw) when the air conditioner is operating, η represents the energy efficiency ratio of the air conditioner, c is the equivalent thermal capacity (j/c), r is the equivalent thermal resistance (c/w), δt represents the interval (s). 𝑆 indicates the operating status of the air conditioner, 𝑆 =0 indicates that the air conditioner is out of service, 𝑆 = 1 indicates that the air conditioner is in operation. the operating cost function of air conditioning consists of its electricity cost function and maintenance cost function. the electricity cost function is used to describe the power cost consumed when air conditioning is in operation, which can be expressed as follows: 𝑁𝑖 (𝑡) = 𝑐𝑑 (𝑡). 𝑝𝑖 𝐴𝐶 (𝑡) (8) thereinto 𝑝𝑖 𝐴𝐶 (𝑡) indicates the cooling/ thermal power of the ith air conditioner, 𝑁𝑖 (𝑡) represents its corresponding cost of electricity. the maintenance cost function of the air conditioner is used to describe the depreciation and maintenance costs generated by the operation of the air conditioner. considering the differences between different air conditioners, the maintenance cost can be defined as the quadratic convex function of the power of the air conditioner. it is expressed as follows: 𝑀𝑖 (𝑡) = 𝑚. (𝑝𝑖 𝐴𝐶 (𝑡)) 2 + 𝑛. 𝑝𝑖 𝐴𝐶 (𝑡) + 𝑙 (9) in above equation, m, n, and l are the maintenance cost factors for air conditioning, respectively. 𝑀𝑖 (𝑡) represents the maintenance cost of the air conditioner. the function of electricity cost and maintenance cost of integrated air conditioning can be obtained as follows: 𝑊(𝑡) = ∑ni(𝑡) + ∑mi(𝑡) (10) at the same time, influencing factors of electricity price are introduced into the comfort constraints of users so that the air conditioners’ set temperature is affected by the two factors of electricity price and user comfort. the set temperature can be formulated as: 𝑇𝑠𝑒𝑡 𝑡+1 = 𝑇𝑠𝑒𝑡 𝑡 + 𝐶1. 𝑐𝑑 −𝑐𝑑 𝑚𝑖𝑛 𝑐𝑑 𝑚𝑎𝑥−𝑐𝑑 𝑚𝑖𝑛 + 𝐶2. 𝑇𝑠𝑒𝑡 𝑡 −𝑇𝑠𝑒𝑡 𝑜𝑟𝑖 𝑇𝑢𝑝−𝑇𝑑𝑜𝑤𝑛 (11) in the above expression, 𝑇𝑠𝑒𝑡 𝑡+1 and 𝑇𝑠𝑒𝑡 𝑡 represent the air conditioning set temperature at t+1 and t-moment, respectively. 𝐶1 and 𝐶2 are the influence coefficients of electricity price and user comfort on the set temperature of air-conditioning. 𝑐𝑑 𝑚𝑎𝑥 and 𝑐𝑑 𝑚𝑖𝑛 are the maximum and minimum value of the electricity price, and tup and tdown are the upper and lower limits of the user's set temperature, 𝑇𝑠𝑒𝑡 𝑜𝑟𝑖 represents the initial set temperature of the air conditioner. the dynamic electricity price increases is directly proportional to the operating cost of air conditioning. at this time, the air conditioning temperature increases and reduces the operating time of air conditioning to save electricity costs. when the set temperature of air conditioning increases to a particular value, by the user's comfort limit, the setting temperature of air conditioning is reduced to ensure the user's comfort. 3.3. generation side cost model in the microgrids operating in an isolated mode, the cost of electricity generation from renewable sources is generally ignored, and only the generation cost from micro-gas turbines is considered. the generation cost is usually defined as the quadratic convex function of power output for conventional micro-gas turbines as (boyd et al., 2012): 𝐶𝑖 = αi(𝑝𝑖 𝑚𝑡 )2 + βi𝑝𝑖 𝑚𝑡 + γi (12) thereinto αi, βi, γi are i th mgt’s cost parameters, and meet αi > 0, βi > 0, γi > 0. 𝑝𝑖 𝑚𝑡 represents the active output of the micro-gas turbine i, 𝐶𝑖 represents the cost of generation for i th mgt. the objective function of the microgrid generation side can be expressed as follows: min 𝐶 = ∑ 𝐶𝑖 (𝑝𝑖 𝑚𝑡 )𝑛𝑖=1 (13) the incremental cost λi of micro-gas turbines is defined as the 1st derivative of its generating cost electricity ci w.r.t. active output pi, represented as follows (wood et al., 2013): λi = ∂ci(pi mt) ∂pi mt = 2αipi mt + βi (14) according to the criterion of equal small increase rate, when the incremental cost of all mgts is the same, the optimal output of each mgt is obtained. at this time, the cost of generation of the mg is the smallest. 4. constraints equality and inequality constraints are introduced to ensure the reliable operation of the mg. equality constraints ensure the power balance between the generation side and the consumption side of the microgrid during operation, which is a prerequisite for the stable operation of the mg. inequality constraints consider the output limit of micro-gas turbines in microgrids and the access limit of adjustable load. 4.1. power balance constraint for an mg, the total output of power generation from rers, esd, and mgts should be balanced with the total load on the user side, which can be expressed as follows: ∑pi mt(t) + ∑p i reg (t) + pess(t) = ∑li fl(t) + ∑li cl(t) + ∑pi ac(t) (15) here, pi reg (t) represents the output power of the ith rer unit at the t-moment, pess(t) represents the output of the esd at tmoment. 4.2. mgt’s output constraint the output of the mgt is limited by its rated capacity and other physical conditions. to ensure an adequate output of the mgt during the optimization process, the constraints are set as follows: min(pi mt) ≤ pi mt(t) ≤ max(pi mt) (16) here, min(pi mt) and max(pi mt) represent the minimum and maximum output of the ith mgt. m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |753 issn: 2252-4940/©2023. the author(s). published by cbiore 4.3. adjustable load constraint the access volume of the adjustable load is adjusted with the change of the dynamic tariff to maximize the user-side revenue. due to the limitation of the physical conditions of the adjustable load, the access volume is restricted as follows: min(li fl) ≤ li fl(t) ≤ max(li fl) (17) hereinto 𝑚𝑖𝑛(𝑙𝑖 𝑓𝑙 ) and 𝑚𝑎𝑥(𝑙𝑖 𝑓𝑙 ) represent the minimum and maximum access volumes of the ith adjustable load. 4.4. user comfort constraints the indoor temperature should be maintained within the user's comfort level. when the indoor temperature exceeds the uppertemperature limit, the air conditioner starts working to reduce the room temperature. when the indoor temperature is less than the lower temperature limit, the air conditioner stops working. when the indoor temperature is within the comfort zone, the air conditioner stops working or keeps working according to the control command. at the same time, influencing factors of electricity price are introduced into the comfort constraints of users so that the set temperature of the air conditioner is affected by the two factors of electricity price and user comfort. the set temperature of the air conditioner can be expressed as follows: tdown < ti(t) < tup (18) in the formula, tup and tdown represent the upper and lower temperature limit set by the user. the upper limit is 2 degrees higher than the user set temperature, and the corresponding lower limit is 2 degrees lower than the user set temperature. 5. simulation model a radial mg operating in islanded mode is established in matlab/simulink to verify the effectiveness of the proposed multi-objective economic dispatch method. the microgrid simulation model consists of 12 distributed power generation units and 12 loads, and its topology is shown in figure 1. the distributed power generation units include micro gas turbines {dgi| i=1,3,5,7,11}, renewable energy generation units {dgi| i=2,6,8,9,10,12} and energy storage systems dg4. there are three types, namely fixed loads {loadi| i=4,8,10,12} adjustable load {loadi| i=1,3,5,11} and acc load {aci| i=2,5,6,9}. renewable energy power generation units, including photovoltaics and wind turbines, are considered to work in maximum power point tracking (mppt) mode. considering fig 1. the topology of microgrid simulation platform https://www.hanspub.org/journal/paperinformation.aspx?paperid=35174#f1 m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |754 issn: 2252-4940/©2023. the author(s). published by cbiore that this paper studies the dynamic economic dispatch of microgrids, the renewable energy power generation units are set not to generate reactive power. in the simulation process, the output curve of the renewable energy power generation unit is shown in fig. 2. the micro gas turbine works in the pq control mode to compensate for the power gap between renewable energy generation units and the load. since the islanded microgrid lacks the support of the main grid, the energy storage system works in the v/f mode as the voltage and frequency support of the microgrid. taking into account the energy storage system’s capacity and output power limitations, a micro gas turbine is set to share the energy storage output. so, the energy storage system only supplies power when the load changes suddenly, and then its output power is shared by the micro gas turbine. resultantly, the overcharge and discharge of the energy storage system are avoided, and the safe and stable operation of the microgrid is ensured. the air-conditioning load in the microgrid is assumed to be inverter air-conditioning with a maximum rating of 3 kw. in addition, the air conditioners in each area are centrally controlled by the corresponding central controller, and the indoor temperature in the same area remains the same. the specific parameters of air conditioning load are shown in table 1. the related parameters of distributed generation units (dg) and loads in the microgrid are listed in table 2 and table 3, respectively. the line losses are introduced in the system to make a real-time simulation model for the microgrid. the line impedance is set to 0.641 + j0.101 ω/km, and the system voltage and frequency are set to 380 v and 50 hz, respectively. fig 2. outputs of renewable energy generation units table 1 setups and parameters of ac serial # no. of ac units initial temperature (0c) 𝒎 𝒏 𝒍 2 15 24 -0.328 3.42 0.80 6 12 24.5 -0.174 2.86 0.83 7 8 23 -0.360 2.32 1.00 9 10 25 -0.198 2.99 1.50 table 2 dg parameters serial # active/ reactive ratings (kw, kvar) mode 𝜶 𝜷 𝜸 1 60, 40 pq 0.059 6.71 80 2 30, 0 mppt 3 55, 42 pq 0.066 6.29 43 4 30 ah v/f 5 65, 50 pq 0.046 7.53 35 6 50, 0 mppt 7 60, 48 pq 0.069 4.57 48 8 35, 0 mppt 9 45, 0 mppt 10 50, 0 mppt 11 60, 45 pq 0.058 0.058 54 12 55, 0 mppt m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |755 issn: 2252-4940/©2023. the author(s). published by cbiore 6. model’s solution based on the specifications of the microgrid simulation model, the matlab function module in the user-defined functions library is added to simulink. then user benefit model, airconditioning operating cost model, and power generation cost model are established in the module. the centralized interiorpoint optimization function 'fmincon' is called to solve the model; a function minimizer with linear and nonlinear constraints. the specific steps are as follows: a) read the current operational time of each micro gas turbine, energy storage system output, adjustable load, acc power, indoor temperature, ambient temperature, and dynamic electricity price. b) based on the current dynamic electricity price, the centralized optimization function 'fmincon' solves the user side revenue model and the air-conditioning operating cost model. and the optimal connection plan of the adjustable load and the optimal operating power of the acc is obtained. c) the output of the v/f controlledenergy storage system is allocated to the micro gas turbine to avoid overcharge and discharge of the energy storage system to ensure the microgrid's stable operation. d) the total output of micro gas turbines is updated. the centralized optimization function 'fmincon' is called to solve the power generation cost model to ensure that all micro gas turbines have incremental costs and minimize power generation costs. e) the adjustable loads, the operating power of the acc, and the output power of the micro gas turbine are updated and the arrival of the following optimization period is entertained in the same way. 7. results & discussion a simulation study is designed for the microgrid simulation model to verify the validity of the proposed multi-objective economic scheduling model. the simulation duration is 6 hours. the operation and economic benefits of the microgrid under normal conditions and multi-time economic dispatch strategy are compared and analyzed, and the simulation results under different strategies are shown in the following figures. comparing the dynamic electricity price and the change of adjustable load in fig. 3 (a) and (b), it can be seen that when the dynamic price increases, the amount of variable load access decreases and the cost of electricity for adjustable load decreases. while the cost of the adjustable load has decreased, fewer devices modify their consumption in response to price table 3 load parameters serial # load type load range (kw) 𝝎 𝝈 1 adjustable 036 -0.123 9.625 2 acc 045 3 adjustable 048 -0.163 13.02 4 fixed 30 5 adjustable 024 6 acc 036 -0.198 13.12 7 acc 0-24 8 fixed 30 9 acc 0-30 -0.207 10.99 10 fixed 35 11 adjustable 0-28 12 fixed 40 (a) (b) fig. 3 (a) connected amount of flexible loads (b) the incremental cost of mts and the dynamic price https://www.hanspub.org/journal/paperinformation.aspx?paperid=35174#f3 m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |756 issn: 2252-4940/©2023. the author(s). published by cbiore signals. this is due to the ability of users to move their electricity usage to cheaper times of the day. when demand is strong or the supply is constrained, both impacts result in a more balanced and effective use of electricity. when the dynamic electricity price decreases, the amount of adjustable load access increases. although the adjustable load electricity cost increases, the utility of the adjustable power load increases more than the cost of electricity, so the total income of the users increases. the expansion of adjustable load access enables consumers to switch their consumption to cheaper times of the day (julia & victor 2022, maximilian 2022). . the increased utility gained by using power during cheaper times leads to an overall rise in total income or benefit for users, notwithstanding the possibility of an increase in the cost of electricity for adjustable loads. fig. 3(b) shows the incremental cost curve of micro-gas turbines, which shows that the incremental costs of micro-gas turbines are consistent throughout the simulation period for all the values of 𝛌, thus minimizing the total cost of power generation on the power generation side of the microgrid. the reason is that they may modify their power output while a demand response program is active based on the grid's requirements and pricing cues. the turbines can continue to operate steadily and efficiently during the simulation time because these modifications often take place within a narrow range. in fig. 3(b), the dynamic cost decreases drastically after 4 hr due to rise in incremental cost of micro-turbine, and then increases gradually after 7 hrs with the decline in the incremental cost of micro-turbine. fig. 4 (a) and (b) show the user room temperature variation and acc power variation with and without introducing the proposed method. it is clear that under the proposed method, the power consumption of the acc is lower than that without the introduction of the proposed method with the constraints of user comfort. the increased temperature variation is caused by equipmentcycling, in which the cooling or heating system frequently switches from on to off operation. during times of peak demand, this cycle can be employed to reduce overall electricity use. as the system alternates between times of active cooling or heating and periods of inactivity, it can also result in temperature changes within the space. there are two reasons behind the decreased acc power variation: • setpoint adjustments: in order to balance the grid's demand for electricity, the dr program requires consumers to change the setpoints on their thermostats or permit a wider range of temperatures. users may need to tolerate slightly higher room temperatures than their preferred comfort levels. the units may cycle on and off more frequently to maintain the greater temperature range as a result of these setpoint alterations, which might lead to increased power variances within the air conditioning cluster (david et al. 2015).. • consumer response heterogeneity: users' reactions to dr signals can differ, resulting in various power fluctuation patterns within the acc. while some users may have more variable usage patterns, others may be more receptive to price signals or demands for demand reduction, leading to higher power variations. this fluctuation in user responses may be a factor in the cluster's higher power variability(waseem et al. 2021).. figure 5 shows the voltage and frequency of the microgrid system and the output of the energy storage system. it is clear from the figure that under the proposed economic dispatch strategy, the voltage and frequency of the system remain stable. for a microgrid's safe and reliable operation, certain voltage limitations must be maintained. the proposed economic dispatch method takes these restrictions into account and allocates generation resources to keep voltage levels within reasonable bounds. the economic dispatch approach contributes to system stability by sticking to these voltage constraints and avoiding voltage instability or voltage collapse. only when the load changes suddenly, there will be small fluctuations, but it will return to normal soon. the economic dispatch strategy aims to balance the generation and load within the power system. by optimizing the allocation of generation resources, it ensures that the total power supplied matches the total power demanded in real-time. this balance helps maintain a stable frequency as the generation and load fluctuations are minimized. dynamic response capabilities allow energy storage systems to respond quickly to changes in load or generation, helping to regulate frequency and voltage within the microgrid. the energy storage system only outputs or absorbs power to maintain the power balance of the microgrid system when the load changes suddenly. but the output is quickly shared by the micro gas turbine and returns to zero, thus avoiding the overcharge and discharge of the energy (a) (b) fig. 4 (a) the room temperature (b) the power of ac m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |757 issn: 2252-4940/©2023. the author(s). published by cbiore storage system to ensure the safety and stability of the microgrid operation fig. 6 compares the benefits of adjustable load users and the operating costs of air-conditioning communities. it can be seen from the figure that the introduction of the demand response economic dispatch method reduces the operating cost of acc. it can be seen from table 4 that the adjustable user benefit of introducing the demand response economic dispatch method is 69.2% higher than the benefit of the user when it is not introduced, and the operating cost of the acc is reduced by 18.2%. the reason is that the proposed method enables load shifting, where the air conditioning cluster can adjust its cooling operation to shift the peak demand to lower-demand periods. by reducing or limiting cooling during peak hours, when electricity prices are typically higher, the cluster can avoid costly peak demand charges and achieve cost savings. fig. 6 the profit of consumers and operation cost of ac table 4 user benefits and air conditioning operating costs scheduling method user benefit (cents) acoperating cost (cents) dr 3,126 1,459 without dr 1,848 1,785 8. conclusion the problem is the current time-of-use electricity price mechanism cannot reflect the real-time status of the microgrid system. so, this paper proposes a dynamic electricity price mechanism that can better reflect the operating status of the microgrid under a high rate of renewable energy penetration. then, air conditioners and adjustable loads are considered as demand response resources to participate in the economic dispatch of the microgrid. the income function of the adjustable load user and the operating cost function of the air conditioner are established, and a series of constraints such as user comfort is considered. according to the criterion of equal small increase rate, the output distribution model of the micro gas turbine is established to minimize the cost of the power generation side of the microgrid. finally, a microgrid simulation model is built in matlab/simulink to verify the proposed multi-objective demand response economic dispatch method. from the simulation results, it can be concluded that under the multiobjective economic dispatch method, the benefit of adjustable users has increased by 69.2%, and the operating cost of the airconditioning cluster has been reduced by 18.2% ensuring user comfort. at the same time, the output of the micro gas turbine is distributed through the equal micro-increase rate criterion to minimize the power generation cost of the microgrid, thereby realizing the optimal economic operation of the micro-grid. references ali, s., ullah, k., hafeez, g., khan, i., albogamy, f.r., haider, s.i. (2022). solving day-ahead scheduling problem with multi-objective energy optimization for demand side management in smart grid, engineering science and technology: an international journal, 36, 101135; https://doi.org/10.1016/j.jestch.2022.101135 . attaviriyanupap, p., kita, h., tanaka, e., hasegawa, j. (2002). a hybrid ep and sqp for dynamic economic dispatch with nonsmooth fuel cost function. ieee transactions on power system, 17, 411–6. https://doi.org/10.1109/tpwrs.2002.1007911 basu, m. (2008). dynamic economic emission dispatch using nondominated sorting genetic algorithm-ii. international journal of electric power and energy systems, 30, 140–9. https://doi.org/10.1016/j.ijepes.2007.06.009 battula, a.r., vuddanti, s., salkuti, s.r. (2021). review of energy management system approaches in microgrids. energies, 14, 5459. https://doi.org/10.3390/en14175459 borenstein, s. (2005). the long-run efficiency of real-time electricity pricing. energy journal, 26 (3). https://doi.org/10.5547/issn0195-6574-ej-vol26-no3-5 . boyd, s., parikh, n., chu, e., peleato, b., & eckstein, j. (2011). distributed optimization and statistical learning via the alternating direction method of multipliers. found. trends mach. learn., 3(1), 1–122. https://doi.org/10.3390/en14175459 calderaro, v,. conio, g,. galdi, v., massa, g., & piccolo, a. (2014). active management of renewable energy sources for maximizing power production. international journal of electric power and energy system, 57, 64–72. https://doi.org/10.1016/j.ijepes.2013.11.040 chen, d.j., gong, q.w., & zhang, m.l. (2011). multi-objective optimal dispatch in wind power integrated system incorporating energyenvironmental efficiency. proc. csee, 31(13), 10-17. dashtdar, m., flah, a., hosseinimoghadam, s.m.s., el-fergany a. (2022). frequency control of the islanded microgrid including energy storage using soft computing. sci rep, 12, 20409. https://doi.org/10.1038/s41598-022-24758-6 . david, p.c., jakob, s., panajotis, a., & nedjib, d. (2015). a new thermostat for real-time price demand response: cost, comfort and energy impacts of discrete-time control without deadband. applied energy, 155, 816-825, https://doi.org/10.1016/j.apenergy.2015.06.048 . fig. 5 the frequency and voltage of the microgrid and the outputs of ess https://doi.org/10.1016/j.jestch.2022.101135 https://doi.org/10.1109/tpwrs.2002.1007911 https://doi.org/10.1016/j.ijepes.2007.06.009 https://doi.org/10.3390/en14175459 https://doi.org/10.5547/issn0195-6574-ej-vol26-no3-5 https://doi.org/10.3390/en14175459 https://doi.org/10.1016/j.ijepes.2013.11.040 https://doi.org/10.1038/s41598-022-24758-6 https://doi.org/10.1016/j.apenergy.2015.06.048 m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |758 issn: 2252-4940/©2023. the author(s). published by cbiore dutta g. & mitra k. (2017). a literature review on dynamic pricing of electricity. journal of operational research society, 68 (10), 11311145. https://doi.org/10.1057/s41274-016-0149-4 . han, x.s., gooi, h.b., & kirschen, d.s. (2001). dynamic economic dispatch: feasible and optimal solutions. ieee transactions on power system, 16, 22–8. https://doi.org/10.1109/59.910777 hao, h., huang, b., & ji, p. (2021). optimal configuration of an island microgrid considering demand response strategy. 2021 36th youth academic annual conference of chinese association of automation (yac), 300-304, https://doi.org/10.1109/yac53711.2021.9486661 . hsiao, c.-h., huang, w.-t., chen, l.-r., lin, w.-c., & li, l.-c. (2021). economic dispatch of microgrids using particle swarm optimization and binning method. 2021 ieee international future energy electronics conference (ifeec), 1-5, https://doi.org/10.1109/ifeec53238.2021.9661738 . imtiaz, b., cui, y., & zafar, i. (2021). economic dispatch of microgrid incorporating demand response using dragonfly algorithm. 2021 ieee international conference on advances in electrical engineering and computer applications (aeeca), 59-68; https://doi.org/10.1109/aeeca52519.2021.9574430 . jindal, a., singh, m., & kumar, n. (2018). consumption-aware data analytical demand response scheme for peak load reduction in smart grid. ieee transactions on industrial electronics, 65, 89939004; https://doi.org/10.1109/tie.2018.2813990 . john, h., yu. d.c., & bhattarai, k. (2008). an economic dispatch model incorporating wind power. ieee transactions on energy conversion, 23(2), 603-611. https://doi.org/10.1109/tec.2007.914171 . julia, f., victor, v-l. (2022). dynamic electricity tariffs: designing reasonable pricing schemes for private households. energy economics, 112, 106146, https://doi.org/10.1016/j.eneco.2022.106146 . kumar, d., verma, y.p. & khanna, r. (2019). demand response-based dynamic dispatch of microgrid system in hybrid electricity market. international journal of energy sector management, 13(2), 318-340. https://doi.org/10.1108/ijesm-12-2017-0008 . liu, x., ning, n., wang, g., liu, d., chen, k., & yuan, j. (2021). distributed optimal dispatch method for smart community demand response based on machine learning. 2021 4th international conference on energy, electrical and power engineering (ceepe), 2021, 1040-1046; https://doi.org/10.1109/ceepe51765.2021.9475821 . ma, r., li, k., li, x., & qin, z. (2015). economic and low-carbon dayahead pareto-optimal scheduling for wind farm integrated power systems with demand response. journal of modern power systems and clean energy, 3(3), 393-401; https://doi.org/10.1007/s40565-014-0094-7 . maximilian, j.b. (2022). dynamic pricing of electricity: enabling demand response in domestic households. energy policy, 164, 112878, https://doi.org/10.1016/j.enpol.2022.112878 mhankale, s.e. & thorat, a.r. (2018). droop control strategies of dc microgrid: a review. 2018 international conference on current trends towards converging technologies (icctct), 372-376; https://doi.org/10.1109/icctct.2018.8550854 . murty, v.v.v.s.n. & kumar, a. (2020). optimal energy management and techno-economic analysis in microgrid with hybrid renewable energy sources. journal of modern power systems and clean energy, 8(5), 929-940; https://doi.org/10.35833/mpce.2020.000273 . nguyen, d.h., narikiyo, t., & kawanishi, m. (2018). optimal demand response and real-time pricing by a sequential distributed consensus-based admm approach. ieee transactions on smart grid, 9(5), 4964-4974; https://doi.org/10.1109/tsg.2017.2676179 . pothireddy, k.m.r., vuddanti, s., salkuti, s.r. (2022). impact of demand response on optimal sizing of distributed generation and customer tariff. energies, 15, 190. https://doi.org/10.3390/en1501019 . pourbabak, h., luo, j., chen, t., & su, w. (2018). a novel consensusbased distributed algorithm for economic dispatch based on local estimation of power mismatch. ieee transactions on smart grid, 9(6), 5930-5942. https://doi.org/10.1109/tsg.2017.2699084 rajkumar, r.k., ramachandaramurthy, v.k., & yong, b. (2011). techno-economical optimization of hybrid pv/wind/battery system using neuro-fuzzy. energy, 36(8), 5148-5153. https://doi.org/10.1016/j.energy.2011.06.017 recalde, a.a., manuel, s., & alvarado, a. (2020). design optimization for reliability improvement in microgrids with wind– tidal– photovoltaic generation. electric power systems research, 188, article no. 106540; https://doi.org/10.1016/j.epsr.2020.106540 ross, d.w. & sungkook, k. (1980). dynamic economic dispatch of generation. ieee transactions on power apparatus and system. pas-99(6), 2060–2068. https://doi.org/10.1109/tpas.1980.319847 saeed, m.h., fangzong, w., kalwar, b.a., & iqbal, s. (2021a). a review on microgrids’ challenges & perspectives. ieee access, 9, 166502-166517; https://doi.org/10.1109/access.2021.3135083 . saeed, m.h., fangzong, w., salem, s., khan, y.a., kalwar, b.a., & fars, a. (2021b). two-stage intelligent planning with improved artificial bee colony algorithm for a microgrid by considering the uncertainty of renewable sources. energy reports, 7, 8912-8928; https://doi.org/10.1016/j.egyr.2021.10.123 . saeed, m. h., fangzong, w., & kalwar, b. a. (2022). control of bidirectional dc-dc converter for micro-energy grid’s dc feeders' power flow application. international journal of renewable energy development, 11(2), 533-546: https://doi.org/10.14710/ijred.2022.41952 . saeed, m.h., iqbal, s., & kalwar, b. a. (2022). electricity market management through optimum installation of distributed generation sources and optimum placement based on lmp and isc. advances in engineering and intelligence systems, 001(01). https://doi.org/10.22034/aeis.2022.147534 salkuti, s.r., abhyankar, a. r., bijwe, p. r. (2015). co-optimization of energy and demand-side reserves in day-ahead electricity markets. international journal of emerging electric power systems, 16(2), 195-206. https://doi.org/10.1515/ijeeps-2014-0020 salkuti, s.r., bijwe, p.r. & abhyankar, a.r. (2016). optimal dynamic emergency reserve activation using spinning, hydro and demandside reserves. frontiers in energy. 10, 409–423. https://doi.org/10.1007/s11708-016-0431-9 . salkuti, s.r. (2017). optimizing energy and demand response programs using multi-objective optimization. electrical engineering, 99, 397–406. https://doi.org/10.1007/s00202-016-0438-6 . salkuti, s.r. (2018). emergency reserve activation considering demandside resources and battery storage in a hybrid power system. electrical engineering, 100, 1589–1599. https://doi.org/10.1007/s00202-017-0637-9 salkuti, s.r. (2022). emerging and advanced green energy technologies for sustainable and resilient future grid. energies, 15, 6667. https://doi.org/10.3390/en15186667 . ulbig, a., borsche, t.s., & andersson, g. (2014). impact of low rotational inertia on power system stability and operation. ifac proceedings, 47 (3), 72907297, https://doi.org/10.3182/20140824-6-za-1003.02615. vardakas, j. s., zorba, n., & verikoukis, c. v. (2015). a survey on demand response programs in smart grids: pricing methods and optimization algorithms. ieee commun. surveys tuts., 17(1), 152178. https://doi.org/10.1109/comst.2014.2341586 wang, j., cary, n.b., hu, z., tan, z. (2010). demand response in china. energy, 35(4), 1592-1597. https://doi.org/10.1016/j.energy.2009.06.020 . wang, y., han, s., & liu, s. (2020). distributed consensus-based algorithm for dynamic economic dispatch with wind turbine and energy storage. 2020 7th international conference on information, cybernetics, and computational social systems (iccss), 618-625, https://doi.org/10.1109/iccss52145.2020.9336855 . waseem, m., lin, z., ding, y., wen, f., liu, s., & palu, i. (2021). technologies and practical implementations of air-conditioner based demand response. journal of modern power systems and clean energy, 9(6), 1395-1413, https://doi.org/10.35833/mpce.2019.000449 . wood, a.j., wollenberg, b.f., & sheblé, g.b. (2013). power generation, operation, and control, 3rd ed. new york, ny, usa: wiley. wu, h., liu, x., & ding, m. (2014). dynamic economic dispatch of a microgrid: mathematical models and solution algorithm. https://doi.org/10.1057/s41274-016-0149-4 https://doi.org/10.1109/59.910777 https://doi.org/10.1109/yac53711.2021.9486661 https://doi.org/10.1109/ifeec53238.2021.9661738 https://doi.org/10.1109/aeeca52519.2021.9574430 https://doi.org/10.1109/tie.2018.2813990 https://doi.org/10.1109/tec.2007.914171 https://doi.org/10.1016/j.eneco.2022.106146 https://doi.org/10.1108/ijesm-12-2017-0008 https://doi.org/10.1109/ceepe51765.2021.9475821 https://doi.org/10.1007/s40565-014-0094-7 https://doi.org/10.1016/j.enpol.2022.112878 https://doi.org/10.1109/icctct.2018.8550854 https://doi.org/10.35833/mpce.2020.000273 https://doi.org/10.1109/tsg.2017.2676179 https://doi.org/10.3390/en1501019 https://doi.org/10.1109/tsg.2017.2699084 https://doi.org/10.1016/j.energy.2011.06.017 https://doi.org/10.1016/j.epsr.2020.106540 https://doi.org/10.1109/tpas.1980.319847 https://doi.org/10.1109/access.2021.3135083 https://doi.org/10.1016/j.egyr.2021.10.123 https://doi.org/10.1016/j.egyr.2021.10.123 https://doi.org/10.14710/ijred.2022.41952 https://doi.org/10.22034/aeis.2022.147534 https://doi.org/10.1515/ijeeps-2014-0020 https://doi.org/10.1007/s11708-016-0431-9 https://doi.org/10.1007/s00202-016-0438-6 https://doi.org/10.1007/s00202-017-0637-9 https://doi.org/10.3390/en15186667 https://doi.org/10.3182/20140824-6-za-1003.02615 https://doi.org/10.1109/comst.2014.2341586 https://doi.org/10.1016/j.energy.2009.06.020 https://doi.org/10.1109/iccss52145.2020.9336855 https://doi.org/10.35833/mpce.2019.000449 m.h saeed et al int. j. renew. energy dev 2023, 12(4),749-759 |759 issn: 2252-4940/©2023. the author(s). published by cbiore international journal of electrical power & energy systems, 63, 336346; https://doi.org/10.1016/j.ijepes.2014.06.002 . xu, h., meng, z., & wang, y. (2020). economic dispatching of microgrid considering renewable energy uncertainty and demand sid response. energy reports, 6(9), 2020, 196-204, https://doi,org/10.1016/j.egyr.2020.11.261. zachar, m. & daoutidis, p. (2016). economic dispatch for microgrids with constrained external power exchange. ifac-papersonline, 49(7), 833-838; https://doi.org/10.1016/j.ifacol.2016.07.293 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1016/j.ijepes.2014.06.002 https://doi.org/10.1016/j.egyr.2020.11.261 https://doi.org/10.1016/j.ifacol.2016.07.293 http://creativecommons.org/licenses/by-sa/4.0/ international journal of renewable energy development int. j. renew. energy dev. 2023, 12(4), 767-778 | 767 https://doi.org/10.14710/ijred.2023.54056 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id assessment of the technical-economic performance and optimization of a parabolic trough solar power plant under algerian climatic conditions khaled bouchareba,b* , nabila ihaddadenea,c, khellaf belkhiria,b , khaoula ikhlefd , aissa boudilmia adepartment of mechanical engineering, med boudiaf university, bp 166, m’sila 28000, algeria. blaboratory of materials and mechanics of structure l.m.m.s, university of m'sila, m’sila 28000, algeria. claboratory of renewable energy and sustainable development (lresd), university of mentouri brothers constantine, constantine 25000, algeria. decole nationale polytechnique d’alger (enp), lgmd laboratory, b.p. 182, el-harrach, algiers, algeria. abstract. in this study, the design, analysis and optimization of the performance of a concentrated solar power plant that is based on the parabolic trough technology with a capacity of 100 mw equipped with a thermal energy storage system were conducted, in two representative sites in algeria (tamanrasset and m’sila). the system advisor model software is used to evaluate the technical and economic performances of the two proposed power plants, in addition to carrying out the process of optimizing the initial design of the two power plants by finding the optimal values of the solar multiple and full load hours of the thermal energy storage system, with the aim of increasing the annual energy production and reducing the levelized cost of electricity. the results of the performance analysis conducted on the optimized design showed that the optimum values of the solar multiple and full load hours of the thermal energy storage system for the proposed power plant at the tamanrasset site were found to be 2.4 and 7 h, respectively, with an annual electricity production of 514.6 gwh, and a minimum value of the levelized cost of electricity of 6.3¢/kwh. while the optimum performance of the proposed plant at the m'sila site can be achieved by selecting a solar multiple of 3 and 7 h for thermal energy storage system, with a high annual energy production of 451.84 gwh and a low value of the levelized cost of electricity of 7.8¢/kwh. the results demonstrate that csp plants using parabolic trough technology can increase energy security in the country, while reducing environmental concerns associated with the use of fossil materials. keywords: solar energy, concentrated solar power, parabolic trough power plant, system advisor model (sam). @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 4th may 2023; revised: 18th june 2023; accepted: 29th june 2023; available online: 5th july 2023 1. introduction the needs of mankind for energy are increasing every year, this is due both to the growth of the population, the development of production and technology, and to the increase in energy consumption in everyday life. as human demand for modern energy supply increases, attention to solar energy becomes more intense. consequently, there are active plans to utilize solar energy for different processes to minimize energy demand from conventional energy supply sources (bouguila & said, 2020). the production of electric energy by exploiting renewable energies, especially solar energy, is a challenge of great importance for the coming years (keykhah et al., 2021). in fact, algeria's electric power needs are rising every day. moreover, algeria will need more energy to implement its development plans. today, most of the energy production in algeria comes from fossil sources, the intensive use of these sources leads to the depletion of its reserves and thus the insecurity of energy in the country, because it is not considered a renewable energy source, in addition to the negative effects on the environment * corresponding author email: khaled.bouchareb@univ-msila.dz (k. bouchareb) (greenhouse gas emissions). with algeria's energy demand expected to increase by about 53%, its current reservoirs of conventional energy resources are expected to sufficiently support the country's electricity production for about 50 years (benhadji serradj et al., 2021). in order to remove all these restrictions, the algerian state must turn to renewable energies, especially solar energy, to exploit it in order to meet the increasing demand for energy in the country. algeria is the largest country in africa in terms of area, located in the center of north africa on the mediterranean coast, between latitudes 19° and 38°n and longitudes 8°w and 12°e, with an area of 2,381,741 km², and a transitional climate, from maritime in the northern regions to semi-arid and arid in the central and southern regions (benhadji serradj et al., 2021; keykhah et al., 2021). as it is located within the sun belt region, algeria has great potential for solar energy. it has one of the highest solar energy deposits in the world (abbas et al., 2013; t. e. boukelia et al., 2015b). the northern region sees approximately 2650 hours of the insolation time annually, while in the southern region it reaches about 3500 hours (stambouli et al., 2012). research article https://doi.org/10.14710/ijred.2023.54056 https://doi.org/10.14710/ijred.2023.54056 mailto:khaled.bouchareb@univ-msila.dz https://orcid.org/0000-0003-3230-7202 https://orcid.org/0000-0003-1962-4481 https://orcid.org/0000-0002-6326-6717 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.54056&domain=pdf k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |768 issn: 2252-4940/© 2023. the author(s). published by cbiore the energy transition is algeria's path to a secure, environmentally friendly and economically prosperous future. the central element of this transition is the restructuring of our energy supply towards the use of renewable energies. this means that renewable energy will become our primary source of electricity. in recent years, there has been an increase in the level of use in the electricity sector. in response to the high demand for energy and the negative impacts on the environment, researchers around the world are working to find more sustainable alternative energy technologies. among the power generation technologies that have been developed, concentrated solar power (csp) systems are a direct alternative to power plants based on fossil fuels. csp plants can contribute 6% of the world's electric power demand by 2030 and 12% by 2050 (islam et al., 2018). csp appears to be the method of choice for large capacity, utility-scale electric generation in the near term. this technology has the possibility of energy storage and auxiliary heat production during sunlight unavailability. in csp systems, sunlight is concentrated using mirrors to create heat, then the heat is used to create steam, which is used to drive turbines and generators, just like in a conventional power station. since solar energy is not very dense, it is necessary to concentrate the solar radiation to obtain high temperatures that can be exploited to produce electricity (islam et al., 2018; stambouli et al., 2012). according to the concentrating geometry, csp systems can be classified into point and linear concentrators; solar tower and dish solar systems use point concentrators, while parabolic trough and linear fresnel collectors use linear concentrators (t. e. boukelia et al., 2015a; el gharbi et al., 2011). the csp plants have economic justification only for locations where direct normal irradiation (dni) values are greater than 5.5 kwh/m2/day or (2000 kwh/m2/year) (hirbodi et al., 2020). the parabolic trough technology power plant is one of the best proven csp systems for its maturity and applicability in arid and semi-arid regions (reddy & kumar, 2012). csp-based technology is suitable for high dni areas (praveen et al., 2018). with sunlight concentrated approximately 70-100 times by parabolic trough mirror technology, the operating temperature achieved is in the range of 350-550°c (ummadisingu & soni, 2011). the thermal energy collected at the solar field level is transported by a heat transfer fluid (htf) that circulates through the solar receivers and returns to a series of heat exchangers in the power block where superheated high-pressure steam is generated. the power block actually used in solar power plants is the steam cycle which uses a steam turbine generator to produce electrical energy (lovegrove & csiro, 2012). most designs of commercial parabolic trough technology csp plants contain a solar field and a power block, and in order to maintain a constant electrical energy production both thermal energy storage (tes) and fuel backup systems can be used (t. e. boukelia et al., 2015a). the majority of csp plants using parabolic trough technology are equipped with tes system to ensure constant energy production and to extend the plant's operating time during times of low or absent solar radiation (bouguila & said, 2020). reddy et al. (reddy & kumar, 2012) conducted a technical and economic study of a 5 mw csp plant with parabolic trough technology at 58 sites in india. the results showed that the annual electricity production in the studied sites ranged between 11 and 18 mw, and the levelized cost of electricity (lcoe) in jodhpur site amounted to 11.00 and 11.84 indian rupees/kwh for the plant that uses oil and water as htf, respectively. kalogirou (kalogirou, 2013) analyzed the technical characteristics, the cost of electricity produced and land area required, for three types of csp technology (parabolic trough, solar tower and solar dish) in cyprus. the results indicate that the csp plant with a parabolic trough and a tes system with a capacity of 4 h is the best option for installation in cyprus, since it has a high annual efficiency and does not require a large land area. a study by guzman et al. (guzman et al., 2014) where the performance of a parabolic trough csp plant with tes system for the city of barranquilla (colombia) is simulated for find the ideal plant design optimization and the key design parameters. the results showed that the studied plant could contribute 50% of the city's electrical consumption, and through the optimization results it was found that the solar multiple (sm) is 2 and 6 hours for the capacity of the tes system. bhuiyan et al. (bhuiyan et al., 2020) carried out a study to optimize key design parameters of a parabolic trough csp plant, in addition to evaluating the optimum design performance of the plant at eight different sites in bangladesh. the results showed that the power plant that uses molten salt as a htf offers better performance compared to the thermal oil plant. tahir et al. (tahir et al., 2021) evaluated the technical and economic feasibility of a csp plant with a parabolic trough collector at six sites in pakistan, and carried out an optimization study to obtain the optimal design of the proposed plants that reduces the lcoe. the results indicated that pishin site provided the lowest value for lcoe compared to other sites, and in terms of the availability of suitable infrastructure, it is noticed that quetta site is the ideal site for the construction of these plants. mohammadi et al. (mohammadi et al., 2021) conducted a technical, economic and environmental analysis of the performance of a solar power plant with a parabolic trough technology for thermal energy production in salt lake city (usa). the results revealed that the annual production of the plant amounted to 15,389.24 mwth, at a levelized cost of heat estimated at 26.3 $/mwth, and the results showed the optimization also has a significant impact of the values of the sm, the investment tax credit, and the total cost of the plant on the levelized cost of heat. bashir et al. (bashir & özbey, 2022) conducted a design study for a hypothetical csp plant with a parabolic trough collector with a capacity of 80 mw in sudan, and in order to determine the appropriate sites for the installation of such plants, they analyzed the thermal performance and economic feasibility of the plant studied in 15 sites in sudan. the results concluded that the city of wadi halfa, located in the northern region of sudan, is one of the suitable sites for the establishment of csp plants, given that it has the highest rates of dni, in addition to its good topographical characteristics and favorable climatic conditions. the annual electrical production of the proposed plant at the wadi halfa site was 281.145 gwh with an overall efficiency and capacity factor (cf) of 15% and 40.1%, respectively. through the economic analysis of the plant, the lcoe was 0.155$/kwh. focusing on algeria, benhadji serradj et al. (benhadji serradj et al., 2021) carried out a design and analysis of the technical and economic performance of a power plant using parabolic trough technology in the city of tamanrasset (southern algeria). they found that the plant could provide about 78% and 60% of the city's electrical consumption during winter and summer respectively, and that the lcoe was about 0.062$/kwh with a payback period of 8.78 years. benabdellah et al. (benabdellah & ghenaiet, 2021) conducted a technoeconomic analysis of the integrated solar combined cycle k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |769 issn: 2252-4940/© 2023. the author(s). published by cbiore (iscc) power plant that uses parabolic trough technology and is currently operating in the hassi r'mel region (southern algeria). the studied plant is equipped with a new tes system. the obtained results show significant improvements in both the overall performance of the studied plant and the efficiency of converting solar energy into electrical energy. the results of the economic evaluation of the studied plant showed that the lcoe was about 9.75 ¢/kwh. in addition, the integration of the tes system into the power plant helps better stability of the grid, and the modified power plant can save about 30 million$ in natural gas consumption. debbache et al. (debbache et al., 2018) conducted an investigation study to find out the effect of some parameters of the design of the parabolic trough collector (aperture width and focal distance) on the energy produced for the csp plant that depends on the parabolic trough technology, proposed in the city of touggourt (southern algeria). the results of the study show that the electricity production increases with the increase in aperture width with the smallest focal distance. in addition, it was found that the best design for a parabolic trough collector is an aperture width of 5 m and a focal distance of 0.5 m which leads to an annual production of 30 mwh. achour et al. (achour et al., 2018) examined the performance of a power plant based on iscc technology in southern algeria by developing a thermodynamic model to evaluate both the overall performance of the hybrid solar power plant and the intensity of solar radiation. the results showed that the efficiency of converting solar energy into electricity during sunny hours reaches 14.4%. in addition, the flow rate of the htf and the solar incidence angle on the collector surface are among the factors that affect the amount of electricity generated. from the above literature review, it is clear that the majority of studies related to the design, performance evaluation and optimization of csp plants with a parabolic trough collector are conducted at sites in asia, india, bangladesh, and north and south america. however, the most of the available research works on the deployment of csp plants with parabolic trough technology in algeria is generally limited to a preliminary evaluation of the advantages of their installation and a study of their economic feasibility. numerous researches related to the design, performance analysis and optimization of the parabolic trough csp plants in algeria, have been performed. however, to the best knowledge of us, the most of these studies were carried out on grounds located in the southern region of the country. the question arises whether these studies can be used to simulate the parabolic trough csp plants in northern algeria such as the m'sila site which has an important potential solar energy as shown by (kherbiche et al., 2021). research in this aspect is very important due to the urgent need to find more sustainable alternative energy technologies such as exploiting renewable energy sources to meet the increasing demand for electricity in algeria and reducing dependence on traditional energy resources and the resulting negative effects on the environment. for this reason, this study is being conducted to design, analyze and optimize the performance of a 100 mw csp plant based on parabolic trough technology with a tes system at two representative sites in algeria. an important aspect of this analysis is the comparison of the results of two representative sites in algeria (tamanrasset and m'sila). 2. methodology the design and analysis of the technical and economic performance of the csp plant based on the proposed parabolic trough technology is carried out at the two selected sites using system advisor model (sam), a software used to design and evaluate the technical and economic potential of solar power plants, and to assist in the decision-making of those involved in the renewable energy industry (achour et al., 2018). it was developed by the national renewable energy laboratory (nrel). the methodology of this study consists of the following steps: (i) collecting meteorological data for selected locations, (ii) design of a 100 mw csp plant with tes, (iii) evaluation of the performance of the preliminary design of the proposed solar power plant in two representative sites in algeria (tamanrasset in the far south and m’sila in the northern region), (iv) study the environmental impacts: water consumption, carbon dioxide (co2) emissions and natural gas preservation, and (v) optimization of the parabolic trough power plant with tes. the main parameters of the optimization process are full load hours of the tes and the sm. 2.1 site selection and resource assessment to evaluate the performance of the proposed csp plant, the sam software needs the meteorological data for the two selected sites, which were obtained by creating a typical meteorological year 3 (tmy3) weather file data format from the metonorm7 software database. the average daily dni in algerian territory ranges between 4.66 kwh/m2 in the northern regions and 7.26 kwh/m2 for the southern areas, and this corresponds to 1700 kwh/ m2/year and 2650 kwh/ m2/year for the northern and southern regions, respectively (taqiy eddine boukelia & mecibah, 2013; kherbiche et al., 2021). in this study, two representative sites in algeria providing average annual dni greater than 5.5 kwh/m2/day were selected to analyze and optimize the performance of the proposed csp plant. the two selected sites are tamanrasset in the far south, and m’sila for the northern region. the characteristics of the two selected sites, tamanrasset and m’sila, are presented in table 1. in figure 1 are shown the monthly changes of average dni and ambient temperature for the tamanrasset and m’sila sites. comparatively, the tamanrasset site is characterized by a high irradiation level of more than 7 kwh/m2/day throughout the year except for the months of september and december. during march, the dni reached its maximum value of 9.019 kwh/m2/day, while december recorded its lowest value at 6.703 kwh/m2/day. it can also be observed that the maximum and minimum values of the average ambient temperature were respectively in the months of june and january, on the other hand the average dni at the m’sila site has a maximum during table 1 characteristics of the selected locations analyzed in this study. location latitude and longitude elevation (m) daily average dni(kwh/m2/day) daily average temperature(°c) tamanrasset 22.78° n, 5.51° e 718 7.70 22.8 m’sila 35.70° n, 4.54° e 476 6.25 21.6 k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |770 issn: 2252-4940/© 2023. the author(s). published by cbiore july of 7.934 kwh/m2/day and a minimum of 4.753 kwh/m2/day during december. in addition, the average maximum and minimum temperature values were recorded in july and january, respectively. 2.2 parabolic trough solar thermal power plant among the csp technologies available, the parabolic trough technology is today the most widespread, the most successful and the most developed for the production of electricity (taqiy eddine boukelia & mecibah, 2013). figure 2 is a schematic diagram of parabolic trough solar power plants with tes. it can be seen that these power plants consist of three main parts, including the solar field, the tes system, and the power block (belgasim & elmnefi, 2014). the solar collectors are arranged in a series configuration known as loops and oriented in a northsouth direction to follow the sun from east to west. tes can be used with solar power plants to ensure the continuity of electricity production. normally, the tes capacity is in the order of several hours during which it is filled with htf during the day and emptied after sunset so that electricity is still produced even after sunset. 2.3 characteristics of the proposed csp plant design the csp plant subject to this study consists of 898160 m2 of solar field reflector based on the one of ls3 model (luz solar collector, third generation). these collectors are equipped with a schott ptr70 2008 type vacuum receiver tube. the ls3 solar collectors are oriented in the north-south direction and its direction axis is parallel to the horizontal plane. the htf used in the solar field is therminol vp-1, and molten salt as the storage fluid, these two traditional htf fluids are often used in csp-based power generation systems (bouguila & said, 2020). the solar multiple (sm) is defined as the ratio between thermal power obtained by the solar field at design point and thermal power required by the power block at nominal conditions, and it can be expressed as (marugán-cruz et al., 2019) : int th sf design po th pb e sm e = (1) where: eth sf is the thermal energy obtained by the solar field and eth pb is the thermal energy required by the power block at nominal conditions. the proposed csp plant has a tes system in the form of two circular tanks containing molten salt which consists of 60% sodium nitrate (nano3) and 40% potassium nitrate (kno3) (purohit & purohit, 2017). tes system allows the supply of thermal energy to the power block when solar radiation is low or absent during the day or night (ghodbane et al., 2021). the full load hours of tes for a csp plant specifies the number of hours thermal storage can supply energy to the power block to operate at the designed input level (t. e. boukelia et al., 2015b), and is given by the expression: des tes tes des cycle p h h  = (2) where: htes is the thermal energy storage system capacity; pdes is the design cycle thermal requirement; htes the total number of fig. 1 average dni per month and ambient temperature for the two selected sites (tamanrasset and m'sila). fig. 2 schematic diagram of a parabolic trough csp plant with tes system. k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |771 issn: 2252-4940/© 2023. the author(s). published by cbiore desired storage hours; and des cycle  the design point cycle efficiency. for the initial analysis, the value of the sm is set as 2 and the full load hours for tes are taken to be 6 h. the initial design of the 100 mw parabolic trough solar thermal power plant with tes system was carried out in order to evaluate its performance in the cities of tamanrasset and m’sila. optimal values of sm and full load hours for tes will be obtained through the initial design optimization process. table 2 summarizes the initial design parameters of the proposed csp plant. 2.3.1 energy analysis the total incident solar energy received by the solar field opening is given as: cos inc q a dni=   (3) where a is the collector’s aperture area and the angle of incidence. the total utile energy delivered by the solar field is presented as: ( ) field f sfo sfi q m h h=  − (4) where: mf is the mass flow rate of the htf, hsfo enthalpy at the outlet of solar field, and hsfi enthalpy at the inlet of the solar field. therefore, the energy efficiency of the solar field can be found as: field sf inc q q  = (5) the power block energy efficiency is expressed as: net pb inp w q  = (6) table 2 characteristics of the proposed parabolic trough power plant (cáceres et al., 2016; guzman et al., 2014; hirbodi et al., 2020). characteristics value solar field total field reflector area solar multiple field htf fluid number of loops single loop aperture field htf min operating temperature field htf max operating temperature design loop inlet temperature design loop outlet temperature water usage per wash number of washes per year 898160 m2 2 therminol vp-1 206 4360 m2 12°c 400 °c 293 °c 391 °c 0.7 l/m2 63 collectors collectors type reflective aperture area aperture width, total structure length of collector assembly number of modules per assembly length of single module luz ls-3 545 m2 5.75 m 100 m 12 8.33 m receivers receiver type absorber tube inner diameter absorber tube outer diameter glass envelope inner diameter glass envelope outer diameter schott ptr 70 2008 0.066 m 0.07 m 0.115 m 0.12 m power cycle design gross output estimated gross to net conversion factor estimated net output at design (nameplate) rated cycle conversion efficiency design inlet temperature design outlet temperature condenser type 111 mwe 0.9 100 mwe 0.356 391 °c 293 °c evaporative thermal storage storage type full load of tes storage volume tes thermal capacity parallel tank pairs storage htf min operating temperature storage htf max operating temperature storage htf fluid two tank 6 h 25304.4m3 1870.79mwht 1 238 °c 593 °c hitec solar salt k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |772 issn: 2252-4940/© 2023. the author(s). published by cbiore where: wnet is the net power generation and qinp is the total thermal energy received by the power block. the plant’s final energy efficiency is calculated as follows: net overall inc w q  = (7) the net capacity factor (cf) of a designed csp plant with a capacity of 100 mw is expressed as: 24 net w cf h nd plant power capacité day =        (8) where: nd is the number of days in a year. 2.3.2 economic analysis the lcoe is one of the most important indicators used in evaluating the economic performance of csp plants (azouzoute et al., 2020); it is calculated by dividing the accumulated construction and operating costs of a solar power plant by the total annual energy produced during the operating life of the plant, as given in the following equation (cáceres et al., 2016): ( ) ( ) 0 1 1 1 1 n n nn n n nn c i d lcoe q d = = + + = +   (9) where: i0 is the initial investment expenditures, cn is the annual total costs for the year n, qn is the electricity produced for the year n, n is the economic life of the power plant, and d is the discount rate. the assumptions and economic data used in the simulations on the sam software for the parabolic trough csp plant are presented in table 3. 3. results and discussions the design, analysis and optimization of the performance of a csp plant that is based on the parabolic trough technology described in the preceding sections yielded results that are now presented and discussed. it will be seen that csp plants using parabolic trough technology are one of the most promising technologies in the field of electric power generation in algeria. 3.1 performance analysis of the csp plant design figure 3 shows the hourly data of the thermal energy incident on the solar field and the thermal energy produced from the solar field, the input of thermal energy for the power block, the thermal energy stored in the tes system, and the net electrical output of the proposed power plant at the tamanrasset site. the net electrical output depends on the incident irradiation and the thermal energy input to the power cycle. the value of the maximum thermal energy incident on the solar field was found to be about 840.53 mw in march, while the maximum value of the thermal energy entered into the power cycle was recorded at about 311.79 mw in march, due to the availability of solar resources for the selected site in this period of the year. during the period from february to october, the solar resources are high, the tes system tank is charged during the day with thermal energy in excess of the power block needs, and at the time of low solar radiation or after sunset, the tes system provides thermal energy to the power block to continue to produce electrical energy. table 3 the main financial input parameters used in the economic modeling of the proposed csp plant (benhadji serradj et al., 2021; enjavi-arsanjani et al., 2015; ikhlef & larbi, 2020; zhang et al., 2013). financial data value analysis period loan term loan rate inflation rate real discount rate nominal discount rate assessed percent insurance rate sales tax state income tax rate 30 years 20 years 8% years 4.6%/year in 2018 4%/year in 2018 8.78%/year 80% of installed cost 0.3% of installed cost 5% of installed cost 15%/year direct costs site improvements solar field cost htf system cost storage cost power block balance of plant contingency 15 $/m2 150 $/m2 60 $/m2 65 $/kwht 1150 $/kwht 120 $/kwht 10% of direct costs indirect costs engineering, procurement and construction other costs 13% of direct costs 3.5% of direct costs operation and maintenance costs fixed cost by capacity variable cost by generation 70 $/kw-year 3 $/mwh k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |773 issn: 2252-4940/© 2023. the author(s). published by cbiore figure 4 presents the hourly data for thermal energy incident on the solar field to the net electrical output of the proposed csp plant at the m'sila site. the peak value of the solar thermal incident was found around 705.08 mw in july, which is lower than the maximum recorded at the tamanrasset site. this is due to the noticeable difference in the values of dni between the two sites. the maximum thermal energy input to the power block was recorded as 319.29 mw in august. in addition, it should be noted that the tes system during the summer months contributes significantly to extending the period of electrical energy production after sunset, due to the availability of solar resources during this period, while the amount of energy stored in the winter months is very low, which fig. 3 field incident solar thermal power, thermal power input for a power cycle and electrical output (tamanrasset). fig. 4 field incident solar thermal power, thermal power input for a power cycle and electrical output (m’sila). k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |774 issn: 2252-4940/© 2023. the author(s). published by cbiore does not allow the system storage extends the duration of electrical energy production until after sunset. figure 5 shows the average monthly electric power generation obtained from the proposed csp plant in both tamanrasset and m’sila sites. the monthly energy generated from the parabolic trough power plant in tamanrasset peaks during the month of march and may, reaching a 48.06 gwh and 45.12 gwh, respectively, the lowest value of the monthly energy generated for the tamanrasset site was found during the month of december, reaching a value of 26.41 gwh. for the proposed power plant in the m’sila site, it is seen that the highest monthly production of electric energy was found during the months of june and july, with a value of 42.23 gwh and 45.99 gwh, respectively, and the minimum values of monthly generated energy were found during january and december, which are 13.65 gwh and 9.22 gwh respectively. it is clear that the monthly variation of the net energy production, for each site, almost follows the monthly variation of the dni. the annual production of energy generated from the two csp plants was 454.51 gwh, 329.66 gwh for tamanrasset and m’sila site, respectively. it can be seen that the proposed power plant in tamanrasset provided the highest annual generation of electric energy compared to the proposed plant in m’sila, and the reason is due to the marked difference in the values of dni, which is shown in figure 1. the waterfall diagram in figure 6 indicates the annual flow of energy from incident solar irradiation on the solar field to net electrical output. it shows the annual energy performance of each component of the proposed power plant at the two selected sites. as for the power plant in the city of tamanrasset, the total annual solar energy incident on the solar collector amounted to 2552.34 gwh/year, while the amount of thermal energy produced by the solar field amounted to 1362.74 gwh/year. however, the heat energy transferred to the power block is 1314.88 gwh/year. the thermal energy produced from the solar field of the proposed plant at the m’sila site amounted to 983.7 gwh/year, and the power block received an amount of 974.14 gwh/year. it can be concluded that the main energy losses that occur during energy transfer between the various components of the proposed plant are at the solar field and power block level. the losses in the solar field are found mainly due to the thermal losses in the receivers, while the losses in the power block are the mechanical losses, and the electrical losses necessary to operate the auxiliary equipment. the annual amount of electrical energy generated from the proposed csp plant in tamanrasset was found to be 454.51 gwh, while m’sila recorded 329.66 gwh, with a cf of 51.9% and 37.7% for tamanrasset and m’sila, respectively. in addition, the efficiency value of solar-to-electrical energy conversion was 18% and 16.08% for the two selected sites, respectively. it is clear that the proposed power plant in the city of tamanrasset offers high performance compared to the power plant in m’sila, because the levels of solar irradiation in tamanrasset are higher than m’sila. table 4 summarizes the annual performance comparison of the two proposed power plants in algeria. 3.2 cost analysis the economic study of the two proposed plants is based on a lcoe calculation, the lcoe value ranged between 6.46 ¢$/kw h in tamanrasset, 8.82 ¢$/kw h in m’sila. according to the international renewable energy agency (irena) 2020 report, 150 mw of new csp plants were commissioned globally in 2020 (irea, 2020). the values of fig. 5 monthly energy generation for the proposed parabolic trough power plants in algeria. fig. 6 annual energy flow of the proposed parabolic trough power plants. table 4 comparison of annual energy yields for proposed parabolic trough plants in algeria. parameter tamanrasset m’sila dni (kwh/m2/year) annual power generation (gwh) capacity factor (%) mean efficiency of the solar field(𝜼𝑺𝑭) mean efficiency of the power block(𝜼𝑷𝑩) mean efficiency of the plant(𝜼𝒐𝒗𝒆𝒓𝒂𝒍𝒍) 2810.5 454.51 51.90 53.96 34.56 18.00 2281.25 329.66 37.70 47.98 33.84 16.08 k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |775 issn: 2252-4940/© 2023. the author(s). published by cbiore lcoes obtained for the two power plants in this study are lower than the irena global weighted average of 0.108 $/kwh in 2020 for csp plants. the plant located in tamanrasset provides the highest annual energy production with the lowest lcoe, while the m’sila plant offers an average annual energy production with a relatively high lcoe. 3.3 environmental impacts analysis considering the energy transition toward clean energy, csp technologies are more beneficial to the environment because it emits very few damaging pollutants and reduces fossil fuel consumption (praveen et al., 2018). therefore, it is necessary to study the environmental impacts of csp plants in order to estimate their potential benefits. among these effects are natural gas preservation, carbon dioxide (co2) emission, and water consumption. the software sam is used to calculate the amount of water consumed by these plants. in general, the water use of csp plants is divided into three main parts, the washing system, the steam generation, and the cooling system. from the obtained results, it is clear that there is a difference in the amount of water consumed by the proposed plants, which is estimated at 1,576,475 m3 and 1,215,756 m3 in tamanrasset and m’sila, respectively. the proposed csp plant at the tamanrasset site consumes more water annually than the proposed power plant in m’sila. reliance on the exploitation of traditional energy to produce electrical energy leads to energy insecurity, in addition to the fact that the exploitation of these resources results in an increase in the amount of co2 emitted into the atmosphere, which leads to global warming and climate change (solomon et al., 2009). to produce 1 kwh of electricity, 0.285 m3 of natural gas are required (hassabelgabo abdelrazig ibrahim & mohammed elmardi suleiman khayal, 2020). and 1 kwh of electricity generation produces 0.35 m3 (0.66 kg) of co2 emissions (brander et al., 2011). the calculations that we have made suggest that the quantities of natural gas and co2 emissions that we avoid through our use of csp plants to generate electricity are very significant quantities compared to conventional electricity generation systems. the results of the environmental impact analysis of the proposed csp plants, including the amount of natural gas preserved, reduced co2 emissions, and annual water consumption are summarized in table 5. 3.4 optimization of the initial parabolic trough plant design the smaller solar field of a csp plant with parabolic trough technology reduces the thermal energy supplied to the power block, thus reducing the amount of electrical energy produced. the presence of a large solar field means an increase in the thermal energy produced, which is greater than the needs of the power block, and with insufficient storage capacity to store the excess thermal energy, there will be thermal energy loss and an increased investment cost for the csp plant. thus, optimization analysis is essential for the whole system. the optimization procedure helps to determine the lowest value of lcoe with a higher amount of annual electrical energy produced (awan et al., 2020). optimization is about finding the combination of the two inputs that minimizes the lcoe while maximizing annual power generation. the variation of two main design parameters, namely the sm and the full load hours of tes, is used to optimize the proposed design. the size of the solar field has a direct impact on annual electricity production and lcoe. an increase in the sm value leads to a corresponding increase in the solar field aperture and, thus, an increase in the thermal energy produced by the solar field. as a result, more electricity is generated, thus lowering the lcoe value. the variations of the annual energy production and lcoe with the sm of the proposed csp plant in both tamanrasset and m’sila sites are shown in figures 7(a) and 7(b), respectively. as shown in figure 7(a) the lcoe decreases with the increase of the sm that reaches 2.4 for the proposed plant in tamanrasset, table 5 annual quantities of preserved natural gas, carbon dioxide mitigation and annual water consumption of the proposed csp plants (m3). parameters tamanrasset m’sila natural gas preservation (m3) co2 mitigation (m3) water usage (m3) 129,536,887 159,080,387 1,576,475 93,953,884.3 115,381,963 1,215,756 fig. 7 the variation of annual energy generation and lcoe with sm of the proposed csp plant in: (a) tamanrasset and (b) m’sila. k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |776 issn: 2252-4940/© 2023. the author(s). published by cbiore beyond this value there is a significant increase in the value of the lcoe with a slight increase in the amount of annual electricity produced. the minimum lcoe was recorded to be around 6.33 ȼ$ /kwh with a sm of 2.4. the lcoe value rises with higher sm values, and this is due to the fact that the capital cost of the plant increases with the increase in the size of the solar field. the annual power output of the plant clearly increases with the increase in sm, but this increase becomes negligible for higher sm values. the optimal values of the sm for the proposed csp plant in m’sila was 3 with a lcoe of 7.84 ȼ$ /kwh, in addition to an increase in the annual production of electric energy, as shown in figure 7(b). the full load hours of the tes system in the csp plant are the second design parameter that has been studied for optimization. when solar radiation is low or after sunset the tes system delivers more thermal energy to the power block, thus allowing the plant to generate electricity for longer time intervals. the variations of the annual energy production and lcoe with the full load hours of the tes for the csp plant proposed at tamanrasset and m’sila sites are presented in figures 8(a) and 8(b), respectively. as shown in figure 8(a), the lcoe decreases with the increase of the full load hours of the tes, reaching a value of 6.3 ȼ$ /kwh for the tamanrasset site, the lcoe increases after another increase in the full load hours of the tes system. the lowest lcoe values correspond to 7 h of tes. the optimum full load hours obtained for the proposed power plant at the m’sila site was 7 h with the lowest lcoe value of 7.8 ȼ$ /kwh, and with an increase in power production as shown in figure 8(b). if the full load hours of tes are increased beyond 7 h for both the proposed csp plants in tamanrasset and m’sila, the lcoe value increases with a slight increase in annual energy production because the solar fields of the two power plants are not able to generate enough surplus thermal energy for a large tes system. as for the increase in the lcoe value, the reason is due to the high investment cost of the tes system for both plants. the optimized design results obtained for the two proposed csp plants are summarized in table 6. as can be seen from the data in table 6, it is evident that there is a significant improvement in the amount of annual electric energy produced, cf and lcoe values for the optimized configuration compared to that of the initial design of the two studied plants. fig. 8 the variations of annual energy generation and lcoe with full load hours of tes of the proposed csp plant in (a) tamanrasset and (b) m’sila. table 6 performance comparison of the optimized and initial design of a proposed csp plant for two sites in algeria. parameters tamanrasset m’sila initial optimized initial optimized annual energy output (gwh) capacity factor (%) solar multiple full load hours of tes lcoe (ȼ$ /kwh) 454.51 51.9 2 6 6.46 514.6 58.8 2.4 7 6.3 329.66 37.67 2 6 8.82 451.84 51.63 3 7 7.8 table 7 performance comparison of the proposed csp plant with other similar literature. author annual energy generation (gw h) mean annual efficiency (%) lcoe (¢$/kwh) (benhadji serradj et al., 2021) (abbas et al., 2013) (awan et al., 2019) current study 390.7 223 ‒ 415 ranges from 355.18to 397.48 451.84‒ 514.6 15.3 13.8 ‒ 16.4 16.73 ‒ 17.93 16 – 18 6.2 11.93 ‒ 29.58 10.5 ‒ 11 6.3 ‒ 7.8 k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |777 issn: 2252-4940/© 2023. the author(s). published by cbiore 3.5 comparison of output performance with similar studies table 7 presents the results of comparing the performance of the csp plant with the proposed parabolic trough technology in two locations in algeria with the results reported in some similar studies on this technology in different locations around the world. according to this table, it is found that the average plant efficiency and lcoe values are in close agreement with the results obtained in the reviewed literature with a significant increase in the value of the annual energy generated. 4. conclusion the present study was conducted to design, analyze and optimize the performance of a concentrated solar power plant based on parabolic trough technology with a thermal energy storage system at two representative sites in algeria with a wet cooling system, the system advisor model software was used to evaluate the performance of the two proposed power plants. the initial analysis of the proposed design showed that the annual production of the proposed concentrated solar power plant in tamanrasset and m’sila amounted to 454.51 gwh and 329.66 gwh respectively. it is noteworthy that the annual energy production has a significant relationship with the dni values and the climatic conditions of the site. evaluation of the economic performance of the proposed concentrated solar power plants showed that the power plant in the tamanrasset site has the lowest lcoe at 6.46 ȼ$/kwh, as for the proposed plant in m’sila. it is observed that the lcoe is less than 8.82 ȼ$/kwh. since the two proposed power plants at the two selected sites have the same investment costs estimated at 627,808,128 $ per site, it is important to note that the lcoe values are inversely proportional to the annual production of each plant. the environmental impact analysis revealed that the proposed csp plant in the tamanrasset site consumes more annual water than the proposed power plant in m'sila. it can be seen that the more arid and desert climate of the site, the greater the annual water consumption of the plant. in terms of the amount of annual carbon dioxide emissions that can be avoided through the use of csp plants to generate electricity instead of conventional electricity generation systems, the environmental analysis showed that the solar power plant in the tamanrasset site is able to avoid the amount of carbon dioxide emissions estimated at 159,080,387 m3, while the proposed plant at the m'sila site has the lowest annual amount of co2 emissions estimated at 115,381,963 m3. the comparison of the production of electricity through csp plants and conventional electricity generation systems that depend on fossil resources, shows that a very large amount of natural gas is preserved when using csp plants. the optimization procedure is based on considering the lowest value of lcoe with the largest annual energy production. the proposed design optimization process is done by modifying two main design parameters, namely the sm and full load hours of tes. the optimization study showed that there is a significant increase in the annual energy production, reaching 514.6 gwh and 451.84 gwh for the two sites, tamanrasset and m’sila, respectively, with a clear decrease in the values of lcoe. this type of csp plant, which is based on parabolic trough technology, has shown good results in terms of power generation and the price of electricity production in algeria. these results may be encouraging for the algerian government to exploit its large potential of solar energy to generate electricity on a large scale and reduce the use of fossil materials, therefore, the spread of csp plants in algeria is a major step for the renewable energy sector in the country, not only to add new energy, but also to increase energy security and address growing environmental problems due to the use of fossil fuels. conflicts of interest: no potential conflict of interest was reported by the authors. references abbas, m., belgroun, z., aburidah, h., & merzouk, n. k. (2013). assessment of a solar parabolic trough power plant for electricity generation under mediterranean and arid climate conditions in algeria. energy procedia, 42(december), 93–102. https://doi.org/10.1016/j.egypro.2013.11.009 achour, l., bouharkat, m., & behar, o. (2018). performance assessment of an integrated solar combined cycle in the southern of algeria. energy reports, 4(march), 207–217. https://doi.org/10.1016/j.egyr.2017.09.003 awan, a. b., zubair, m., & chandra mouli, k. v. v. (2020). design, optimization and performance comparison of solar tower and photovoltaic power plants. energy, 199, 117450. https://doi.org/10.1016/j.energy.2020.117450 awan, a. b., zubair, m., praveen, r. p., & bhatti, a. r. (2019). design and comparative analysis of photovoltaic and parabolic trough based csp plants. solar energy, 183(march), 551–565. https://doi.org/10.1016/j.solener.2019.03.037 azouzoute, a., alami, a., & touili, s. (2020). overview of the integration of csp as an alternative energy source in the mena region. energy strategy reviews, 29(december 2019), 100493. https://doi.org/10.1016/j.esr.2020.100493 bashir, a. a. a., & özbey, m. (2022). modelling and analysis of an 80mw parabolic trough concentrated solar power plant in sudan. clean energy, 6(3), 512–527. https://doi.org/10.1093/ce/zkac032 belgasim, b., & elmnefi, m. (2014). evaluation of a solar parabolic trough power plant under climate conditions in libya. 13th international conference on sustainable energy technologies (set2014), august, 1–7. https://doi.org/10.13140/2.1.3788.4485 benabdellah, h. m., & ghenaiet, a. (2021). energy, exergy, and economic analysis of an integrated solar combined cycle power plant. engineering reports, 3(11), 1–25. https://doi.org/10.1002/eng2.12404 benhadji serradj, d. e., sebitosi, a. b., & fadlallah, s. o. (2021). design and performance analysis of a parabolic trough power plant under the climatological conditions of tamanrasset, algeria. international journal of environmental science and technology, may. https://doi.org/10.1007/s13762-021-03350-x bhuiyan, n., ullah, w., islam, r., ahmed, t., & mohammad, n. (2020). performance optimisation of parabolic trough solar thermal power plants–a case study in bangladesh. international journal of sustainable energy, 39(2), 113–131. https://doi.org/10.1080/14786451.2019.1649263 bouguila, a., & said, r. (2020). optimization of a small scale concentrated solar power plant using rankine cycle. journal of thermal engineering, 6(3), 268–281. https://doi.org/10.18186/thermal.711287 boukelia, t. e., mecibah, m. s., kumar, b. n., & reddy, k. s. (2015a). investigation of solar parabolic trough power plants with and without integrated tes (thermal energy storage) and fbs (fuel backup system) using thermic oil and solar salt. energy, 88, 292– 303. https://doi.org/10.1016/j.energy.2015.05.038 boukelia, t. e., mecibah, m. s., kumar, b. n., & reddy, k. s. (2015b). optimization, selection and feasibility study of solar parabolic trough power plants for algerian conditions. energy conversion and management, 101, 450–459. https://doi.org/10.1016/j.enconman.2015.05.067 boukelia, t. e, & mecibah, m. s. (2013). parabolic trough solar thermal power plant: potential, and projects development in algeria. https://doi.org/10.1016/j.egypro.2013.11.009 https://doi.org/10.1016/j.egyr.2017.09.003 https://doi.org/10.1016/j.energy.2020.117450 https://doi.org/10.1016/j.solener.2019.03.037 https://doi.org/10.1016/j.esr.2020.100493 https://doi.org/10.1093/ce/zkac032 https://doi.org/10.13140/2.1.3788.4485 https://doi.org/10.1002/eng2.12404 https://doi.org/10.1007/s13762-021-03350-x https://doi.org/10.1080/14786451.2019.1649263 https://doi.org/10.18186/thermal.711287 https://doi.org/10.1016/j.energy.2015.05.038 https://doi.org/10.1016/j.enconman.2015.05.067 k. bouchareb et al int. j. renew. energy dev 2023, 12(4), 767-778 |778 issn: 2252-4940/© 2023. the author(s). published by cbiore renewable and sustainable energy reviews, 21, 288–297. https://doi.org/10.1016/j.rser.2012.11.074 brander, a. m., sood, a., wylie, c., haughton, a., lovell, j., reviewers, i., & davis, g. (2011). electricity-specific emission factors for grid electricity. ecometrica, august, 1–22. cáceres, g., montané, m., nasirov, s., & o’ryan, r. (2016). review of thermal materials for csp plants and lcoe evaluation for performance improvement using chilean strategic minerals: lithium salts and copper foams. sustainability (switzerland), 8(2). https://doi.org/10.3390/su8020106 debbache, m., karoua, h., laissaoui, m., hazmoune, m., takilalte, a., bouhallassa, a., lecheheb, s., bouaichaoui, s., hamidat, a., & imessad, k. (2018). design parameters effect on annual energy production of proposed design of parabolic trough solar plant. proceedings of 2018 6th international renewable and sustainable energy conference, irsec 2018, 1–6. https://doi.org/10.1109/irsec.2018.8702844 el gharbi, n., derbal, h., bouaichaoui, s., & said, n. (2011). a comparative study between parabolic trough collector and linear fresnel reflector technologies. energy procedia, 6, 565–572. https://doi.org/10.1016/j.egypro.2011.05.065 enjavi-arsanjani, m., hirbodi, k., & yaghoubi, m. (2015). solar energy potential and performance assessment of csp plants in different areas of iran. energy procedia, 69, 2039–2048. https://doi.org/10.1016/j.egypro.2015.03.216 ghodbane, m., boumeddane, b., hussein, a. k., li, d., & sivasankaran, s. (2021). optical numerical investigation of a solar power plant of parabolic trough collectors. journal of thermal engineering, 7(3), 550–569. https://doi.org/10.18186/thermal.888167 guzman, l., henao, a., & vasqueza, r. (2014). simulation and optimization of a parabolic trough solar power plant in the city of barranquilla by using system advisor model (sam). energy procedia, 57, 497–506. https://doi.org/10.1016/j.egypro.2014.10.203 hassabelgabo abdelrazig ibrahim, d. s., & mohammed elmardi suleiman khayal, d. o. (2020). solar energy potential and assessment of csp plant accounting for sustainability in sudan. international journal of engineering applied sciences and technology, 5(7), 12–19. https://doi.org/10.33564/ijeast.2020.v05i07.003 hirbodi, k., enjavi-arsanjani, m., & yaghoubi, m. (2020). technoeconomic assessment and environmental impact of concentrating solar power plants in iran. renewable and sustainable energy reviews, 120(december 2019), 109642. https://doi.org/10.1016/j.rser.2019.109642 ikhlef, k., & larbi, s. (2020). techno-economic optimization for implantation of parabolic trough power plant: case study of algeria. journal of renewable and sustainable energy, 12(6). https://doi.org/10.1063/5.0013699 irea. (2020). renewable power generation costs in 2020. in international renewable energy agency. https://www.irena.org//media/files/irena/agency/publication/2018/jan/irena_ 2017_power_costs_2018.pdf islam, m. t., huda, n., abdullah, a. b., & saidur, r. (2018). a comprehensive review of state-of-the-art concentrating solar power (csp) technologies: current status and research trends. renewable and sustainable energy reviews, 91(november 2017), 987–1018. https://doi.org/10.1016/j.rser.2018.04.097 kalogirou, s. a. (2013). solar thermoelectric power generation in cyprus: selection of the best system. renewable energy, 49, 278– 281. https://doi.org/10.1016/j.renene.2012.01.014 keykhah, s., assareh, e., moltames, r., taghipour, a., & barati, h. (2021). thermoeconomic analysis and multi-objective optimization of an integrated solar system for hydrogen production using particle swarm optimization algorithm. journal of thermal engineering, 7(4), 746–760. https://doi.org/10.18186/thermal.915413 kherbiche, y., ihaddadene, n., ihaddadene, r., hadji, f., mohamed, j., & beghidja, a. h. (2021). solar energy potential evaluation. case of study: m’sila, an algerian province. international journal of sustainable development and planning, 16(8), 1501–1508. https://doi.org/10.18280/ijsdp.160811 lovegrove, k., & csiro, w. s. (2012). introduction to concentrating solar power (csp) technology. concentrating solar power technology, 3–15. https://doi.org/10.1533/9780857096173.1.3 marugán-cruz, c., serrano, d., gómez-hernández, j., & sánchezdelgado, s. (2019). solar multiple optimization of a dsg linear fresnel power plant. energy conversion and management, 184(january), 571–580. https://doi.org/10.1016/j.enconman.2019.01.054 mohammadi, k., khanmohammadi, s., immonen, j., & powell, k. (2021). techno-economic analysis and environmental benefits of solar industrial process heating based on parabolic trough collectors. sustainable energy technologies and assessments, 47(june), 101412. https://doi.org/10.1016/j.seta.2021.101412 praveen, r. p., baseer, m. a., awan, a. b., & zubair, m. (2018). performance analysis and optimization of a parabolic trough solar power plant in the middle east region. energies, 11(4), 1–18. https://doi.org/10.3390/en11040741 purohit, i., & purohit, p. (2017). technical and economic potential of concentrating solar thermal power generation in india. renewable and sustainable energy reviews, 78(april), 648–667. https://doi.org/10.1016/j.rser.2017.04.059 reddy, k. s., & kumar, k. r. (2012). solar collector field design and viability analysis of stand-alone parabolic trough power plants for indian conditions. energy for sustainable development, 16(4), 456– 470. https://doi.org/10.1016/j.esd.2012.09.003 solomon, s., plattner, g. k., knutti, r., & friedlingstein, p. (2009). irreversible climate change due to carbon dioxide emissions. proceedings of the national academy of sciences of the united states of america, 106(6), 1704–1709. https://doi.org/10.1073/pnas.0812721106 stambouli, a. b., khiat, z., flazi, s., & kitamura, y. (2012). a review on the renewable energy development in algeria: current perspective, energy scenario and sustainability issues. renewable and sustainable energy reviews, 16(7), 4445–4460. https://doi.org/10.1016/j.rser.2012.04.031 tahir, s., ahmad, m., abd-ur-rehman, h. m., & shakir, s. (2021). techno-economic assessment of concentrated solar thermal power generation and potential barriers in its deployment in pakistan. journal of cleaner production, 293. https://doi.org/10.1016/j.jclepro.2021.126125 ummadisingu, a., & soni, m. s. (2011). concentrating solar power technology, potential and policy in india. renewable and sustainable energy reviews, 15(9), 5169–5175. https://doi.org/10.1016/j.rser.2011.07.040 zhang, h. l., baeyens, j., degrève, j., & cacères, g. (2013). concentrated solar power plants: review and design methodology. renewable and sustainable energy reviews, 22, 466– 481. https://doi.org/10.1016/j.rser.2013.01.032 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1016/j.rser.2012.11.074 https://doi.org/10.3390/su8020106 https://doi.org/10.1109/irsec.2018.8702844 https://doi.org/10.1016/j.egypro.2011.05.065 https://doi.org/10.1016/j.egypro.2015.03.216 https://doi.org/10.18186/thermal.888167 https://doi.org/10.1016/j.egypro.2014.10.203 https://doi.org/10.33564/ijeast.2020.v05i07.003 https://doi.org/10.1016/j.rser.2019.109642 https://doi.org/10.1063/5.0013699 https://www.irena.org/-/media/files/irena/agency/publication/2018/jan/irena_2017_power_costs_2018.pdf https://www.irena.org/-/media/files/irena/agency/publication/2018/jan/irena_2017_power_costs_2018.pdf https://www.irena.org/-/media/files/irena/agency/publication/2018/jan/irena_2017_power_costs_2018.pdf https://doi.org/10.1016/j.rser.2018.04.097 https://doi.org/10.1016/j.renene.2012.01.014 https://doi.org/10.18186/thermal.915413 https://doi.org/10.18280/ijsdp.160811 https://doi.org/10.1533/9780857096173.1.3 https://doi.org/10.1016/j.enconman.2019.01.054 https://doi.org/10.1016/j.seta.2021.101412 https://doi.org/10.3390/en11040741 https://doi.org/10.1016/j.rser.2017.04.059 https://doi.org/10.1016/j.esd.2012.09.003 https://doi.org/10.1073/pnas.0812721106 https://doi.org/10.1016/j.rser.2012.04.031 https://doi.org/10.1016/j.jclepro.2021.126125 https://doi.org/10.1016/j.rser.2011.07.040 https://doi.org/10.1016/j.rser.2013.01.032 http://creativecommons.org/licenses/by-sa/4.0/ international journal of renewable energy development int. j. renew. energy dev. 2023, 12 (3), 448-458 | 448 https://doi.org/10.14710/ijred.2023.49759 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id design, optimization and economic viability of an industrial low temperature hot water production system in algeria: a case study karim kacia* , mustapha merzouka , nachida kasbadji merzoukb , mohammed missoumc , mohammed el ganaouid , omar behare , rabah djedjigd alaboratoire de physique fondamental et appliqué département des energies renouvelables, faculté de technologie, université blida, w. blida, algeria. bunité de développement des equipements solaires, udes, centre de développement des energies renouvelables, cder, 42004, w. tipaza, algérie. cdepartment of mechanical engineering, faculty of technology, university center of morceli, abdellah, tipaza, algeria. duniversity of lorraine, lermab, iut de longwy, 186 rue de lorriane, 54400 cosnes-et-romain, france. eccrc, king abdullah university of science and technology (kaust), thuwal, 23955, saudi arabia. abstract. solar energy has a great potential in many areas of industrial activity in algeria. this is because most of algeria has high levels of sustainable solar insulation. unfortunately, few industries use solar energy for hot water generation, but some industrial processes require hot water at temperatures that can be easily obtained from solar thermal panels. this paper presents a case study to investigate the technical and financial feasibility of a solar-powered industrial agro-processing system in algiers. based on the solar collectors connection type for which the economic feasibility study was carried out, an appropriate design of the system was determined. the latter was actually done by analyzing the levelized cost of energy savings. the design of the thermo-solar process is carried out based on f-chart method with a new approach by integrating the incidence angle modifier and of using real and experimental data requirements to determine realistic achievable performance of the solar process. the results showed that, in comparison to the currently used electrical system, the electrical energy savings achieved by the solar-powered system make it an economically viable option with a solar coverage rate of 80%. the investment depreciation balance shows that the use of such a thermal solar energy system will be more competitive than fossil fuels system if the price of electricity in the country increases from 0.048 to 0.075 €/kwh. keywords: process industrial application, solar thermal collector, thermo-solar system, techno-economic assessment, water heating. @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 12th nov 2022; revised: 24th january 2023; accepted: 6th march 2023; available online: 21st march 2023 1. introduction in recent years, interest in solar development has increased in algeria since the publication of the renewable energy and energy efficiency plan 2016-2030 (the minister of energy transition and renewable energies., 2022). the industrial sector is a promising area for the development of solar thermal technology, accounting for more than 25% of total final energy consumption (international energy agency, 2020). the efficiency of converting solar energy into heat (up to 70%) is much higher than converting it into electricity (about 15%), demonstrating the benefits of integrating solar thermal energy into industrial processes. the heating process consumes a lot of thermal energy. globally, 50% of energy consumption is used for heating applications (valderrama et al., 2022). a great fraction (about 90%) of this heat comes from fossil fuels, which emit large amounts of greenhouse gas emissions and thus exacerbate the effects of climate change (intergovernmental panel on climate change ipcc (2022) mitigation of climate change; assessment report on climate change). statistics show that 60% of industrial processes use heat at temperatures below 400°c, while more than 30% operate at temperatures * corresponding author email: kkaci2022@gmail.com (k. kaci) below 100°c (zühlsdorfa et al., 2019). therefore, integrating solar energy into industry can help to reduce the effects of climate change, especially in low-temperature processes. nowadays, around 456 gwth of solar thermal output has been installed globally (national renewable energy laboratory. 2021). low temperature thermal industrial processes recorded growth of 1.5% in 2016 (meyers et al 2018). according to a recent study, published by solirco, the industrial sector used 416 414 m2of installed space, 40% of which was used in the agro-food sector (the international renewable energy agency. 2015) in fact, the heat generated by solar thermal systems is less used in industry than in domestic applications (renewable energy policy network for the 21st century. 2016). however, several studies related to the use of solar heat for industrial processes revealed that solar thermal industrial heating systems can achieve higher efficiencies compared to domestic applications, especially at lower temperatures (farjana et al., 2017 and sharma., et al 2017). a study conducted by schweiger et al. (2000) in the framework of the european project poship, which focuses on evaluating the potential of industrial solar systems in spain and portugal, shows that the latter can supply research article https://doi.org/10.14710/ijred.2023.49759 https://doi.org/10.14710/ijred.2023.49759 mailto:kkaci2022@gmail.com https://orcid.org/0000-0002-8787-2025 https://orcid.org/0000-0003-0362-5674 https://orcid.org/0000-0001-5900-6675 https://orcid.org/0000-0003-4910-6052 https://orcid.org/0000-0003-4910-6052 https://orcid.org/0000-0002-7058-5935 https://orcid.org/0000-0002-8898-6737 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.49759&domain=pdf k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 449 issn: 2252-4940/© 2023. the author(s). published by cbiore 5804 gwh, or 3.6% of the total energy demand (schweigeretal.,2000). muller et al. (2004) investigated solar heating systems as part of the european project promise, and found that austria has an energy solar potential of 264 pj, representing 30% of total energy consumption. about 32 % is used to generate thermal energy below 200 °c. kalogirou et al. (2003) studied the potential of solar industrial process heating in cyprus for different temperature levels. similar studies have also been reported for australia by fuller et al., (2011). kummert et al. (2000) carried out an experimental study on installation of a thermo-solar process, considering typical daily profiles. several studies have proven that the application of solar thermal systems in industry can lead to significant energy savings. as a part of an iea project shc task 33 solar peces task iv, several configurations for thermo-solar processes thermal systems have been carried out in several european countries (weiss., 2005). lima et al., (2015) demonstrated the feasibility of integrating a thermal solar process in a hospital in brazil. surech el et al., (2017) presented a techno-economic study on integrated thermal solar processes in the indian textile industry. anubhav et al., (2016) developed a model of a thermal solar process for the automotive industry in 2016. the economic analysis shows a payback period of 18 months. quijera et al., (2011) studied the feasibility of integrating a solar thermal system into the dairy industry through mathematical modeling. therefore, the work suggests that the solar thermal potential of the industrial processes studied is important and should be considered for future energy. moreover, akssas et al, (2013) conducted a technoeconomic analysis of a solar water heater for a hospital center in batna in algeria, to study the technoeconomic feasibility of solar heating water integration. the results showed the possibility of significant energy savings with installation (total annual provided energy = 1427,1mwh and a total annual net reduction of ghg = 905,84 tons of co2 (pahlavana 2018). review of previously published studies and various applications demonstrate the great potential of integrating solar heat into industrial processes. to encourage the deployment of solar energy, real case studies must be investigated to provide accurate information about the performance and the costs of solar-powered industrial processes. in this direction, the objective of this work is to design, evaluate the energy performance and determine the costs of a thermo-solar process that would be integrated to a food industrial process in algeria. the design of the thermo-solar process is carried out based on f-chart method with a new approach by integrating the incidence angle modifier and of using real and experimental data requirements to determine realistic achievable performance of the solar process. this methodology was implemented into trnsys software to facilitate modeling and device management of the entire system. in the following, firstly the case study is presented, then the methodology used to design the different solar system components is detailed and finally the energy and economic performances of the whole thermo-solar system are evaluated. 2. methodology 2.1. presentation of the case study the thermo-solar process system considered in this study is an agro-food industrial process producing flavors and perfumes located in algiers, algeria (longitude: 2.95°e, latitude: 36.7°n and elevation: 350 m). fig. 1. thermo-solar process into the industrial food process the system consists of primary and secondary circuits as shown in figure 1. the primary circuit includes a solar collector's field and a heat exchanger. water is heated up in flat plat collectors and it is used as a working fluid to power a secondary circuit through water/water heat exchanger. the secondary circuit feeds the industrial process and it is equipped with a storage tank which serves to store hot water to be used when solar energy is not enough available. another heat exchanger is integrated in storage tank to provide high modularity to the system. a back-up electric heater is used in addition to the solar heating to ensure the continuous supply of hot water to the industrial process. for starting or stopping the pump, a regulation system was used. the thermo-solar process is designed to produce 3 m3/day of hot water at 60°c. the hourly daily load profile imposed by the used process is illustrated in figure 2. the 3 m3 of hot water demand is situated between 11 to 12 am and 14 to 15 pm, which coincide with the availability of solar radiation. 2.2. weather data and validation validation of the weather data model involves comparing predicted and measured solar radiation and ambient temperature. the measurements are provided by the algiers meteorological station (centre for the development of renewable energy of algeria., 2021), while the forecasts are obtained from the trnsys software (meteonormsoftware8., 2022). figure (3a) and (3b) highlight the measured solar radiation intensity and ambient temperature at the site of algiers in 2020. a good agreement between predicted and measured results was observed. the model demonstrated high accuracy in predicting the site's solar radiation and ambient temperature, allowing the system's performance to be estimated with reasonable accuracy. 6 7 8 9 10 11 12 13 14 15 16 17 18 0 500 1000 1500 2000 2500 3000 w at er lo ad (l ite r) times (hours) fig. 2 hourly daily load profile imposed by the process k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 450 issn: 2252-4940/© 2023. the author(s). published by cbiore janu febr marc apri may june july augu sept octo nove dece 0 50 100 150 200 250 m on th ly s ol ar ir ra di at io n (k w h/ m 2) months results meteonorm results measured a) janu febr marc apri may june july augu sept octo nove dece 0 5 10 15 20 25 30 a m bi en t t em pe ra tu re (° c ) months results meteonorm results measured b) fig.3 the radiometric parameters of the site a) solar radiation data b) ambient temperature it can also be noted that the location has a high solar radiation intensity, reaching 250 kwh/m2 in july, one of the highest potentials in the world, with an average ambient temperature of 18ºc. the temperature of cold water is an interesting parameter to determine the performance of the solar system. the monthly average ambient air and network water temperature provided by the software are shown in figure 4 for a period of one year. the ambient temperature varies between 10 °c and 25 °c and average annual cold water temperature is about 18 °c. these values are the same of those provided by the algerian water company (ministry of water resources and water security, algerian waters, 2021). this means that the data provided by the software is accurate. the difference between summer and winter is significant. it can be seen that the temperature of cold water is 15°c in winter and reaches 25°c in summer, a difference of 10°c between these two periods, which affects strongly the system performance (software meteonorm 8. 2022). table 1 conditions of parameters during testing measurement parameters test conditions angle of incidence (θ) (°) -55° to +55° total radiation on tilted surface ig (w/m2) 840 <ig< 950 diffuse radiation id(w/m2) 130 < id< 220 ambient temperature ta(°c) 20°c < ta< 25°c mass flow rate, �̇�𝑚 (kg/s.m2) 0.02 +/10% wind velocity (v)-(m/s) 1 < v < 3 source: kaci et al (2012) jan feb mar apr may jun jul aug sep oct nov dec 8 10 12 14 16 18 20 22 24 26 t em pe ra tu re (° c ) months ambient temperature the mains water temperature fig.4 variation of the mains water and ambient temperatures during a year 2.3. determination of the solar collector’s parameters before evaluating the energy performance of the solar waterheating system under investigation, an experimenal study was carried out to determine the real characteristics of the solar collector used; namely the instantaneous efficiency and the modified incidence angle. according to the european standard, the fluctuations of the variables (incidence angle θ, ambient temperature ta, global radiation ig, mass flow rate �̇�𝑚 and wind speed v) are checked within the limit tolerated by the standard as given in table 1. the experimental study was carried out during different moments of four days, where inlet and outlet collector’s temperatures, ambient temperature as well as the global radiation are measured. these parameters allow calculating the instantaneous efficiency of the solar collector η. four inlet temperatures are considered (ta, ta+10, ta+20, ta+30). each temperature is recorded for four times, two before noon and two in the afternoon, for a total of sixteen points. table 2 shows the parameters measured during the test period. the instantaneous efficiency may be given by the equation (1) (kaci et al2012.) 𝜂𝜂 = 𝜂𝜂0 − 𝑎𝑎1𝑇𝑇 ∗ (1) where: 𝑇𝑇∗ = (𝑇𝑇𝑒𝑒 − 𝑇𝑇𝑎𝑎) 𝐼𝐼𝑔𝑔 𝜂𝜂0 = f𝑅𝑅(τα) is the collector optical efficiency; and a1=( frug) is the collector overall energy loss coefficient. 0,00 0,05 0,10 0,15 0,20 0,25 0,30 0,35 0,0 0,1 0,2 0,3 0,4 0,5 ef fic ie nc y (-) t* fig. 5 representation of the instantaneous performance of the collector k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 451 issn: 2252-4940/© 2023. the author(s). published by cbiore table 2 tests carried out to calculate the instantaneous yield time te ts ta ig t* η (hour) (°c) (°c) (°c) (w/m2) (km2/w) 11:29 a.m 21.45 25.04 18.61 815 0.056 0.36 11:44 a.m 21.14 27.18 22.25 941 0.02 0.53 12:54 p.m 20.46 25.72 20.28 950 0.029 0.46 01:09 pm 20.82 26.31 20.01 954 0.037 0.48 11:09 a.m 31.86 36.58 23.20 839 0.131 0.46 11:24 a.m 31.72 36.39 23.79 850 0.12 0.45 13:44 p.m 32.16 36.69 22.34 935 0.129 0.40 13:59 p.m 31.67 35.77 24.62 897 0.101 0.38 11:41 a.m 39.08 43.38 23.24 940 0.191 0.38 11:56 a.m 39.75 43.92 22.59 951 0.202 0.36 13:14 p.m 40.22 43.82 20.26 915 0.237 0.32 13:29 p.m 40.67 44.46 21.72 889 0.234 0.35 12:01 p.m 49.15 52.82 22.42 878 0.325 0.34 12:16 p.m 49.32 52.65 21.50 906 0.325 0.30 13:11 p.m 49.89 52.52 22.01 918 0.318 0.23 13:26 p.m 49.15 51.65 21.68 918 0.312 0.22 source: kaci et al (2012) from equation (1) and obtained measured values (table 2), the characteristic curve of the instantaneous efficiency related to the inlet temperature can be plotted. this curve will satisfy the equation 2: 𝜂𝜂 = 0.49 – 5.7 t* (2) the value 0.49 represents the optical efficiency of the collector 𝜂𝜂0. it is the intersection of the curve with the y-axis. the value 5.7 represents the overall thermal losses of the collector a1 which is the slope of the line. the incidence angle modifier is a correction factor for the effective transmissivity-absorptivity product 𝐾𝐾𝜏𝜏𝜏𝜏is defined as the ratio of the instantaneous efficiency at any incidence to the efficiency at normal incidence. it allows you tocalculate the instantaneous return at any incidence of the day (kaci et al 2012). 𝐾𝐾𝜏𝜏𝜏𝜏 = 𝜂𝜂(𝜃𝜃) 𝜂𝜂(𝜃𝜃=0) (3) with (𝜏𝜏𝜏𝜏)𝜃𝜃: optical efficiency of the collector at normal incidence; (𝜏𝜏𝜏𝜏)𝑛𝑛: optical efficiency of the collector at 𝜃𝜃 incidence. the incidence angle modifier tests were carried out by measuring the previous parameters (te, ts, ta, ig) at different angle of incidence θ (°). six points are recorded: two readings in the morning at an incidence of -40° < θ < -50°; two readings around noon tsv (θ < 15°) and two in the afternoon (40°<θ<50°). table 3 shows the experimental values for the of the modified angle of incidence of the solar collector studied. the adjustment of the measurement points of the modified angle of incidence kτα as a function of the angle of incidence θ is presented in parabolic form, as shown in figure 6. table 3 experimental values noted for the calculation of k𝜏𝜏𝜏𝜏 (kaci et al, 2012) points θ (°) t*(km2/w) η [(1/cosθ)-1] kα 01 -55 0.028 0.391 0.6689 0.836 02 -50 0.034 0.343 0.555 0.734 03 0 0.076 0.468 0 1 04 15 0.049 0.452 0.025 0.966 05 50 0.110 0.320 0.657 0.685 06 55 0.063 0.327 0.750 0.699 0 10 20 30 40 50 60 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 k τα angle of incidence θ (°) fig. 6 angle of incidence modified according to the angle of incidence θ table 4 main parameters used in the simulation parameter value unit weather data algiers (algeria) (-) collector aperture area 2.6 m2 collector type flat plate (-) optical efficiency 0.49 (-) heat loss coefficients a1 5.7 w/m2k the incidence angle modifier𝐾𝐾𝜏𝜏𝜏𝜏 0.87 (-) collector loop-specific mass flow 0.02 kg/s m2 tilt angle 36 ° storage 3000 liters process temperature 60 °c source : experimental tests (kaci et al 2012); laboref company report (2020) the results of the evolution kτα as a function of θ give a second-order curve in parabolic shape. it can be noticedthat kτα is more or less constant in the range of angle of incidence lower than 40°. with prismed solar glass, the instantaneous efficiency of the flat solar collector can be constant up to an angle of incidence of 60°. it reaches a maximum for a value equal to 1. the values of kτα are all less than one 1, which is true for the flat solar collector, because the latter reaches a maximum efficiency for a normal incidence the minimum value is 0.74, for this the average value is of 0.87 will be taken throughout our study. the latter represents a correction coefficient that must be multiplied by the expression of the performance of the capture field. table 4 summarizes the technical characteristics of the experienced solar collector and a technical report obtained from the company (laboref., 2020), (laboref: agro-food industry for the production of aromas). 2.4. system design the objective this part is to determine the collector field area and its optimal configuration (series-parallel arrangements). the f-chart method, which is the most used method in the sizing of solar heating systems, is used to determine the solar collector field size. the f-chart method was developed by klein and beckman, using results obtained on a large number of simulations with trnsys. according to this model, the useful energy gained by a solar collector can be given by (bangura et al 2022): 𝑄𝑄𝑢𝑢 = 𝐴𝐴𝐶𝐶�𝐹𝐹𝑅𝑅(𝜏𝜏𝜏𝜏)𝑒𝑒𝐼𝐼𝑔𝑔 − 𝐹𝐹𝑅𝑅𝑈𝑈𝑔𝑔(𝑇𝑇𝑒𝑒 − 𝑇𝑇𝑎𝑎)� (4) k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 452 issn: 2252-4940/© 2023. the author(s). published by cbiore in our work, this model was improved by introducing the measured performance correction factor 𝐾𝐾𝜏𝜏𝜏𝜏 at any incidence for flat plate solar collectors. this correction is among the originalities of this work (bangura et al 2022): 𝑄𝑄𝑢𝑢 = 𝐴𝐴𝐶𝐶�𝐾𝐾𝜏𝜏𝜏𝜏𝐹𝐹𝑅𝑅(𝜏𝜏𝜏𝜏)𝑒𝑒𝐼𝐼𝑔𝑔 − 𝐹𝐹𝑅𝑅𝑈𝑈𝑔𝑔(𝑇𝑇𝑒𝑒 − 𝑇𝑇𝑎𝑎)� (5) using the average values of collector’s parameters obtained in the previous section (𝐾𝐾𝜏𝜏𝜏𝜏 =0.87, 𝐹𝐹𝑅𝑅(𝜏𝜏𝜏𝜏)𝑒𝑒= 0.49 and 𝐹𝐹𝑅𝑅𝑈𝑈𝑔𝑔=6.53), equation (5) become: 𝑄𝑄𝑢𝑢 = 𝐴𝐴𝐶𝐶�0.87. 0.49 𝐼𝐼𝑔𝑔 − 6.53(𝑇𝑇𝑒𝑒 − 𝑇𝑇𝑎𝑎)� (6) this method provides an useful mean for estimating the fraction of total heating load that will be supplied by solar energy for a given solar heating system. the primary designvariable is collector area and secondary variables are collector type. then, the solar fraction is defined by the equation 7 (zwalnana et al 2021): 𝑓𝑓 = 𝑄𝑄𝑢𝑢 𝐿𝐿 𝑜𝑜𝑜𝑜𝑓𝑓 = 1 − 𝐴𝐴𝑢𝑢𝐴𝐴 𝐿𝐿 (7) with l : monthly total heating load for heating hot water; 𝐴𝐴𝐴𝐴𝐴𝐴 : monthly total heating auxiliary energy hot water; by replacing the expression of the useful energy in equation (7): 𝑓𝑓 = 1 𝐿𝐿 ∫ 𝐴𝐴𝐶𝐶𝐹𝐹𝑅𝑅 �(𝜏𝜏𝜏𝜏)𝑒𝑒 𝐼𝐼𝑔𝑔 − 𝑈𝑈𝑔𝑔(𝑇𝑇𝑒𝑒 − 𝑇𝑇𝑎𝑎)�𝛥𝛥𝛥𝛥 𝑑𝑑𝑑𝑑 (8) where : 𝑍𝑍 = 𝑇𝑇𝑒𝑒−𝑇𝑇𝑎𝑎 𝑇𝑇𝑟𝑟𝑒𝑒𝑟𝑟−𝑇𝑇𝑎𝑎 therefore, the equation (8) becomes: 𝑓𝑓 = 𝐴𝐴𝐶𝐶𝐹𝐹𝑅𝑅 𝐿𝐿 ∫  �(𝜏𝜏𝜏𝜏)𝑒𝑒 𝐼𝐼𝑔𝑔 − 𝑈𝑈𝑔𝑔�𝑇𝑇𝑟𝑟𝑒𝑒𝑟𝑟 − 𝑇𝑇𝑎𝑎�𝑍𝑍�𝛥𝛥𝛥𝛥 𝑑𝑑𝑑𝑑 (9) the method is a correlation of many hundreds simulation results of thermal solar system performances. the conditions of the simulations were varied over appropriate ranges of parameters for practical system designs. the resulting correlations give f, the fraction of monthly heating loads (hot water) supplied by solar energy as a function of two dimensionless parameters. the first is related to the ratio of collector losses and the second is related to the ratio of absorbed solar radiation. this equation can be decomposed into 2 dimensionless terms x (collector loss) and y (collector gain), which are given in equations (10) and (11)(klein et al 2013 and okafor et al 2012): 𝑋𝑋 = 𝐴𝐴𝐶𝐶𝐹𝐹𝑅𝑅 𝐿𝐿 𝑈𝑈𝑔𝑔�𝑇𝑇𝑟𝑟𝑒𝑒𝑟𝑟 − 𝑇𝑇𝑎𝑎�𝛥𝛥𝑑𝑑 (10) 𝑌𝑌 = 𝐴𝐴𝐶𝐶𝐹𝐹𝑅𝑅 𝐿𝐿 (𝜏𝜏𝜏𝜏) 𝐻𝐻𝛽𝛽𝑁𝑁 (11) for the system configuration studied, the solar fraction f of the monthly total load supplied by water heating system, given as a function of x and y in equations 10 and 11, becomes in the form (deepika et al 2016): 𝑓𝑓 = 1.029𝑌𝑌 − 0.065𝑋𝑋 − 0.245𝑌𝑌2 + 0.0018𝑋𝑋2 + 0.0215𝑌𝑌3 (12) the iterative dichotomy method is used to solve equations (10), (11) and (12) which allow to obtain the collector’s area for any incidence angle modifier. concerning the optimal configuration of solar system, the latter is designed according to the water temperature of the storage tank. the thermal collectors of the solar field may be connected in series, in parallel, or in combinations. once the collector parameters, the storage size and loss coefficient, the magnitude of the load, and the meteorological data are specified, the storage tank temperature can be calculated as a function of time. also, the energy gained from the collector, losses from storage, and energy to load can be determined for any desired period of time by integration of the appropriate rate quantities with the following equation (kaci et al 2014). (𝑀𝑀𝑀𝑀𝑀𝑀)𝑠𝑠 𝑑𝑑𝑇𝑇𝑠𝑠 𝑑𝑑𝛥𝛥 = 𝐴𝐴𝐶𝐶𝐹𝐹𝑅𝑅 ��(𝐾𝐾𝜏𝜏𝜏𝜏)𝑒𝑒𝐼𝐼𝑔𝑔 − 𝑈𝑈𝑔𝑔(𝑇𝑇𝑖𝑖 − 𝑇𝑇𝑎𝑎)�� − (𝑈𝑈𝐴𝐴)𝑆𝑆(𝑇𝑇𝑆𝑆 − 𝑇𝑇𝑎𝑎) − 𝜀𝜀2��̇�𝑚𝐶𝐶𝑝𝑝�𝑚𝑚𝑖𝑖𝑛𝑛(𝑇𝑇𝑆𝑆 − 𝑇𝑇𝑟𝑟) (13) equation (13) can be integrated using simple euler method, where temperature derivative 𝑑𝑑𝑇𝑇𝑠𝑠 𝑑𝑑𝛥𝛥 is expressed as 𝑇𝑇𝑠𝑠+−𝑇𝑇𝑠𝑠 ∆𝛥𝛥 . the time step should be taken equal to weather data increments in order to ensure numerical stability. the storage temperature ts can be given by (kaci et al 2014): 𝑇𝑇𝑆𝑆 + = 𝑇𝑇𝑆𝑆 + 𝑑𝑑𝛥𝛥 (�̇�𝑚𝑐𝑐𝑝𝑝)𝑠𝑠 �𝜀𝜀𝐴𝐴𝐶𝐶𝐹𝐹𝑅𝑅�𝐾𝐾(𝜏𝜏𝜏𝜏)𝐼𝐼𝑔𝑔 − 𝐹𝐹𝑅𝑅𝑈𝑈𝑔𝑔(𝑇𝑇𝑒𝑒 − 𝑇𝑇𝑎𝑎)� − (𝑈𝑈𝐴𝐴)𝑆𝑆(𝑇𝑇𝑆𝑆 − 𝑇𝑇𝑎𝑎) − 𝜀𝜀(�̇�𝑚𝑀𝑀𝑀𝑀)𝑠𝑠(𝑇𝑇𝑒𝑒 − 𝑇𝑇𝑟𝑟)� (14) 𝑇𝑇𝑆𝑆 +: the new tank temperature calculated for the end of the hour. 2.5. economical parameters the levelized cost of energy (lcoe) is the most indicator to evaluate the competitiveness of the proposed solar thermal process. indeed, lcoe represents the average revenue per unit of energy generated that would be required to recover the costs of operating the thermo-solar process during an assumed financial life cycle. according to the international energy agency (iea), the costs of the thermo-solar process can be classified into three different sets: investment costs, operation and maintenance costs and electricity costs. the factor multiplying the initial investment is called capital recovery factor (crf). it converts a present value into a stream of equal annual payments over a specified time (in this case 25 years), at a specified discount rate (in this case 8%)(fernández 2013) . 𝐿𝐿𝐶𝐶𝐿𝐿𝐿𝐿 = 𝐶𝐶𝑟𝑟𝑟𝑟.𝐾𝐾𝑖𝑖𝑖𝑖𝑖𝑖+𝐾𝐾𝑂𝑂,𝑀𝑀+ 𝐾𝐾𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑟𝑟𝑖𝑖𝑒𝑒𝑖𝑖𝑒𝑒𝑒𝑒 𝐸𝐸𝑖𝑖𝑒𝑒𝑒𝑒 (15) 𝐶𝐶𝑜𝑜𝑓𝑓 = 𝐾𝐾𝑑𝑑(1+𝐾𝐾𝑑𝑑) 𝑖𝑖 (1+𝐾𝐾𝑑𝑑)𝑖𝑖−1 + 𝐾𝐾𝑖𝑖𝑛𝑛𝑠𝑠𝑢𝑢𝑟𝑟𝑎𝑎𝑛𝑛𝑐𝑐𝑒𝑒 (16) where lcoe is the levelized cost of energy, crf is the capital recovery factor, kinv is the total investment of the plant, ko&m is the annual operation and maintenance costs, kelectricity is the annual electricity costs (7 dzd/kwh), enet is the annual net thermal energy, kd is the real debt interest rate = 8 %, n is the depreciation period in years = 25 years, kinsurance is the annual insurance rate = 1 %. 3. results and discussions 3.1. solar field area and optimal configuration in this section, the optimal solar collector area and system configuration using f-chart method are investigated. the thermo-solar system is sized based on the real parameters of the k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 453 issn: 2252-4940/© 2023. the author(s). published by cbiore collector measured in the experimental section. in figure 7, it is shown the variation of the surface area of solar collector in function of the incidence angle and incidence angle modifier during a typical day of the year. figure 7a shows that the collector area varies inversely proportional with the θ and kτα. that means that the closer the collector is to normal incidence (kτα =1 or θ=0) the more the collector’s efficiency increases which obviously induces a minimum surface. this result is in accordance with that found by duffie et al (2013). in our case, as shown in figure 7a and figure 7b, for θ=30° (east) and θ=+30° (west) and where a value of incidence angle modifier of 0.87 is considered (figure 6), an area of 64 m² is obtained, which corresponds to a number of 24 collectors (the collector used has a unit area of 2.6 m2). in order to determine the optimal configuration of the solar field, four cases are considered in function of the number of solar collectors in series. the cases consider two, three, four collectors in series and in the last case all collectors are connected in parallel. the variation of the storage tank temperature for the four different cases studied is shown in figure 8. it was observed that the higher the number of collectors in series, the higher the water outlet temperature. however, only the configuration with four collectors in series is able to reach the operating temperature of the industrial process (60°c) during the period of hot water utilization, as well nunes et al (2014) reported in their work. therefore, the solar system will be composed of 6 blocks of collectors in parallel with 4 collectors in series in each block, (i.e. 24 collectors). -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 0 10 20 30 40 50 60 70 80 90 a re a (m 2 ) θ a) 0,65 0,70 0,75 0,80 0,85 0,90 0,95 1,00 0 10 20 30 40 50 60 70 80 90 a re a (m 2 ) kτα b) fig. 7 determination of the collector’s areas: a) a function of incidence angle; b) a function of incidence angle modifier 6 7 8 9 10 11 12 13 14 15 16 17 18 0 10 20 30 40 50 60 70 te m pe ra tu re (° c) times (hours) parallel collector two series of collectors trie series of collectors four collectors in series process temperature fig. 8 storage tank temperature during a day 6 7 8 9 10 11 12 13 14 15 16 17 18 0,0 0,2 0,4 0,6 0,8 1,0 e ff ic ie nc y times (hours) the blocks of collector in parallel two blocks of collector in series three blocks of collector in series four blocks of collector in series five blocks of collector in series six blocks of collector in series fig.9 variation of the collector’s efficiency in function of the block’s configurations effectively, the result shows that the efficiency of the configuration of 6 blocks in parallel is better than that of configurations with blocks inseries, as shown in figure 9. indeed, the blocks of collectors connected in seriesinduce greater heat losses by convection and radiation and therefore result in a lower thermal efficiency. this design is more efficient from an economic point of view because the comparison of different configurations also indicates that series configurations result in additional pressure drop, which also increases the pumping power (duffie et al.2013) 3.2. solar fraction of the solar system the energy performance the solar system under investigation is determined by calculating the solar fraction of the solar system. the solar fraction is defined as the ratio of the thermal energy produced by the solar collectors and the thermal energy needed for hot water production required by the industrial process (ghafour, 2014). in this work, eq.12 was used to calculate the hourly, monthly and yearly solar fraction considering the variations of the mains water temperature (shown in fig.4) and rhythms of industrial production(hourly daily hot water need, fig.2), the monthly energy needs of hot water production are represented on figure 10. this figure illustrates also the global monthly solar radiation available in this location. we note that, unfortunately, the maximum needs k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 454 issn: 2252-4940/© 2023. the author(s). published by cbiore for hot water correspond to the periods when the availability of solar energy is minimal. indeed, solar radiation energy is high in summer. it reaches up to (6000 kwh) while in winter it is about 2500 kwh. solar energy could be enough to deliver the required amount of hot water during summer. during winter, the solar coverage rate is inevitably lower than in summeras shown in figure 10. it is therefore necessary to properly size the installation to obtain the best energy/economic compromise. figure 11 represents the monthly solar fraction of the solar system. it can reach a maximum of 80% in summer, with an average annual solar fraction of about 50 %. this is due to the best proposed design, the optimal operating parameterschosen for the system and the high intensity of solar radiation at the site. in the work conducted by pahlavan et al (2018),where a water heating solar system is applied in 37 regions in algeria, a yearly solar fraction of 60% was found, which is very close to our results.this mans that such systems are very interesting. 3.3. parametrical study in order to determine the optimum flow rate for the optimum solar collector’s configuration found in the previous section, water mass flow rate varying between 0.01 kg/m2.s and 0.04 kg/m2.s are considered, which are the most useful flow rate range for solar heating installations (duffie et al 2013). the aim was to determine the optimum flow for this configuration of solar collectors. it was observed that this configuration offers the best solar fraction when the mass flow rate is 0.02 kg/s m2. 6 8 10 12 14 16 18 0 40000 80000 120000 160000 200000 0,03 kg/sm2 0,02 kg/sm2 0,01 kg/sm2 u se fu l e ne rg y (j ou le )) times (hours) fig. 12 effect of mass flow rate on the useful energy gain 6 7 8 9 10 11 12 13 14 15 16 17 18 0 1000 2000 3000 4000 5000 6000 w at er lo ad (l ite r) times (hours) a) 6 7 8 9 10 11 12 13 14 15 16 17 18 0 1000 2000 3000 4000 5000 6000 w at er lo ad (l ite r) times (hours) b) 6 7 8 9 10 11 12 13 14 15 16 17 18 0 1000 2000 3000 4000 5000 6000 w at er lo ad (l ite r) times (hours) c) fig. 13 storage capacity for three different casesload profile a) in morning b) afternoon c) shared between hand and afternoon jan feb mar apr may jun jul aug sep oct nov dec -0 1000 2000 3000 4000 5000 6000 e ne rg y (k w h) month solar radiation monthls needs fig.10 monthly solar radiation and monthly energy needs jan feb mar apr may jun jul aug sep oct nov dec 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 so la r f ra ct io n months fig. 11variation of solar energy efficiency during a year k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 455 issn: 2252-4940/© 2023. the author(s). published by cbiore 6 8 10 12 14 16 18 20 0 1x105 2x105 3x105 4x105 5x105 case 1 case 2 case 3 in te rn al e ne rg y ch ar ge (j ou le ) times (hours) fig. 14 effect of the water load moment on the performance of the installation this is the optimum value of mass flow rate for solar water heating systems recommended by iso 9806 and en 12975-2 (kramer et al. 2017) and which is used throughout this work. the results find in this study are also in agreement with those of hang et al. (2012) and ge et al. (2012). as shown in figure 12, for all cases considered the mass flow rate increase gradually with heat flux until reaching its maximum in the middle of the day and then decrease with the falling of the sun. the collector efficiency shows a proportional relationship with the mass flow rate (mandal and ghosh, 2020; yassen et al., 2019). the influence of the load profile is investigated for different cases. a real load profile was used as an input for the model in the brewing water loop to consider the manual interference of the discharge control. it has been considered three cases to highlight the effect of water load on the effectiveness of the installation, as shown in figure 13. the hot water demand occurs, in the first, second and third cases, respectively, in morning (fig.13a), afternoon (fig.13b), and shared between morning and afternoon (fig.13c). it can be seen from figure 14, which represents the effect of the water load moment on the performance of the installation, that the efficiency of the case with a water load in the afternoon (case 2) is higher than that in the morning (case 1). indeed, in the latter case, the storage tank is less charged with solar energy and the storage tank will be recharged with cold water, which induces significant heat losses. it has been observed that the water load in the morning or in the evening has a significant effect on the performance of the installation. therefore, the best system performance is achieved when the water load is occurred in the afternoon. it can also be observed from figure 14, which shows the effect of the number of waters loads on the performance of the installation, that the efficiency of the case with one water load (case 1 and 2) is higher than that with two water loads (case 3), nearly double. this is because in the latter case the storage tank is re-charged with cold water, which also induces significant heat losses. it has been observed that the number of water load during the day has strongly effect of the performance of the installation. this means that processes that operate during many days in the week are not always preferable unless the total load is higher. it can be concluded that the load profile is expected to be the most important influence factors for the system performance of a solar process. indeed, the work of schmitt et al (2012) also shows that the load profile is the most important influencing factor for the system performance of a solar process heat system; 4. economical study after the design phase, the system is investigated in economical point of view. for this reason, various costs of the project components are estimated. table 5 illustrates the costs of each component of the thermo-solar process. as illustrated in table 6, three cases have been considered: case 1: solar system without support (subvention) from the algerian government program of renewable energy; case 2: solar system with support. the algerian government program of renewable energy offers 45% of the total costs of solar collectors (aprue 2022). case 3: conventional system. the heating system is powered by the electrical energy from the grid. the net annual energy produced is calculated taking into account the technical report from the company, that is to say process temperature of 60 °c and the average daily hot water of 3000 liters. the solar processes (cases 1 and 2) are supposed to ensure 50 % of the energy needs (result found in figure 6 average annual solar fraction of about 50 %),which results in annual electricity demand of 25 080 kwh. table 6 shows the actual lcoe in the algerian market for an electricity price of 7dzd/kwh, which is equivalent to 0.05 €/kwh. table 5 costs of each part of the industrial process component unit cost (€) number total cost (€) investment costs collectors 505,52 24 12 132,48 piping 25,28 200 5 056 gate valve 27.33 30 819,9 safety valve 13.66 12 163.92 monometer 21.86 5 109.3 purge 10.25 30 307.5 check valve 13.66 10 136.6 tank 1024.7 2 2049.4 pumps 314.57 5 1572.85 vase expansion 51.23 1 51.23 heat exchanger 512.35 2 1024.7 temperature regulator 81.98 2 163.96 operation and maintenance costs collectors 6.83 24 163.92 tanks 153.7 2 307.4 piping 170.78 1 170.78 source : missoum et al (2021) table 6 total costs of the water heating processes case kinvest (€) ko&m (€) enet (kwh) electricity needs (kwh) lcoe (dzd/ kwh) lcoe (euro/k wh) case 1 23587.84 742.1 50160 25080 12.650 0.108 case 2 12973.31 742.1 50160 25080 9.071 0.067 case 3 3415.67 3204.94 50160 50160 8.246 0.061 1dzd = 0.0069 € (source : external bank of algeria,. 2023) k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 456 issn: 2252-4940/© 2023. the author(s). published by cbiore 8 12 16 20 24 28 32 36 40 5 10 15 20 25 30 35 40 45 lc o e (d zd /k w h) electricity price (dzd/kwh) case 1 case 2 case 3 fig. 15 variation of lcoe as a function of electricity price. the latter value is so much lower than international markets. for instance, the electricity price in germany is about six times more expensive than in algeria (0.30 €/kwh). under the current scenario, the lcoe of the solar thermal process is 0.1€/kwh, but its price with support is 0.067 €/kwh, slightly higher than the one based on electric heating (fossil fuels-based processes) is 0.061€/kwh, a very encouraging result. figure 15 shows the variation of the lcoe of the three processes (cases 1, 2, 3) as a function of electricity prices. a range from 0.048€ to 0.27€/kwh is intentionally considered. the former value is the actual price in the algerian market while the latter is the price in the german market. as showing in figure 15, the solar process with support is becoming competitive to the fossil-based heating process at the electricity price of 0.061€/kwh; a price corresponds to us and canada markets. for the case of a solar process without support from the government, the electricity price should rise to about 0.12€/kwh to ensure competitiveness. this price corresponds to the french electricity market. 5. comparison of results to verify and justify the results we obtained, they were analyzed and compared with the results of two different process heating systems, as shown in table 7, with the same collector technology (flat plate collector), flow rate and set-point water heating temperature tc (60 °c). therefore, the solar fractions of the obtained systems are relatively similar. to verify and justify the obtained results, they were analyzed and compared with the results of two different process heating systems, as shown in table 7, with the same collector technology (flat plate collector), flow rate and set-point water heating temperature tc (60 °c). therefore, the solar fractions of the obtained systems are relatively similar. bahria et al., (2016) investigated the performance of a solar heating system in different climates of algeria. the solar fraction for domestic hot water of 80% has been found. kalogirous et al.(2003) studied a solar industrial process heat in chypre. the solar fraction of system was 75%. in other work, the authors kalogirous et al.(2019), have found for the same system, a lcoe of 0.2 euro/kwh. moreover, for the work presented by strum et al. (2015) on the case study of a medium sized beijing based factory, the result also gave the fraction of 14% and the lcoe of 0.14, which also confirms the accuracy of our result. however, the resulting energy costs (lcoe) obtained for solar heat are 0.1 € /kwh. these results are applicable to all countries with similar weather and economic conditions to these regions. the small difference in the lcoe lies mainly in the price of conventional energy in algeria is subsidized which makes its price a bit low compared to other cases table 7 comparison of results to others studies case solar potential (kwh/m2) collectors type flow rate (kg/sm2) tc (°c) solar fraction lcoe (€/kwh) this study 1651 (fpc) 0.02 60 50 % 0.1 bahria et al 2016 1800 (fpc) 0.02 60 82% kalogirou 2003, 2019) 1727 (fpc) 0.015 60 75% 0.2 sturm et al. (2015) 1400 (fpc) 0.02 60 14% 0.1 6. conclusion this study presents the results of the design, size and energy efficiency of a solar thermal system for the food industry in algeria. the method used to size the system is based on the fchart method of the hwb model, improved by introducing a new power correction factor at each flat-plate incidence angle, which was one of the original features of this work. this value has been experimentally tested and calculated according to the en 12975-2 standard, using real data of energy demand and water consumption curves to determine the actual solar process performance. the results obtained after simulating the model used resulted in a collection arrangement taking into account the requirements, process temperature and the available area of 64 m² (i.e. 24 collectors) on the company's roof. the system consists of 6 collector blocks connected in parallel, with 4 collectors connected in series in each block. simulation results show that a solar coverage of 50% per year can be achieved, which is very high and has sparked great interest in integrating this solar thermal process at selected locations. the parametric study confirmed that the selected mass flow rate is in good agreement with the en 12975-2 and iso 9806 standards. the results also showed that the amount and duration of water exposure during the day had a large impact on plant performance. therefore, to keep up with the fluctuations of the sun, the best time to brew is in the afternoon or early evening. the sensitivity analysis shows that operating temperature, collector selection, and required power are the most important factors affecting system efficiency. the economic evaluation shows that the assisted solar process is competitive with the fossil heating process at an electricity price of 0.062€ /kwh, while the unassisted solar process may be more competitive when the electricity price reaches 0.12€ /kwh. in the current context of global warming, algeria's energy transition is still hampered by the relatively high installation costs of renewable energy compared to the kwh price of fossil electricity. the same is true for most oil and gas production operations. in fact, the latter's current offer is unbeatable. therefore, it is more urgent than ever to develop national economic development plans that promote energy and environmental transition and sustainable development. nomenclature ac1 collectors’ area of the first collector (m2) ac2 collectors’ area of the second collector (m2) ac collectors’ area (m2) cp heat capacity of water (kj/kg°c) f fraction of the monthly heating load(-) f’ corrected fr factor (-) fr collector heat exchanger efficiency factor (-) k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 457 issn: 2252-4940/© 2023. the author(s). published by cbiore f’r corrected collector heat exchanger efficiency factor (-) frug collector overall energy loss coefficient (w/m2°c) gh monthly averaged, daily radiation incident (mj/m2) 𝐻𝐻𝛽𝛽 monthly average daily radiation incident on collector surface per unit area (mj/m2) hm sun height (degree) ig global solar radiation (w/m2) l monthly total heating load hot water (kwh/m2) m actual mass of storage capacity (kg) �̇�𝑚 mass flow rate (kg/s) n days in month(-) qu1 solar useful heat gain rate of the first collector(w) qu2 solar useful heat gain rate of the second collector (w) te inlet fluid temperature (°c) ta ambient temperature (°c) tref an empirically derived reference temperature (°c) tr the mains water temperature (°c) ts storage temperature (°c) 𝑇𝑇𝑎𝑎��� monthly average ambient temperature (°c) td the minimum desired temperature(°c) x modified sensibility factor of the thermal losses (-) 𝑋𝑋𝐶𝐶��� dimensionless average daily critical level of the solar collector (-) y ratio of the absorbed solar energy to the cooling load (-) greek symbols δt total number of seconds in month (s) β collector tilt angle (degree) ε heat exchanger effectiveness (-) (τα)e monthly average transmittance-absorptance product (-) abbreviations iam incidence angle modifier (-) tsv true solar time (hours) references aprue. the national agency for the promotion and rationalization of energy use, (2022) development of the regulatory and incentive framework for energy efficiency in algeria), alsol-programm. https://www.aprue.org.dz mounir, a., rima, z., aziz, n. (2013). technical-economic analysis of solar water heating systems at batna in algeria. in: hakansson, a., höjer, m., howlett, r., jain, l. (eds) sustainability in energy and buildings. smart innovation, systems and technologies, vol 22. springer, berlin, heidelberg. https://doi.org/10.1007/978-3642-36645-1_70 anubhav, u., kesari, j. p., zunai. m., (2016) designing of solar process heating system for indian automobile industry. international journal of renewable energy research, 6(4). 1627-1636. https://doi.org/10.20508/ijrer.v6i4.4490.g6956 bahria, s., amirat, m., hamidat, a, el ganaoui., m, slimani, m. a. (2016). parametric study of solar heating and cooling systems in different climates of algeria–a comparison between conventional and high-energy-performance buildings. energy, 113, 521-535 https://doi.org/10.1016/j.energy.2016.07.022 bangura, a. b. m., hantoro, r., fudholi. a. (2022) mathematical model of the thermal performance of double-pass solar collector for solar energy application in sierra leone. international journal of renewable energy development, 11(2), 347-355. https://doi.org/10.14710/ijred.2022.41349 center for the development of renewable energies in algeria., (2021) radiometric station. https://www.cder.dz/?lang=en deepika, d., baig, m. a., reddy, a. r., maneaih, d. (2016). utilization of f-chart method for designing solar thermal heating system. journal of mechanical and civil engineering, 5, 23-28, https://doi.org/10.9790/1684-16053042328 desai, d.d., raol, j.b, patel, s, chauhan, i. (2013) application of solar energy for sustainable dairy development. eur j sustain. dev. 2, 131-40. https://doi.org/10.14207/ejsd.2013.v2n2p131 duffie, j. a., william, a., beckman, a. w. (2013) solar engineering of thermal processes. 4th ed. published by john wiley & sons; https://doi.org/10.1002/9781118671603 farjana, s. h., huda, n., parvez., mahmud, m.a. and saidur. r. (2017) solar process heat in industrial systems – a global review. renewable and sustainable energy reviews. 82(3), 2270-2286, https//doi.org/10.1016/j.rser.2017.08.065 fernández, a. c. (2013) economic study of solar thermal plant based on gas turbines. master thesis; lth school of engineering. http://hdl.handle.net/2099.1/19690 fuller, r.j., (2011) solar industrial process heating in australia, past and current status. renewable energy, 36, 216-221; https://doi.org/10.1016/j.renene.2010.06.023 ge, t.s., dai, y.j., li, y., wang, r.z. (2012) simulationinvestigation on solar powered desiccant coated heatexchanger cooling system. applied energy, 93, 532-540; https://doi.org/10.1016/j.apenergy.2011.11.089 ghafoor, a., munir, a. (2014) thermo-economic optimization of solar assisted heating and cooling (sahc) system, international journal of renewable energy development, 3(3), 217-227. https://doi.org/10.14710/ijred.3.3.217-227 hang, y., qu, m., zhao, f. (2012) economic and environmental life cycle analysis of solar hot water systems in the united states. energy and buildings, 45, 181-188; https://doi.org/10.1016/j.enbuild.2011.10.057 intergovernmental panel on climate change ipcc (2022) mitigation of climate change; assessment report on climate change. https://www.ipcc.ch/report/ james & james., (2010) planning &installing, solar thermal systems. fully revised & updated 2nd edition. london: earthscan https://doi.org/10.4324/9781849775120 kaci, k., laoudj, o., merzouk, m., kasbadji, n., sami, h. (2014) optimization of collector area for solar heating. j. macro trends health med. https://macrojournals.com/yahoo_site_admin/assets/docs/8e s21al.960840.pdf kaci, k., merzouk, m., kasbadji, n. (2012) effect of tests norms on the instantaneous efficiency of a plate solar collector. procedia engineering 33, 392-403; https://doi.org/10.1016/j.proeng.2012.01.121 kalogirou, s. (2003).the potential of solar industrial process heat applications. appl. energy. 76(4), 337–361; https://doi.org/10.1016/s0306-2619(02)00176-9 kalogirou, s. (2019) applications of renewable energy systems in buildings and industry; report of department of mechanical engineering and materials sciences and engineering. http://www.oeb.org.cy/wp-content/uploads/2019/11/3 klein, k. a., beckman, w. a., duffie, j. a. (2013) a design procedure for solar heating systems.solar energy. 18, 113-127. https://doi.org/10.1016/0038-092x(76)90044-x kramer, k., mehnert, s., (2017) guide to standard iso 9806:2017 a resource for manufacturers, testing laboratories, certification bodies and regulatory, technical report, https://doi.org/10.13140/rg.2.2.30241.30562 kummert, m., andre, p., nicolas, j. (2001). optimal heating control in a passive solar commercial building. solar energy, 69, 103-116. https://doi.org/10.1016/s0038-092x(01)00038-x laboref, sarl., food and beverage company(2013), the technical report from the company.laboref . https://www.laboref-dz.com lima, t., charamba, c.j. (2015) solar water heating for a hospital laundry: a case study, solar energy. 122, 737-748, https://doi.org/10.1016/j.solener.2015.10.006 mandal, s., ghosh, s.k. (2020) experimental investigation of the performance of a double pass solar water heater withreflector. renewable energy, 149, 631-640; https://doi.org/10.1016/j.renene.2019.11.160 . meyers, s., schmitt, b., vajen, k . (2018) the future of low carbon industrial process heat: a comparison between solar thermal and heat pumps. solar energy. 173, 893-904. https://doi.org/10.1016/j.solener.2018.08.011 meteonorm 8. (2022). meteonorm software worldwide europe irradiation data. intersolar europe. https://meteonorm.com/en/news ministry of water resources and water security, algerian waters , (2021), annual report. https://www.aps.dz/en https://www.aprue.org.dz/ https://doi.org/10.1007/978-3-642-36645-1_70 https://doi.org/10.1007/978-3-642-36645-1_70 https://doi.org/10.20508/ijrer.v6i4.4490.g6956 https://doi.org/10.1016/j.energy.2016.07.022 https://doi.org/10.14710/ijred.2022.41349 https://www.cder.dz/?lang=en https://doi.org/10.9790/1684-16053042328 https://doi.org/10.14207/ejsd.2013.v2n2p131 https://doi.org/10.1002/9781118671603 http://dx.doi.org/10.1016/j.rser.2017.08.065 http://hdl.handle.net/2099.1/19690 https://doi.org/10.1016/j.renene.2010.06.023 https://doi.org/10.1016/j.apenergy.2011.11.089 https://doi.org/10.14710/ijred.3.3.217-227 https://doi.org/10.1016/j.enbuild.2011.10.057 https://www.ipcc.ch/report/ https://doi.org/10.4324/9781849775120 https://macrojournals.com/yahoo_site_admin/assets/docs/8es21al.960840.pdf https://macrojournals.com/yahoo_site_admin/assets/docs/8es21al.960840.pdf https://doi.org/10.1016/j.proeng.2012.01.121 https://doi.org/10.1016/s0306-2619(02)00176-9 http://www.oeb.org.cy/wp-content/uploads/2019/11/3 https://doi.org/10.1016/0038-092x(76)90044-x https://doi.org/10.13140/rg.2.2.30241.30562 https://orbi.uliege.be/browse?type=author&value=kummert%2c+micha%c3%abl https://orbi.uliege.be/browse?type=author&value=andre%2c+philippe+p004269 https://orbi.uliege.be/browse?type=author&value=nicolas%2c+jacques+p004274 https://doi.org/10.1016/s0038-092x(01)00038-x https://www.laboref-dz.com/ https://doi.org/10.1016/j.solener.2015.10.006 https://doi.org/10.1016/j.renene.2019.11.160 https://doi.org/10.1016/j.solener.2018.08.011 https://meteonorm.com/en/news https://www.aps.dz/en k. kaci et al int. j. renew. energy dev 2023, 12(3), 448-458 | 458 issn: 2252-4940/© 2023. the author(s). published by cbiore missoum, m., loukarfi, l. (2021) investigation of a solar poly generation system for a multi-storey residential building-dynamic simulation and performance analysis. int. journal of renewable energy development. 10(3), 445-458; https://doi.org/10.14710/ijred.2021.34423 müller, t., weiß, w., schnitzer, h., brunner, c. (2004) potential study on the use of thermal solar energy in austrian commercial and industrial companies. annual report. https://doi.org/10.1016/j.rser.2012.04.032 national renewable energy laboratory. (2021) decarbonizing industrial process heat, annual report. https://www.frontiersin.org/journals/energy-research núñez, m. p., odríguez. g. m., amanda. l., silva. f. (2014) design of solar collector networks for industrial applications. applied thermal engineering, 1-8. https://doi.org/10.1016/j.applthermaleng.2014.05.005 okafor, i. f., akubue, g. (2012) f-chart method for designing solar thermal water heating systems. international journal of scientific & engineering research, 3-9, https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf& doi=094d081464742e2226f7700239fcb6dc953cd0c5 pahlavana, s., jahangirib, m., shamsabadic, a., khechekhouche, a., (2018). feasibility study of solar water heaters in algeria, a review. journal of solar energy research, 3(2), 135-146. https://jser.ut.ac.ir/article_67424.html palaniappan, c., (2009). perspectives of solar food processing in india:1–11. iinternational solar food processing conference. https://solarcooking.fandom.com/wiki/international_solar_fo od_processing_conference_2009 quijera, j.a., alriols, m.g., labidi, j. (2011). usage of solar energy in an industrial process. chem eng trans; 25, 875–80. https://doi.org/10.3303/cet1125146 renewable energy policy network for the 21st century . (2016) global status report. annual report. https://renewablepower.theiet.org schmitt, b., lauterbach, c., dittmar, m., vajen, k., (2012). guideline for the utilization of solar heat in breweries. proceedings eurosun rijeka, croatia. schweiger, h., farinha, m., schwenk, c., hennecke, k., (2000)the potential of solar heat for industrial processes, barcelona, lissabon, münchen, köln, madrid, spain, poship. final report. http://www.solarpaces.org/library/docs/poship_final_repo rt.pdfs . sharma, a., k, sharma, c., mullicka. s. c and kandpala, t. c. (2017) solar industrial process heating: a review. renewable and sustainable energy reviews. 78, 124–137. https://doi.org/10.1016/j.rser.2017.04.079 sturm, b., meyers, s., zhang, y., law, r., eric, j., valencia, s., bao, h., wang, y., chen, h. (2015) process intensification and integration of solar heat generation in the chinese condiment sector – a case study of a medium sized beijing based factory. energy conversion and management, 106, 1295-1308; https://doi.org/10.1016/j.enconman.2015.10.045 suresh, n.s. (2017) solar energy for process heating: a case study of select indian industries . journal of cleaner production, (151), 439451. https://doi.org/10.1016/j.jclepro.2017.02.190 the international energy agency.,(2015) process heat collectors: state of the art and available medium temperature collectors, task 49. annual report. https://task49.ieashc.org/data/sites/1/publications/task%2049%20deliverable %20a1.3_20160504.pdf the international renewable energy agency . (2015) solar heat for industrial processes, technology brief e21. annual report. https://www.irena.org//media/files/irena/agency/publication/2015/irena_etsa p_tech_brief_e21_solar_heat_industrial_2015.pdf the international renewable energy agency. (2015) quality infrastructure for renewable energy technologies solar water heater, annual report, https://www.irena.org/publications/2015/dec/qualityinfrastructure-for-renewable-energy-technologies-solar-waterheaters the international renewable energy agency. (2015) renewable energy options for the industry sector, global and regional potential until 2030. annual report, https://www.irena.org//media/files/irena/agency/publication/2014/aug/irena_ re_potential_for_industry_bp_2015.pdf the international energy agency (iea). (2020) renewables 2019 analysis and forecast to 2024. annual report; https://iea.blob.core.windows.net/assets/a846e5cf-ca7d-4a1fa81b-ba1499f2cc07/renewables_2019.pdf the minister of energy transition and renewable energies., (2022). the energies and energy efficiency program. annual report; https://www.energy.gov.dz valderrama, c., cortina, j. l., aliakbar, a., bawahab, m. (2022) solar ponds. storing energy (second edition), 2 537-558; https://doi.org/10.1016/b978-0-12-824510-1.00004-0 weiss, w., (2005) solar heating worldwide, markets and contribution tot he energy supply , iea, solar heating and cooling program. annual report. https://www.ieashc.org/data/sites/1/publications/solar-heat-worldwide2022.pdf zühlsdorf, b., bühlera, f., bantlec, m., elmegaarda, b. (2019) analysis of technologies and potentials for heat pump-based process heat supply above 150 °c. energy conversion and management x, 2. https://doi.org/10.1016/j.ecmx.2019.100011 zwalnana, s. j., duvunab, g. a., abakra, y. a., bandaa, t., (2021) design and performance evaluation of a multi-temperature flat plate solar collector. int. journal of renewable energy development, 10(3),537-549. https://doi.org/10.14710/ijred.2021.33213 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.14710/ijred.2021.34423 https://doi.org/10.1016/j.rser.2012.04.032 https://www.frontiersin.org/journals/energy-research https://doi.org/10.1016/j.applthermaleng.2014.05.005 https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=094d081464742e2226f7700239fcb6dc953cd0c5 https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=094d081464742e2226f7700239fcb6dc953cd0c5 https://jser.ut.ac.ir/article_67424.html https://solarcooking.fandom.com/wiki/international_solar_food_processing_conference_2009 https://solarcooking.fandom.com/wiki/international_solar_food_processing_conference_2009 https://doi.org/10.3303/cet1125146 https://renewablepower.theiet.org/ http://www.solarpaces.org/library/docs/poship_final_report.pdfs http://www.solarpaces.org/library/docs/poship_final_report.pdfs https://doi.org/10.1016/j.rser.2017.04.079 https://doi.org/10.1016/j.enconman.2015.10.045 http://dx.doi.org/10.1016%2fj.jclepro.2017.02.190 https://task49.iea-shc.org/data/sites/1/publications/task%2049%20deliverable%20a1.3_20160504.pdf https://task49.iea-shc.org/data/sites/1/publications/task%2049%20deliverable%20a1.3_20160504.pdf https://task49.iea-shc.org/data/sites/1/publications/task%2049%20deliverable%20a1.3_20160504.pdf https://www.irena.org/-/media/files/irena/agency/publication/2015/irena_etsap_tech_brief_e21_solar_heat_industrial_2015.pdf https://www.irena.org/-/media/files/irena/agency/publication/2015/irena_etsap_tech_brief_e21_solar_heat_industrial_2015.pdf https://www.irena.org/-/media/files/irena/agency/publication/2015/irena_etsap_tech_brief_e21_solar_heat_industrial_2015.pdf https://www.irena.org/publications/2015/dec/quality-infrastructure-for-renewable-energy-technologies-solar-water-heaters https://www.irena.org/publications/2015/dec/quality-infrastructure-for-renewable-energy-technologies-solar-water-heaters https://www.irena.org/publications/2015/dec/quality-infrastructure-for-renewable-energy-technologies-solar-water-heaters https://www.irena.org/-/media/files/irena/agency/publication/2014/aug/irena_re_potential_for_industry_bp_2015.pdf https://www.irena.org/-/media/files/irena/agency/publication/2014/aug/irena_re_potential_for_industry_bp_2015.pdf https://www.irena.org/-/media/files/irena/agency/publication/2014/aug/irena_re_potential_for_industry_bp_2015.pdf https://iea.blob.core.windows.net/assets/a846e5cf-ca7d-4a1f-a81b-ba1499f2cc07/renewables_2019.pdf https://iea.blob.core.windows.net/assets/a846e5cf-ca7d-4a1f-a81b-ba1499f2cc07/renewables_2019.pdf https://www.energy.gov.dz/ https://www.sciencedirect.com/science/book/9780128245101 https://doi.org/10.1016/b978-0-12-824510-1.00004-0 https://www.iea-shc.org/data/sites/1/publications/solar-heat-worldwide-2022.pdf https://www.iea-shc.org/data/sites/1/publications/solar-heat-worldwide-2022.pdf https://www.iea-shc.org/data/sites/1/publications/solar-heat-worldwide-2022.pdf https://doi.org/10.1016/j.ecmx.2019.100011 https://doi.org/10.14710/ijred.2021.33213 design, optimization and economic viability of an industrial low temperature hot water production system in algeria: a case study karim kacia0f , mustapha merzouka, nachida kasbadji merzoukb, mohammed missoumc, mohammed el ganaouid, omar behare, rabah djedjigd 1. introduction 2. methodology 2.1. presentation of the case study 2.2. weather data and validation 2.3. determination of the solar collector’s parameters 3. results and discussions 3.1. solar field area and optimal configuration 3.2. solar fraction of the solar system 3.3. parametrical study 4. economical study 6. conclusion int. j. renew. energy dev. 2023, 12(2), 288-299 | 288 https://doi.org/10.14710/ijred.2023.48672 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id wind speed prediction based on statistical and deep learning models ilham tyassa* , tajeddine khalilib , mohamed rafika , bellat abdelouaheda , abdelhadi raihania , khalifa mansouria aenset mohammedia, eeis laboratory, hassan ii university of casablanca, morocco blicpm laboratory, ensa béni mellal, sultan moulay sliman university, béni mellal, morocco. abstract. wind is a dominant source of renewable energy with a high sustainability potential. however, the intermittence and unstable nature of wind source affect the efficiency and reliability of wind energy conversion systems. the prediction of the available wind potential is also heavily flawed by its unstable nature. thus, evaluating the wind energy trough wind speed prevision, is crucial for adapting energy production to load shifting and user demand rates. this work aims to forecast the wind speed using the statistical seasonal auto-regressive integrated moving average (sarima) model and the deep neural network model of long short-term memory (lstm). in order to shed light on these methods, a comparative analysis is conducted to select the most appropriate model for wind speed prediction. the errors metrics, mean square error (mse), root mean square error (rmse), mean absolute error (mae), and mean absolute percentage error (mape) are used to evaluate the effectiveness of each model and are used to select the best prediction model. overall, the obtained results showed that lstm model, compared to sarima, has shown leading performance with an average of absolute percentage error (mape) of 14.05%. keywords: wind energy, wind speed, time series, forecasting models, sarima, deep learning, lstm. @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 1st sept 2022; revised: 16th dec 2022; accepted: 10th january 2023; available online: 23rd january 2023 1. introduction renewable energies are a serious alternative to fossil fuels. these type of energy sources are sustainable, extremely abundant within nature and can be a reliable alternative to classic energy generation methods. wind energy, particularly, is one of the most exploited energy sources; it is a clean source, with a highperformance coefficient, it assures a good energy independence. the most imposing drawback of wind energy is the high intermittence within the year cycle (asari et al., 2002). because of its fluctuating nature, wind speed is an extremely unpredictable meteorological factor. increasing the accuracy of wind forecasting is necessary to optimize wind farm operations, maximize their yield, and ensure a steady development (devis et al., 2018). wind energy production plants use special equipment to determine wind direction in order to control the rotor axis direction (kodjo et al., 2008). however, the wind direction has a low influence on the energy production, especially since wind tends to have one or two dominant directions for which most of the energy is produced (adekunle, 2017). thus, the unpredictable nature of wind potential, affecting the yield of wind turbines is related to the variation of wind speed itself for a given direction. currently, there are two methods to forecast the wind speed. the first one, named indirect technique, involves predicting wind speed based on environmental data, such as wind direction, temperature, air humidity, etc. the second way is to * corresponding author email: ilham.tyass-etu@etu.univh2c.ma (i. tyass) use past wind speed data to forecast wind speed for the next hours or days. according to the time horizon, the time series prediction is divided into three categories: (i) an extremely short-term prediction, which is very useful for intraday market trading and it represents a few minutes to an hour, (ii) a shortterm prediction approach that is suitable for maintenance planning and it is from one hour to 12 hours. finally, (iii) a medium to long-term prediction, which is useful for maintaining non-renewable energy production. the prediction horizon ranges from several hours to several days. to grasp wind speed forecasting, several approaches have been proposed. these methods can be separated into physical and statistical methods. generally, the physical models require ample physical background information, so that they are not appropriate for wind speed forecasting in electric power systems (mi et al., 2019). in the statistical methods, the historical data are used to train time series models, e.g., auto-regression (ar) model, autoregressive moving average (arma) model (erdem and shi, 2011), kalman filter, and artificial neural network (ann) (zhang et al., 2016). the purpose of this article is to compare the predicting results of the seasonal autoregressive integrated moving average (sarima) model with the deep neural network model of long short-term memory (lstm) using the same dataset and diverse evaluation metrics. in the first section of this paper, various kinds of forecasting models are described, and then compared based on the analysis of several contents in the literature. in the second section, wind research article https://doi.org/10.14710/ijred.2023.48672 https://doi.org/10.14710/ijred.2023.48672 mailto:ilham.tyass-etu@etu.univh2c.ma https://orcid.org/0000-0001-5803-6776 https://orcid.org/0000-0001-9559-585x https://orcid.org/0000-0002-4044-9035 https://orcid.org/0000-0002-2137-2618 https://orcid.org/0000-0002-6295-274x https://orcid.org/0000-0003-2242-0973 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.48672&domain=pdf i. tyassl et al int. j. renew. energy dev 2023, 12(2), 288-299 | 289 issn: 2252-4940/© 2023. the author(s). published by cbiore speed time series data is studied using the time-series decomposition method. the third paragraph discusses a shortterm wind speed forecasting using two of the most frequently employed time-series forecasting methods: sarima and the lstm models. in the last section of the paper, the obtained results are carefully analyzed. to evaluate the performance of the wind speed forecasting model, data from three different velocity month were used, and four kinds of error metrics were considered: the mean squared error (mse), root mean squared error (rmse), mean absolute error (mae) and mean absolute percentage error (mape). 2. related work according to the methodology, there are three major types of weather forecasting systems namely, physical, deterministic and hybrid systems. physical systems utilize probabilistic approaches to indicate weather event probability, deterministic approaches produce more precise weather forecasts for a given location, and hybrid models which combine multiple individual prediction models to overcome several prediction limits (jaseena and kovoor, 2020). in the literature, all these approaches are applied for wind speed forecasting systems: 2.1. physical models in general, a physical model is an interpretable model, which permits the identification of causal relationships and has the possibility to generate data by performing various numerical simulation experiments. however, it can only approximate complex cases and only simulates well-known physical laws. physical models are used to forecast the wind resource, beginning with meteorological data and adjusting it to local physical effects. the physical method does not require any training inputs from older data; they provide real-time physical information of wind farm, such as wind speed, wind direction, temperature, humidity, air pressure, air density (bessac et al., 2018). combined with the information of topography around the wind farm (terrain, obstacle, contour etc.), the wind speed at the wind turbine’s mast height is calculated, and the output of the turbine may then be determined. the use of inaccurate meteorological parameters induces system faults, which subsequently causes a cumulative impact of error, and reduce the forecast accuracy. the physical model uses the mathematical approaches of the atmosphere, they are mainly based on numerical weather prediction model; the nwp model is a program that resolves complex equations and describes atmospheric processes and how they evolve over time. authors in (lorenc, 1986), have proposed a nwp model that uses bayesian probabilistic arguments. however, the model does not exhibit accurate results in short-term prediction. indeed, the performance of the physical model is relatively weak when the wind speed is very random, it is only used for long-term forecasting or it is considered as an input for the statistical method. evidence illustrates that, to achieve the objective of accurate wind speed prediction, the statistical model built based on probability theory and mathematical statistical method is identified as the most widely used model. 2.2. deterministic models deterministic models can be classified into statistical models and artificial intelligence models. (i) the statistical methods consider the training of history of wind speed data and produces a result without considering the effect physical phenomena. a statistical method contains kalman filters model; in this method, wind speed is considered a state variable leading to the establishment of state space representation. kalman filters algorithm is then used to forecast the future wind speed. this approach commonly increases the accuracy of nwp models predictions (cassola and burlando, 2012). the fuzzy logic model, is a polyvalent statistical approach for wind forecasting, which the values of variables are real numbers between 0 and 1. it is based on qualitative variables to model complex systems, it is generally best suited for systems and phenomenon, for which it’s not possible to establish an exact model (damousis et al., 2004). another statistical method is the conventional statistical model; they are similar to the direct random time-series model. model identity, parameter estimation, and model validation are used to develop a mathematical solution to the problem. these models can be categorized into the following categories: autoregressive model (ar), moving average model (ma), autoregressive moving average model (arma), auto regressive integrated moving average model (arima) (yatiyana et al., 2017). (ii) artificial intelligence (ai) models have proven to be more accurate and efficient at handling nonlinear data sets and providing better forecasting results. these models are divided between machine learning (ml) approaches for forecasting and deep learning predictors. among the most commonly used machine-learning predictors for forecasting are ann and support vector machine (svm) models. thanks to its ability to convey non-linear correlations between previous weather patterns and future weather conditions, ann-based models (faniband and shaahid, 2020) are one of the most common approaches for weather prediction. neural networks are nonlinear regression models that can rapidly process large amounts of data and efficiently match input and output variables. svm are supervised machine learning models focused on solving mathematical discrimination and regression problems. svms are appreciated for their simplicity of application, they can handle classification and regression issues (brereton and lloyd, 2010). deep learning is a subcategory of machine learning that uses an artificial neural network approach to extract intelligence from big data sets. this technique employs supervised or unsupervised methods in deep architectures to create hierarchical representations. rnns are employed in deep learning and in the generation of models that simulate the activity of the human brain system. they are particularly powerful in forecasting results. for classical rnns, the absence of the gradient is a severe issue, since the neural network, unable to be properly trained, would surely lose performance. lstm (long short-term memory) units are one of the solutions to this problem (tokgöz and ünal, 2018). lstm based architectures are capable of capturing long-term dependencies with much more precision. 2.3. hybrid models hybrid predictive models were introduced to address the limits of single models and increasing the performance of wind speed forecasting, a single model is not sufficient to handle complex real-world systems with unknown mixed models. such as the combination of ann and arima model forecasts proposed in (li et al., 2011) (nair et al., 2017), the combination of long shortterm memory (lstm) network and the decomposition methods using the grey wolf optimizer (gwo) (altan et al., 2021). in literature, the most common and extensively used hybrid models are parameter optimization and data preprocessing based models (hajirahimi and khashei, 2019). the data preprocessing-based hybrid models, generally rely on data preparation approaches, the time series is converted split into many subsets of data. authors in (nguyen and novák, 2019) created a hybrid model based on preprocessing to forecast seasonal time series. i.tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 290 issn: 2252-4940/© 2023. the author(s). published by cbiore table 1 comparison of wind speed forecasting models wind speed forecasting model strengths weaknesses models based on bayesian approach. (miranda and dunn, 2006) (tascikaraoglu and uzunoglu, 2014). it offers a high flexibility in the modeling, allowing the inclusion of information from physical phenomena that may have an influence on the variable studied. this approach demands more work and time and requires a certain level of skill on the part of the user. nwp models (yang et al., 2018) (bennitt and schueler, (2012).) (martinez-garcía et al., 2021). provides good results for longer prediction terms. the spatial resolution of numerical weather prediction models is continuously growing, providing a better representation of weather characteristics. requires longer processing times and advanced computational resources. not appropriate for short forecast times. difficult to model. requires the analysis and understanding of a variety of conditions. kalman filter models (wu et al., 2004) (babazadeh et al., 2012) (hide et al., 2003) good performance in linear regression methods because of its recursive form it is not needed to store historical data needs extensive previous information of the system. divergent filter estimates due to stability problems. fuzzy logic models (bououden et al., 2012). appropriate for structures that are harder to model precisely. very complicated, and needs a considerable processing time when there are numerous rules. time series models (ar, arma, arima) (torres et al., 2005) (eldali et al., 2016) the structure is relatively simple, it’s possible to correct local trends in data, requires a much longer historic records, difficult to model nonlinear problems. ann-based models (navas et al., 2020) (nazir et al., 2020). before analysis, there is no requirement to establish a mathematical model, not very susceptible to input data errors, better adaptability to online measures. needs a training process and a large number of training data sets. svm-based models (ranganayaki and deepa, 2017) (pinto et al., 2014) good efficiency in generalization requires a good setting of parameters, using a complicated optimization approach, and long training time. lstm based models (araya et al., 2020) (geng et al., 2020). capable of capturing long-term dependencies more precisely. needs a training process and a big number of training data sets. hybrid models (soman et al., 2010) (zhang et al., 2020). achieves a good prediction ability with higher precision, and maximizes the approximation of the actual value. achieves excellent stability performance. difficulty of analyzing correlated observations and their temporal orders. sequential nature of the data. require complex and specific resolution techniques. the parameter optimization models are established using optimization algorithm, specifically meta heuristic techniques, thanks to its ample searching benefits (qian et al., 2019) 2.4. literature synthesis different forecasting models have been used trough literature. each model includes a special set of strengths and weaknesses. table 1 summarizes numerous references with an explication of the strengths and weaknesses of the methods used. based on this literature analysis, numerous researches work related to wind speed forecasting and power using various models have been used; these approaches often produce credible results, and each method has strengths and weaknesses. the appropriate model is chosen based on the specific data characteristics of the site and the application area of the method, though, the developed prediction models are typically site-specific and greatly influenced by the change in the prediction times required. 3. mathematical fundamentals 3.1. problem formulation the problem consists of predicting the future wind speed values from historical wind data. assuming that 𝑊 denotes the input set that includes the recorded wind speed data in the past up to the time 𝑡, and �̂� means the output vector that refers to the predicted wind speed values over a given prediction horizon ℎ. figure 1 shows the wind speed forecasting paradigm. 𝑊 and �̂� are defined below: { 𝑊 = [𝑤𝑡−𝑁 , 𝑤𝑡−(𝑁−1), … , 𝑤𝑡 ] �̂� = [�̂�𝑡+1, �̂�𝑡+2, �̂�𝑡+3, … , �̂�𝑡+ℎ−1, �̂�𝑡+ℎ ] (1) fig. 1 wind speed forecasting paradigm 𝑥𝑡−𝑁 𝑥1 𝑥𝑡 𝑥𝑡+ℎ 𝑁 + 1 𝑣𝑎𝑙𝑢𝑒𝑠 ℎ 𝑣𝑎𝑙𝑢𝑒𝑠 𝑇𝑖𝑚𝑒 w in d s p e e d i. tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 291 issn: 2252-4940/© 2023. the author(s). published by cbiore the prediction operation consists to forecast the �̂� values from the known 𝑊 records with the objective to minimize the prediction error (𝑤𝑡+𝑖 − �̂�𝑡+𝑖 ) for each 𝑖 = 0 𝑡𝑜 ℎ. thus, the prediction problem can be formulated as follow: [�̂�𝑡+1, �̂�𝑡+2, … , �̂�𝑡+ℎ−1, �̂�𝑡+ℎ ] = 𝑓(𝑤𝑡−𝑁 , 𝑤𝑡−(𝑁−1), … , 𝑤𝑡 ) (2) where 𝑓 denotes the forecasting method. in this study, sarima and lstm methods are used to predict the time related wind speed. those models would be compared based on several metrics as detailed below. 3.2. arima model arima model, which means auto-regressive integrated moving average is the most popular stochastic models in time series forecasting (siami-namini et al., 2018). arima is a time series model that can be employed to analyze and forecast next values in the series. arima models are determined by the combination of three different features (𝑝,𝑑 and 𝑞). the parameters 𝑝, 𝑑 and 𝑞 are integers superior or equals to 0 and are described as follow: p: represents the seasonality, it’s specifying the number or order of the ar terms. d: the trend, it’s the number or order of the differences q: the noise, it represents the number or order of the ma, moving average terms. the fundamental forecasting equation in terms of y is: �̂�𝑡 = ϕ1 y𝑡−1 + ⋯ + ϕ𝑝 y𝑡−𝑝 + 𝑒𝑡 − θ1𝑒𝑡−1 − ⋯ θ𝑞 𝑒𝑡−𝑞 (3) where ϕi are the auto-regression coefficients, θj are the moving average coefficients of the model, et is the noise, and yt represents the wind speed value at time t. a more detailed version of the arima model is seasonal arima (sarima), this technic proposes to model the seasonality of the time series by adding the period parameter s and the coefficients p, d and q equivalent to the parameters p, d and q of the differentiated time series (farida and zeghdoudi, 2020) the factors p, d, q, s, p, d and q are fixed by improving the akaike information criteria (aic). this measure is based on striking a balance between a model's complexity and its fit (mantalos et al., 2010). 3.3. deep learning models 3.3.1. recurrent neural network (rnn) a recurrent neural network (rnn) is a developed branch of artificial neural network (ann) that aids in sequence modeling. the recurrent neural networks have the capacity to build on previous types of networks. in rnn, an input sample is added to previously recorded outputs that are included as new inputs. although rnns are efficient, they are affected by the vanishing gradient problem, which makes learning large data sequences very difficult. the gradient transmits information used in updating the rnr parameters and when the gradient becomes significantly smaller, the parameter updates become inconsequential, indicating that no meaningful learning is done. instead, better variation of rnns is used: long short-term networks (lstm) (tian et al., 2018). 3.3.2. long short-term memory (lstm) lstm networks are a category of rnn that employs a combination of special and standard units. a “memory cell” in a lstm unit can store information and keep it for lengthy periods, and a collection of gates is used to manage information. as a result, they use an activation function to determine the activation of a weighted sum. the structure of lstm cell is shown in figure 2. the lstm cell contains the following components: forget gate f, candidate layer c, input gate i, output gate o, hidden state h, memory state c, inputs of the lstm cell at any step are 𝑋𝑡 (current input), ht-1 (previous hidden state) and 𝑐𝑡−1 (previous memory state). the lstm cell produces two outputs: ht (current hidden state) and 𝑐𝑡 (current memory state). firstly, the lstm cell get the prior memory state 𝑐𝑡−1 and performs by element multiplication with the forget gate f. if the forget gate value is 0, the prior memory state is totally forgotten; if it is 1, the previous memory state is completely transferred to the cell. (f gate values are between 0 and 1). 𝑐𝑡 = 𝑐𝑡−1 × 𝑓𝑡 (4) calculating the new memory state: 𝑐𝑡 = 𝑐𝑡 + (𝑖𝑡 × 𝑐 ′ 𝑡 ) (5) thus, the output is ℎ𝑡 = 𝑡𝑎𝑛ℎ(𝑐𝑡 ) (6) 3.4. evaluation metrics validating a prediction model is indispensable to ensure that the model is indeed capable of accurately predicting the values of a variable of interest. the focus is on whether the values predicted by the model are close to the true values in the validation data set. the statistical error metrics described in the following are the most commonly reported indicators: the mean square error (mse), the root mean square error (rmse), mean absolute error (mae) and mean absolute percentage error (mape) (botchkarev, 2019): • mse and rmse the mean square error of an estimator of a parameter is a measure characterizing the precision of this estimator; it measures the average of the square of the error’s deviancies. however, the rmse is the square root of the second sampling moment of discrepancies between forecasted and observed values. because the influence of each error is proportional to the amount of the error squared, greater errors have a disproportionate effect on rmse. (hossin and sulaiman, 2015), (kamble and deshmukh, 2017). 𝑅𝑀𝑆𝐸𝑦,�̂� = √ 1 𝑛 ∑ (𝑦𝑗 − �̂�𝑗 ) 2𝑛 𝑗=1 (7) fig. 2 the structure of lstm cell i. tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 292 issn: 2252-4940/© 2023. the author(s). published by cbiore • mae mean absolute error (mae) is a statistical measure for comparing two continuous variables. given that, y and y1 are variables of paired observations expressing the same phenomena. comparisons of expected against observed, subsequent time versus starting time, and one measuring technique versus another measurement technique are examples of y1 versus y (botchkarev, 2019): 𝑀𝐴𝐸𝑦,�̂� = 1 𝑛 ∑ |𝑦𝑗 − �̂�𝑗 | 𝑛 𝑗=1 (8) the mean absolute error is one of the main methods used for comparing predictions with their eventual results. • mape the mean absolute percentage error (mape) is one of the most widely used indicators of prediction performance. it was used as the primary measure in the m-competition (kim and kim, 2016). 𝑀𝐴𝑃𝐸𝑦,�̂� = 1 𝑛 ∑ | 𝑦𝑗−�̂�𝑗 𝑦𝑗 |𝑛𝑗=1 (9) mape has a severe disadvantage in generates undefined values when the real values are null or near to zero (tayman and swanson, 1999). 4. materials and methods 4.1.methodical approach for the work process a methodical approach was developed to better implement and assess the two methods for wind speed prediction; the work process is depicted in the diagram shown in figure 3. fig. 3 methodical approach for the work process fig. 4 wind data visualization : (a) march 2018, (b) july 2018 and (c) october 2018 4.2.data preparation the meteorological data for this study was recorded at the site "abdelkhalak torres" in "al koudia al baida -tetouen" located at (latitude: 35° 45' 35.1, longitude: -5° 41' 19.9''). the data used in this is recorded daily with a 10 𝑚𝑖𝑛 sampling period at a height of 100 m above the ground. the data that was specifically used for this study was recorded between january 1st and december 31st of the year 2018. records includes date, time of the record and the wind speed. the data that was considered for the validation of the prediction model is related to three months: march, july and october of the year. data cleaning is a crucial step of data preparation. missing data might be the result of an instrument malfunction or a discrepancy with previously recorded data. in the suggested method, the mean of the previous five observations is used to fill in the missing data. the analysis and processing of the data was carried out by the python 3.8 programming language, with the spyder 5.0.0 application available on the graphic interface of the anaconda distribution. data preprocessing data cleaning data scaling data analysis data visualization time series decomposition prediction model development sarima model data collection performance metrics calculation lstm model (a) m e a n s p e e d m /s m e a n s p e e d m /s (b) m e a n s p e e d m /s (c) i. tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 293 issn: 2252-4940/© 2023. the author(s). published by cbiore 4.3. time series wind data analysis a time series is defined as a sequence of numerical values indexed in time, habitually with the same time step separating two successive observations. in this paper, the data is a univariate time-series data. after importing the data set from the recorded csv file, and to better apprehend the data recording for each studied month, the data is plotted as a time series with the date along the x-axis and the wind speed on the y-axis. the data visualization is showed in figure 4. time-series forecasting is the process of fitting a model to previous data and employing it to forecast future observations. decomposition is a time series analysis technique, it gives a systematic method of thinking about a time series forecasting issue, in terms of modeling complexity and in terms of how to effectively capture each of these components in a given model. a time series must be decomposed to systematic and nonsystematic components. systematic components are time series components with consistency or recurrence that can be characterized and modeled. non-systematic components are time series components that cannot be directly represented. time series have three systematic components: level, trend, and seasonality, with an additional non-systematic noise component. fig. 5. time series decomposition the average value is called level; the trend in the series is the increasing or decreasing value. seasonality is the recurring short-term cycle, while noise is the random fluctuation in time series (athiyarath et al., 2020). results obtained for each month are shown in figure 5. for each month, the plot on figure 4 shows clearly that the wind speed is unstable and extreme winds are rare. it is noted that the seasonal component is more significant on october and evolves slowly in time, because the data are from separate months of the year. however, for consecutive months these will be different seasonal patterns with more important seasonal components. in addition, random fluctuations are more frequent in october and appear to be approximately stable in size over time. according to the trend line, the data and the trend are on the same scale and there is no long-term evolution of the series. 5. simulations and results 5.1 wind speed forecasting based on sarima model 5.1.1 parameter’s selection and model fitting the objective of this subsection is to conduct a grid search to identify the best combination of parameters that generates the highest performance for the model. to accomplish this objective, time series predictive model sarima was used. the code's output indicates that respectively: sarima(1, 1, 1)x(0, 0, 1, 2) , sarima(1, 1, 1)x(0, 0, 1, 2), and sarima(1, 0, 1)x(0, 1, 1, 2) generate the lowest aic (the akaike information criteria ) value of 14 869,12 for march, 9 443,72 for july, and 12 864,92 for october. therefore, this optimal combination was adopted. after fitting seasonal arima models, it is important to run model diagnostics to ensure that the assumptions made by the model are correct. figure 6 illustrates the diagnostics produced by the proposed model for each month. the conducted study has shown that the model residuals are normally distributed. the correlogram reveals that the autocorrelation of time series residuals are near to zero. this is confirmed by the plot of the residuals over time, which shows no significant seasonality. the kernel density estimator (kde) tracks tightly the normal distribution n (0, 1); this demonstrates that the residuals are distributed regularly. from the q-q plot, it can be observed that in the middle of the plot, the distribution of residuals closely follows the linear trend of the data obtained from a standard distribution; however, the residuals distribution diverges at the extremities, which corresponds to the extreme values of the data. subsequently, the model generates a satisfactory fit that can be used to predict future values. 5.1.2 forecasting validation in order to evaluate the accuracy of the proposed forecasts, 25% of the time series data was used as a test set. the forecasted wind speed targeted by the present model is related to the periods between the 24th and the end of the months march, july and october 2018. the forecasts were generated at one-step forward, meaning that the forecasts at each date are compiled using the complete history up to that time. the graph in figure 7, compares the actual values to the predicted values. overall, the forecasted data tightly aligns with the actual values; this confirms that the proposed model produces reliable prediction results. this performance will be quantified using the errors metrics introduced previously. r e s id u a l s e a s o n a l t r e n d march 2018 r e s id u a l s e a s o n a l t r e n d july 2018 r e s id u a l s e a s o n a l t r e n d october 2018 i. tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 294 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 6. the model diagnostics: (a) march 2018, (b) july 2018 and (c) october 2018 fig. 7. sarima forecasting validation: (a) march 2018, (b) july 2018 and (c) october 2018 5.2 wind speed forecasting with the lstm network 5.2.1 experimental test setup for each month, the global dataset is split in training dataset and test dataset of 75% and 25% of data respectively. a walkforward validation forecasting model is used. each step of the dataset test is browsed one at a time. afterwards, the model is ready to make predictions. the new estimated value from the test data set is obtained and made available to predict the next time step. 5.2.2 . lstm data preparation to create a lstm model, the data must be processed. the time series data is therefore turned into a supervised learning model. data is separated into input (x) and output (y). the observation from the prior time step (t-1) is used as the input, while the observation from the current time step (t) is used as the output. the hyperbolic tangent function (𝑡𝑎𝑛ℎ), having an average range (-1, 1), is the default activation function for lstms. therefore, the training dataset was employed to generate the (a) (b) (c) (a) m e a n s p e e d m /s (b) m e a n s p e e d m /s m e a n s p e e d m /s (c) i. tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 295 issn: 2252-4940/© 2023. the author(s). published by cbiore scaling coefficients (min and max) and then used the results to scale the test dataset and the observations were transformed to get a certain scale: the scale was reversed in order to restore the values to the original scale while computing error score in order to interpret and compare the acquired results. 5.2.3 lstm model development the lstm layer requires that the inputs be arranged in a matrix with three following dimensions: samples, time steps, and features. samples are sets of data in rows, time steps are discrete periods inside an observation, and features are distinct measurements taken at the time of the observation. accordingly, each time step in the original sequence was treated as a distinct sample with a single time step and feature. the lstm layer specifies the projected number of observations to be received in each batch. the frequency of updating the weights per time step is determined by the batch size and the number of epochs. the number of neurons is the last characteristic to define the lstm layer, a number between 1 and 5 is sufficient. to forecast the wind speed at the following time step, the network requires a single neuron in the output layer. once the network is described, it must be compiled into an appropriate symbolic representation. adam optimization algorithm was applied to compile the network since it is a realistic approach with advantages over other stochastic optimization methods (chang et al., 2019) (zhou et al., 2019). after, the network is fitted to the training data, which is taken in a supervised learning format. as be needed to make one-step predictions, the batch size is fixed to 1, the best configuration found is: batch size: 1, number of epochs: 50, number of neurons: 4. the number of epochs specifies how many times the learning algorithm will work on the training data set. the result will be a two-dimensional array with a single value because they are supplied with a single input. at this point, the lstm network model for the wind speed dataset can be submitted for performance evaluation. table 2 shows an example of the expected and predicted values in the test dataset for each month studied. figure 8 illustrates a line graph of the test values versus the predicted values. as observed, the forecasted data firmly aligns with the observed values, this, further confirms the model’s forecasting capability. table 2 the expected values versus predicted values march july october h=10, pred= 8.523, expec=11.068 h=11, pred=11.109, expec=11.534 h=12, pred=11.545, expec=10.359 h=13, pred=10.248, expec=10.107 h=14, pred= 9.981, expec=12.074 h=15, pred=12.072, expec=11.480 h=10, pred=4.226, expec=5.700 h=11, pred=5.820, expec=6.169 h=12, pred=6.173, expec=5.354 h=13, pred=5.164, expec=7.654 h=14, pred=7.901, expec=8.754 h=15, pred=8.851, expec=7.180 h=10, pred=10.553, expec=13.057 h=11, pred=13.191, expec=12.621 h=12, pred=12.745, expec=11.590 h=13, pred=11.738, expec=11.424 h=14, pred=11.589, expec=11.612 h=15, pred=11.787, expec= 9.983 (a) (b) (c) fig. 8. expected values vs predicted values: (a) march 2018, (b) july 2018 and (c) october 2018 m e a n s p e e d m /s m e a n s p e e d m /s m e a n s p e e d m /s i.tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 296 issn: 2252-4940/© 2023. the author(s). published by cbiore 5.2.4 reinforcement of the model the outputs of neural networks vary according to the initial conditions. for this specific problem, the experiment from the previous section were repeatedly conducted and the average of each error metric as a measure of the configuration performance envisaged on unknown data was used. in a loop with a set number of repetitions, the model's fit and walk-forward performance were validated. every iteration, the executions of error metrics are captured. next, a summary of the scores for all of the error metrics (mse, rmse, mae, and mape) is produced. table 3 shows the errors metrics results for 15 iterations for the three periods studied. the box and whisker graph of the distribution of test mape results for each of the 15 iterations is shown in figure 9. mape values obtained with different periods for the same number of iterations show significant variances. however, it is interesting to note that the model produced the lowest mape value in july when compared to the other months. looking at the three boxes medians, the median for march is closest to the minimum error, but for october, 50% of mape values are close to the maximum value. the lowest interquartile range (q3-q1) calculated, iqr = 0.060%, corresponds to the month of july, which indicates that the mape value distribution is the most homogeneous and the errors values are evenly distributed around the median. 5.3 results discussion the objective of this analysis section is to evaluate the numerical predictive models and determine which one performed better in terms of producing the lowest metrics error and minimal computation time. overall, both, lstm and sarima models offers good accuracy values of errors metrics comparing to results cited in the literature, for short term wind speed forecast (liu et al., 2021) (haddad et al., 2019) (duan et al., 2021). as shown in table 4, besides the improvement, lstm model still provides the highest predictive performance in terms of minimal metrics error in march and october the months with a high velocity and more extreme values. table 3 error metrics values resulting from 15 iterations march july october mse rmse mae mape mse rmse mae mape mse rmse mae mape minimum 1.1050 1.0449 0.7102 15.9400% 0.3947 0.6234 0.4544 10.5000% 1.5131 1.2298 0.9300 15.7000% q1 1.1177 1.0493 0.7147 16.0094% 0.4022 0.6270 0.4573 10.6704% 1.5240 1.2398 0.9350 15.7667% median 1.1203 1.0556 0.7164 16.0346% 0.4050 0.6310 0.4587 10.7006% 1.5295 1.2420 0.9370 15.8133% q3 1.1286 1.0567 0.7177 16.2518% 0.4070 0.6352 0.4599 10.7307% 1.5414 1.2442 0.9372 15.8309% maximum 1.1449 1.0586 0.7231 16.3737% 0.4089 0.6377 0.4889 10.9224% 1.5546 1.2464 0.9399 15.8926% mean 1.1238 1.0525 0.7168 16.1607% 0.4042 0.6313 0.4606 10.6982% 1.5318 1.2413 0.9360 15.8025% range 0.0399 0.0137 0.0129 0.4337% 0.0142 0.0143 0.0345 0.4224% 0.0415 0.0166 0.0099 0.1926% fig. 9. lstm repeated experiment box and whisker graph 15.69% 15.71% 15.73% 15.75% 15.77% 15.79% 15.81% 15.83% 15.85% 15.87% 15.89% mape -october10.48% 10.53% 10.58% 10.63% 10.68% 10.73% 10.78% 10.83% 10.88% 10.93% mape -july15.90% 15.95% 16.00% 16.05% 16.10% 16.15% 16.20% 16.25% 16.30% 16.35% 16.40% mape -marchi.tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 297 issn: 2252-4940/© 2023. the author(s). published by cbiore table 4 errors comparison the remarkable performance observed through lstm -based approaches are related to the used iterative optimization algorithm, with the goal of finding the best results, furthermore, taking a look at the rmse values, the statistical sarima model gives slightly better results on july, than the lstm model. therefore, in periods with small fluctuations, the neural network achieved comparable performances to the traditional statistical models. it should be mentioned that lstm would give more accurate results if the data contained structural changes with frequent fluctuation. in effect, the sarima method is unable to interpret the non-linear component of the wind speed time series data, thus, the model cannot capture all of the available details of the dataset. contrarily, lstm is a deep learning tool developed to learn temporal patterns, capture non-linear relations, and store relevant memory for a longer period. furthermore, the required computational time for forecasting becomes critical to make any comparison. the lstm based model is most computationally demanding than sarima, because of the optimization algorithm, 50 epochs were trained for each optimizer for 15 times and found that training and testing time ranges from 16 to 20 minutes, while the compilation time using sarima does not exceed 6 min. previous study has also supported this (shivani et al., 2019). indeed, authors in (makridakis et al., 2018), confirm that the complexity reported by ml approaches still substantially higher than statistical methods. the utility and advantage of a large data-learning model such as lstm, is more solicited when the size of data is much larger (liu et al., 2021). in this study, relatively small series data sets (one month) were used for predictive model training ant testing. otherwise, the performance differences favoring the lstm model would have been considerably more interesting. in previous academic research, the arima model provided favorable performance with a smaller volume of data (elsaraiti and merabet, 2021). 6. conclusion the present study is a comparative analysis of sarima and lstm forecasting models. the main goal is to evaluate the prediction performance of each one. the study uses wind speed time series recorded at three different months as common data. furthermore, sarima and lstm were compared based on the same evaluation metrics: mse, rmse, mae and mape. the mape values ranges from 10.5% to 16.10% for both models which indicates that performance is acceptable in both cases. the results analysis shows that the lstm exudes higher performance than the sarima model. this distinction is basically due to the use of iterative optimization algorithm in lstm model. practically speaking, the sarima model is more functional, because it requires the tuning of six unique parameters (p, q, d, p, q, d), while the lstm requires the evaluation of numerous additional hyper parameters, such as the number of units in each layer, the number of layers, the batch size, the number of epochs and the activation function. regarding the forecasting operation conducted in this study, related to wind speed in particular, sarima model was faster to train and less complicated to implement. as a perspective study, the obtained results can be supported by extending the comparative analysis. thus, in the future work, other machine learning prediction approaches are to be considered such as the support vector machine. the upcoming study will also use larger datasets that extend over a year, allowing us to assess the impact of seasons on the forecasting ability. references adekunle, s.a., (2017). prédiction de la moyenne horaire de la vitesse du vent sur le site de lomé par réseau de neurones. sci. appliquées ing. 2, 1–12. http://publication.lecames.org/index.php/ing/article/view/10 76 altan, a., karasu, s., zio, e., (2021). a new hybrid model for wind speed forecasting combining long short-term memory neural network, decomposition methods and grey wolf optimizer. appl. soft comput. 100, 106996. https://doi.org/10.1016/j.asoc.2020.106996 araya, i.a., valle, c., allende, h., (2020). a multi-scale model based on the long short-term memory for day ahead hourly wind speed forecasting. pattern recognit. lett. 136, 333–340. https://doi.org/10.1016/j.patrec.2019.10.011 asari, m., nanahara, t., maejima, t., yamaguchi, k., sato, t., (2002). a study on smoothing effect on output fluctuation of distributed wind power generation, in: ieee/pes transmission and distribution conference and exhibition. presented at the ieee/pes transmission and distribution conference and exhibition, pp. 938–943 vol.2. https://doi.org/10.1109/tdc.2002.1177602 athiyarath, s., paul, m., krishnaswamy, s., (2020). a comparative study and analysis of time series forecasting techniques. sn comput. sci. 1, 175. https://doi.org/10.1007/s42979-020-00180-5 babazadeh, h., gao, w., cheng, l., lin, j., (2012). an hour ahead wind speed prediction by kalman filter, in: 2012 ieee power electronics and machines in wind applications. presented at the 2012 ieee power electronics and machines in wind applications, pp. 1–6. https://doi.org/10.1109/pemwa.2012.6316394 bennitt, g.v., schueler, t., (2012) an assessment of zenith total delay corrections from numerical weather prediction models 1. geophysical research abstracts 14, egu2012-11292, https://meetingorganizer.copernicus.org/egu2012/egu201211292.pdf bessac, j., constantinescu, e., anitescu, m., (2018). stochastic simulation of predictive space–time scenarios of wind speed using observations and physical model outputs. ann. appl. stat. 12. https://doi.org/10.1214/17-aoas1099 botchkarev, a., (2019). a new typology design of performance metrics to measure errors in machine learning regression algorithms. interdiscip. j. inf. knowl. manag. 14, 045–076. https://doi.org/10.28945/4184 bououden, s., chadli, m., filali, s., el hajjaji, a., (2012). fuzzy model based multivariable predictive control of a variable speed wind turbine: lmi approach. renew. energy 37, 434–439. https://doi.org/10.1016/j.renene.2011.06.025 brereton, r.g., lloyd, g.r., (2010). support vector machines for classification and regression. analyst 135, 230–267. https://doi.org/10.1039/b918972f cassola, f., burlando, m., (2012). wind speed and wind energy forecast through kalman filtering of numerical weather prediction model output. appl. energy 99, 154–166. https://doi.org/10.1016/j.apenergy.2012.03.054 chang, z., zhang, y., chen, w., (2019). electricity price prediction based on hybrid model of adam optimized lstm neural network and wavelet transform. energy 187, 115804. https://doi.org/10.1016/j.energy.2019.07.134 damousis, i.g., alexiadis, m.c., theocharis, j.b., dokopoulos, p.s., (2004). a fuzzy model for wind speed prediction and power generation in wind parks using spatial correlation. ieee trans. march july october sarima lstm sarima lstm sarima lstm mse 1.19 1.10 0.38 0.39 1.54 1.51 rmse 1.09 1.04 0.62 0.62 1.24 1.23 mae 0.72 0.71 0.44 0.45 0.93 0.93 mape 16.10% 15.94% 10.67% 10.50% 16.03% 15.70% http://publication.lecames.org/index.php/ing/article/view/1076 http://publication.lecames.org/index.php/ing/article/view/1076 https://doi.org/10.1016/j.asoc.2020.106996 https://doi.org/10.1016/j.patrec.2019.10.011 https://doi.org/10.1109/tdc.2002.1177602 https://doi.org/10.1007/s42979-020-00180-5 https://doi.org/10.1109/pemwa.2012.6316394 https://meetingorganizer.copernicus.org/egu2012/egu2012-11292.pdf https://meetingorganizer.copernicus.org/egu2012/egu2012-11292.pdf https://doi.org/10.1214/17-aoas1099 https://doi.org/10.28945/4184 https://doi.org/10.1016/j.renene.2011.06.025 https://doi.org/10.1039/b918972f https://doi.org/10.1016/j.apenergy.2012.03.054 https://doi.org/10.1016/j.energy.2019.07.134 i. tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 298 issn: 2252-4940/© 2023. the author(s). published by cbiore energy convers. 19, 352–361. https://doi.org/10.1109/tec.2003.821865 devis, a., lipzig, n.p.m.v., demuzere, m., (2018). should future wind speed changes be taken into account in wind farm development? environ. res. lett. 13, 064012. https://doi.org/10.1088/17489326/aabff7 duan, jikai, zuo, h., bai, y., duan, jizheng, chang, m., chen, b., (2021). short-term wind speed forecasting using recurrent neural networks with error correction. energy 217, 119397. https://doi.org/10.1016/j.energy.2020.119397 eldali, f.a., hansen, t.m., suryanarayanan, s., chong, e.k.p., (2016). employing arima models to improve wind power forecasts: a case study in ercot, in: 2016 north american power symposium (naps). presented at the 2016 north american power symposium (naps), pp. 1–6. https://doi.org/10.1109/naps.2016.7747861 elsaraiti, m., merabet, a., (2021). a comparative analysis of the arima and lstm predictive models and their effectiveness for predicting wind speed. energies 14, 6782. https://doi.org/10.3390/en14206782 erdem, e., shi, j., (2011). arma based approaches for forecasting the tuple of wind speed and direction. appl. energy 88, 1405–1414. https://doi.org/10.1016/j.apenergy.2010.10.031 faniband, y.p., shaahid, s.m., (2020). forecasting wind speed using artificial neural networks – a case study of a potential location of saudi arabia. e3s web conf. 173, 01004. https://doi.org/10.1051/e3sconf/202017301004 farida, m., zeghdoudi, h., (2020). on modelling seasonal arima series: comparison, application and forecast (number of injured in road accidents in northeast algeria). wseas trans. syst. control 15, 235–246. https://doi.org/10.37394/23203.2020.15.25 geng, d., zhang, h., wu, h., (2020). short-term wind speed prediction based on principal component analysis and lstm. appl. sci. 10, 4416. https://doi.org/10.3390/app10134416 haddad, m., nicod, j., boubacar mainassara, y., rabehasaina, l., al masry, z., péra, m., (2019). wind and solar forecasting for renewable energy system using sarima-based model, in: international conference on time series and forecasting. gran canaria, spain. hajirahimi, z., khashei, m., (2019). hybrid structures in time series modeling and forecasting: a review. eng. appl. artif. intell. 86, 83– 106. https://doi.org/10.1016/j.engappai.2019.08.018 hide, c., moore, t., smith, m., (2003). adaptive kalman filtering for low-cost ins/gps. j. navig. 56, 143–152. https://doi.org/10.1017/s0373463302002151 hossin, m. and sulaiman, m.n. (2015). a review on evaluation metrics for data classification evaluations. int. j. data min. knowl. manag. process 5, 01–11. https://doi.org/10.5121/ijdkp.2015.5201 jaseena, k.u., kovoor, b.c., (2020). deterministic weather forecasting models based on intelligent predictors: a survey. j. king saud univ. comput. inf. sci. s1319157820304729. https://doi.org/10.1016/j.jksuci.2020.09.009 kamble, v.b., deshmukh, s.n., (2017). comparision between accuracy and mse,rmse by using proposed method with imputation technique. orient. j. comput. sci. technol. 10, 773–779. https://doi.org/10.13005/ojcst/10.04.11 kim, s., kim, h., (2016). a new metric of absolute percentage error for intermittent demand forecasts. int. j. forecast. 32, 669–679. https://doi.org/10.1016/j.ijforecast.2015.12.003 kodjo, m.k., bédja, k., ajavon, a.s.a., faye, r.m., lishou, c., (2008). neural networks for predictive control of the mechanism of orientation of a wind turbine. j. sci. pour ing. 9, 75–85. https://doi.org/10.4314/jspi.v9i1.30061 li, g., shi, j., zhou, j., (2011). bayesian adaptive combination of shortterm wind speed forecasts from neural network models. renew. energy 36, 352–359. https://doi.org/10.1016/j.renene.2010.06.049 liu, x., lin, z., feng, z., (2021). short-term offshore wind speed forecast by seasonal arima a comparison against gru and lstm. energy 227, 120492. https://doi.org/10.1016/j.energy.2021.120492 lorenc, a.c., (1986). analysis methods for numerical weather prediction. q. j. r. meteorol. soc. 112, 1177–1194. https://doi.org/10.1002/qj.49711247414 makridakis, s., spiliotis, e., assimakopoulos, v., (2018). statistical and machine learning forecasting methods: concerns and ways forward. plos one 13, e0194889. https://doi.org/10.1371/journal.pone.0194889 mantalos, p., mattheou, k., karagrigoriou, a., (2010). forecasting arma models: a comparative study of information criteria focusing on mdic. j. stat. comput. simul. 80, 61–73. https://doi.org/10.1080/00949650802464137 martinez-garcía, f.p., contreras-de-villar, a., muñoz-perez, j.j., (2021). review of wind models at a local scale: advantages and disadvantages. j. mar. sci. eng. 9, 318. https://doi.org/10.3390/jmse9030318 mi, x., liu, h., li, y., (2019). wind speed prediction model using singular spectrum analysis, empirical mode decomposition and convolutional support vector machine. energy convers. manag. 180, 196–205. https://doi.org/10.1016/j.enconman.2018.11.006 miranda, m.s., dunn, r.w., (2006). one-hour-ahead wind speed prediction using a bayesian methodology, in: 2006 ieee power engineering society general meeting. presented at the 2006 ieee power engineering society general meeting, p. 6 pp.-. https://doi.org/10.1109/pes.2006.1709479 nair, k.r., vanitha, v., jisma, m., (2017). forecasting of wind speed using ann, arima and hybrid models, in: 2017 international conference on intelligent computing, instrumentation and control technologies (icicict). presented at the 2017 international conference on intelligent computing, instrumentation and control technologies (icicict), ieee, kerala state,kannur, india, pp. 170–175. https://doi.org/10.1109/icicict1.2017.8342555 navas, r.k.b., prakash, s., sasipraba, t., 2020. artificial neural network based computing model for wind speed prediction: a case study of coimbatore, tamil nadu, india. phys. stat. mech. its appl. 542, 123383. https://doi.org/10.1016/j.physa.2019.123383 nazir, m.s., alturise, f., alshmrany, s., nazir, h.m.j., bilal, m., abdalla, a.n., sanjeevikumar, p., m. ali, z., (2020). wind generation forecasting methods and proliferation of artificial neural network: a review of five years research trend. sustainability 12, 3778. https://doi.org/10.3390/su12093778 nguyen, l., novák, v., (2019). forecasting seasonal time series based on fuzzy techniques. fuzzy sets syst. 361, 114–129. https://doi.org/10.1016/j.fss.2018.09.010 pinto, t., ramos, s., sousa, t.m., vale, z., (2014). short-term wind speed forecasting using support vector machines, in: 2014 ieee symposium on computational intelligence in dynamic and uncertain environments (cidue). presented at the 2014 ieee symposium on computational intelligence in dynamic and uncertain environments (cidue), pp. 40–46. https://doi.org/10.1109/cidue.2014.7007865 qian, z., pei, y., zareipour, h., chen, n., (2019). a review and discussion of decomposition-based hybrid models for wind energy forecasting applications. appl. energy 235, 939–953. https://doi.org/10.1016/j.apenergy.2018.10.080 ranganayaki, v., deepa, s.n., (2017). svm based neuro fuzzy model for short term wind power forecasting. natl. acad. sci. lett. 40, 131–134. https://doi.org/10.1007/s40009-016-0521-6 shivani, sandhu, k.s., ramachandran nair, a., (2019). a comparative study of arima and rnn for short term wind speed forecasting, in: 2019 10th international conference on computing, communication and networking technologies (icccnt). presented at the 2019 10th international conference on computing, communication and networking technologies (icccnt), ieee, kanpur, india, pp. 1–7. https://doi.org/10.1109/icccnt45670.2019.8944466 siami-namini, s., tavakoli, n., siami namin, a., (2018). a comparison of arima and lstm in forecasting time series, in: 2018 17th ieee international conference on machine learning and applications (icmla). presented at the 2018 17th ieee international conference on machine learning and applications (icmla), ieee, orlando, fl, pp. 1394–1401. https://doi.org/10.1109/icmla.2018.00227 soman, s.s., zareipour, h., malik, o., mandal, p., (2010). a review of wind power and wind speed forecasting methods with different time horizons, in: north american power symposium 2010. https://doi.org/10.1109/tec.2003.821865 https://doi.org/10.1088/1748-9326/aabff7 https://doi.org/10.1088/1748-9326/aabff7 https://doi.org/10.1016/j.energy.2020.119397 https://doi.org/10.1109/naps.2016.7747861 https://doi.org/10.3390/en14206782 https://doi.org/10.1016/j.apenergy.2010.10.031 https://doi.org/10.1051/e3sconf/202017301004 https://doi.org/10.37394/23203.2020.15.25 https://doi.org/10.3390/app10134416 https://doi.org/10.1016/j.engappai.2019.08.018 https://doi.org/10.1017/s0373463302002151 https://doi.org/10.5121/ijdkp.2015.5201 https://doi.org/10.1016/j.jksuci.2020.09.009 https://doi.org/10.13005/ojcst/10.04.11 https://doi.org/10.1016/j.ijforecast.2015.12.003 https://doi.org/10.4314/jspi.v9i1.30061 https://doi.org/10.1016/j.renene.2010.06.049 https://doi.org/10.1016/j.energy.2021.120492 https://doi.org/10.1002/qj.49711247414 https://doi.org/10.1371/journal.pone.0194889 https://doi.org/10.1080/00949650802464137 https://doi.org/10.3390/jmse9030318 https://doi.org/10.1016/j.enconman.2018.11.006 https://doi.org/10.1109/pes.2006.1709479 https://doi.org/10.1109/icicict1.2017.8342555 https://doi.org/10.1016/j.physa.2019.123383 https://doi.org/10.3390/su12093778 https://doi.org/10.1016/j.fss.2018.09.010 https://doi.org/10.1109/cidue.2014.7007865 https://doi.org/10.1016/j.apenergy.2018.10.080 https://doi.org/10.1007/s40009-016-0521-6 https://doi.org/10.1109/icccnt45670.2019.8944466 https://doi.org/10.1109/icmla.2018.00227 i. tyass et al int. j. renew. energy dev 2023, 12(2), 288-299 | 299 issn: 2252-4940/© 2023. the author(s). published by cbiore presented at the north american power symposium 2010, pp. 1– 8. https://doi.org/10.1109/naps.2010.5619586 tascikaraoglu, a., uzunoglu, m., (2014). a review of combined approaches for prediction of short-term wind speed and power. renew. sustain. energy rev. 34, 243–254. https://doi.org/10.1016/j.rser.2014.03.033 tayman, j., swanson, d.a., 1999. on the validity of mape as a measure of population forecast accuracy. popul. res. policy rev. 18, 299– 322. https://doi.org/10.1023/a:1006166418051 tian, y., xu, y.-p., yang, z., wang, g., zhu, q., 2018. integration of a parsimonious hydrological model with recurrent neural networks for improved streamflow forecasting. water 10, 1655. https://doi.org/10.3390/w10111655 tokgöz, a., ünal, g., 2018. a rnn based time series approach for forecasting turkish electricity load, in: 2018 26th signal processing and communications applications conference (siu). presented at the 2018 26th signal processing and communications applications conference (siu), pp. 1–4. https://doi.org/10.1109/siu.2018.8404313 torres, j.l., garcía, a., de blas, m., de francisco, a., 2005. forecast of hourly average wind speed with arma models in navarre (spain). sol. energy 79, 65–77. https://doi.org/10.1016/j.solener.2004.09.013 wu, w., shaikhouni, a., donoghue, j.r., black, m.j., 2004. closed-loop neural control of cursor motion using a kalman filter, in: the 26th annual international conference of the ieee engineering in medicine and biology society. presented at the the 26th annual international conference of the ieee engineering in medicine and biology society, pp. 4126–4129. https://doi.org/10.1109/iembs.2004.1404151 yang, j., astitha, m., monache, l.d., alessandrini, s., 2018. an analog technique to improve storm wind speed prediction using a dual nwp model approach. mon. weather rev. 146, 4057–4077. https://doi.org/10.1175/mwr-d-17-0198.1 yatiyana, e., rajakaruna, s., ghosh, a., 2017. wind speed and direction forecasting for wind power generation using arima model, in: 2017 australasian universities power engineering conference (aupec). presented at the 2017 australasian universities power engineering conference (aupec), pp. 1–6. https://doi.org/10.1109/aupec.2017.8282494 zhang, c., wei, h., xie, l., shen, y., zhang, k., 2016. direct interval forecasting of wind speed using radial basis function neural networks in a multi-objective optimization framework. neurocomputing 205, 53–63. https://doi.org/10.1016/j.neucom.2016.03.061 zhang, w., zhang, l., wang, j., niu, x., 2020. hybrid system based on a multi-objective optimization and kernel approximation for multi-scale wind speed forecasting. appl. energy 277, 115561. https://doi.org/10.1016/j.apenergy.2020.115561 zhou, b., ma, x., luo, y., yang, d., 2019. wind power prediction based on lstm networks and nonparametric kernel density estimation. ieee access 7, 165279–165292. https://doi.org/10.1109/access.2019.2952555 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1109/naps.2010.5619586 https://doi.org/10.1016/j.rser.2014.03.033 https://doi.org/10.1023/a:1006166418051 https://doi.org/10.3390/w10111655 https://doi.org/10.1109/siu.2018.8404313 https://doi.org/10.1016/j.solener.2004.09.013 https://doi.org/10.1109/iembs.2004.1404151 https://doi.org/10.1175/mwr-d-17-0198.1 https://doi.org/10.1109/aupec.2017.8282494 https://doi.org/10.1016/j.neucom.2016.03.061 https://doi.org/10.1016/j.apenergy.2020.115561 https://doi.org/10.1109/access.2019.2952555 international journal of renewable energy development int. j. renew. energy dev. 2023, 12(2), 313-327 | 313 https://doi.org/10.14710/ijred.2023.47686 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id theoretical and experimental study on the performance of photovoltaic using porous media cooling under indoor condition ismail al-masalha1,2* , siti ujila binti masuri1 , omar badran2 , mohd khairol anuar bin mohd ariffin1 , abd. rahim abu talib3 , fadi alfaqs2 1department of mechanical and manufacturing engineering, faculty of engineering, universiti putra malaysia, upm serdang, selangor, malaysia 2mechanical engineering department, faculty of engineering technology, al–balqa applied university, jordan. 3department of aerospace engineering, faculty of engineering, universiti putra malaysia, malaysia abstract. this paper presents the theoretical and experimental investigation on performance of a photovoltaic (pv) panel cooled by porous media under indoor condition. porous media offer a large exterior surface area and a high fluid permeability, making them ideal for pv cells cooling. the photovoltaic panel was cooled using 5 cm thick cooling channel filled with porous media (gravel). several sizes of porosity (0.35, 0.4, 0.48, and 0.5) at different volume flow rates (1, 1.5, 2, 3, and 4 l/min) were tested to obtain the best cooling process. the theoretical analysis was performed at the optimum case found experimentally, which has a porosity of 0.35 and a volume flow rate of 2 l/min, to test various experimental results of the pv hot surface temperature, related power output, efficiency and i-v characteristic curve. the enhancement obtained in pv power output and efficiency is compared against the case without cooling and the case using water alone without porous media. results showed that cooling using small size porous media and moderate flow rate is more efficient which reduces the average pv hot surface temperature of about 55.87% and increases the efficiency by 2.13% than uncooled pv. the optimum case reduced the pv hot surface temperature to 38.7°c, and increased the power output to 19 w, efficiency to 6.26%, and the open voltage to 22.77 v. the results showed that the presence of small porous media of 0.35 in the pv cooling process displayed the maximum effectiveness compared to the other two scenarios, because the heat loss from pv surface through porous media layer have developed a homogenous heat diffusion removed much quicker at high flow rate (2 l/min). a good agreement was obtained between experimental and theoretical results for different cases with a standard deviation from 3.2% to 5.6%. keywords: cooling of pv panel; solar cell efficiency; porous media; operating temperature; indoor test. @ author (s). published by cbiore. this is an open-access article under the cc by-sa license. (http://creativecommons.org/licenses/by-sa/4.0/) received: 17th july 2022; revised: 28th nov 2022; accepted: 19th january 2023; available online: 5th feb 2023 1. introduction fossil fuels including oil, gas, and coal are the primary energy sources in the majority of countries. as a result, the globe confronts a major dilemma today because of the tremendous consumption that will occur in the next several years. for this reason, jordan's economy is in freefall. jordan imports most of its fossil fuels, and this drives up the cost of generating power (badran et al. 2010; elayyan et al.2020). with the country's current circumstances, thinking toward renewable energy applications have been considered as a main solution, and as a result, jordan has some of the world's best renewable energy resources (masalha et al. 2017) particularly in the area of solar energy, where daily solar radiation averages 5-8 hours (hammad, et al. 2015). solar energy is a clean, renewable, safe, and cost-effective way to generate electricity. solar cells generate electricity, but one of the challenges is that when cell temperature rises, it affects the efficiency and this is a major issue. solar cells can be more efficient by adding a cooling * corresponding author: email: masalha55@yahoo.com (i. al-masalha) system in order to keep them at the proper operating temperature while still producing maximum power. the performance and efficiency of pv systems have been examined by many researchers, and a variety of cooling strategies have been discovered. among these studies, the work which was carried out by ahmed, et al. (2019) attempted to improve the performance of a hybrid photovoltaic/trombe wall (pv/tw) system through employing a porous medium, dc fan and glass cover. the porous medium with dc fan increased the values of thermal efficiency by 13% and electrical efficiency by 4%, while the combination of porous medium with dc fan and glass cover increased the thermal efficiency by 20% and electrical efficiency by 0.5%. özakina et al. (2019) studied fins inside a channel experimentally and numerically. they observed an increase in power; electrical efficiency and thermal efficiency. soliman et al. (2019) investigated the performance of concentrated photovoltaic cells cooled by spreadermicrochannels system. three configurations for water cooling flowing inside the microchannel were tested. their results showed an increase in the pv net power by about 9.2%. research article https://doi.org/10.14710/ijred.2023.47686 https://doi.org/10.14710/ijred.2023.47686 http://creativecommons.org/licenses/by-sa/4.0/ mailto:masalha55@yahoo.com https://orcid.org/0000-0003-3796-8384 https://orcid.org/0000-0001-8170-8788 https://orcid.org/0000-0002-6984-5015 https://orcid.org/0000-0002-5390-8202 https://orcid.org/0000-0003-4021-4460 https://orcid.org/0000-0003-3427-6454 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.47686&domain=pdf i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 314 issn: 2252-4940/©2023. the author(s). published by cbiore kabeel et al. (2019) conducted a comparative study to improve three pv unit cooling technologies under egyptian conditions. these technologies use air and water with the presence of reflectors. the results showed that the best way to cool the pv unit was to use water with the presence of reflectors. elminshawy et al. (2019) investigated an integrated pv panel cooling system with buried heat exchanger (bhe) experimentally. forced atmospheric air with temperatures of 35°c, 40°c and 45°c and flow rates of 0.0228 m3/s, 0.0248 m3/s, and 0.0268 m3/s were used. it was found that the output power and electrical efficiency were improved. elminshawy et al. (2019) performed experimental and numerical investigations using a buried water heat exchanger. they noticed an increase in the output power, electrical efficiency, and thermal efficiency. issa et al. (2018) investigated the effects of pure water, sio2/water nano fluid, and phase-change material (pcm) coolants on the performance of a photovoltaic thermal (pvt) system numerically. their simulations were performed on pvt/pcm module and a pcm module without pvt. also, they compared nanofluid with pure water as a working fluid. using the water-based pvt/pcm, the average pv cell temperature was decreased by 16°c compared with the pvt system, while the electrical efficiency increased by 8%, and thermal efficiency by 25%. on the other hand, using nanofluid (sio2 with 1 and 3% mass fraction) as a coolant caused an increase in the thermal efficiency by 3.51% and 10.40%, respectively, compared to the pvt/pcm with coolant. alizadeh et al. (2018) used plastic heat pipes as well as copper fins, in order to reduce the pv panels’ temperature; they noticed an increase in electrical efficiency. golzari et al. (2018) conducted an experimental work to enhance the temperature reduction by electro hydrodynamics (ehd) process. heat transfer coefficient was increased by halo wind due to a secondary flow and vortex. therefore, an increase in voltage leads to an increase in electrical energy and efficiency. idoko et al. (2018) used the multi-concept cooling technique, involving three types of passive cooling: conductive cooling, air passive cooling, and water passive cooling. two solar panels of 250 w were mounted at a height of 37 cm to create room for air-cooling. one was attached to an aluminum heat sink and iced water-cooled lead to surface temperature reduction of 20°c. an increase of 20.96 w in output power was also noticed for the iced water-cooling, and an increase in efficiency by 3% were recorded. jamali et al. (2018) proposed a novel system to cool semi-transparent photovoltaic (stpv) system by employing it as a roof for a solar chimney. they found that the system could reduce the average temperature of the stpv to 15°c and power generation enhanced by 29%. al-jamea et al. (2022) studied pv using water-cooling channel numerically. they investigated the temperature of the pv cell, inlet and outlet water, radiation, increase in efficiency, and electrical power. mojumder et al. (2016) studied and designed a prototype of pv collector with cooling fins and induced air as a working fluid. their results showed an increase in thermal efficiency by 56.19% and they validated their experimental results using theoretical and statistical analysis techniques. aldossary et al. (2016) used a method based on two heat sinks cooling system with water passing between these two sinks. they realized that the temperature was reduced by 14oc, while the electrical efficiency increased to 39.5%. bahaidarah et al. (2016) used contrasting techniques of cooling (non-uniform and uniform) using immersion cooling; the temperature of the concentrated pv system (cpv) cell has been decreased to 20– 45°c while maintaining uniform cooling. when using heat sinks to passively cool cells, they were able to lower the temperature by as much as 37°c at the price of increasing the heat sink's surface area. stropnik et al. (2016) showed the cooling process was accomplished by the use of pcm (phase change material). the temperature of the pv panel was reduced to 35.6oc, while output power rose to 8.7% and efficiency improved by 1% in comparison to standard pv panels, according to their testing data. bai et al. (2016) focused on the impact of sprinkling and refrigerant on cooling methods under hungary climate conditions. their pv panel’s temperature was decreased down to around 29 °c while the efficiency increased by 25%. amelia et al. (2016) installed cooling mechanisms of 4 dc (direct current) fans at pv panel. the average temperature was decreased by 22.22%, and the power output increased from first fan to the fourth fan by 12.93, 37.17, 41.28, and 44.34% respectively. radwan et al. (2016) employed a new method represented by a microchannel heat sink with nanofluids, nanofluids such as aluminum oxide (al2o3)–water and silicon carbide (sic), and water were used as a coolant. the electrical efficiency was a direct proportionality to nanoparticles volumes. ebrahimi et al. (2015) created a novel approach to cool pv cell by using natural vapor as coolant depending on the daylight on backward of the pv cells. it was found that their method decreased the temperature to about 16oc and increased the electrical efficiency to about 22.9%. nižetić et al. (2015) studied one of most common ways of cooling pv panels’ surfaces by using spray water, which gave good results through increasing the electrical efficiency and output power. micheli et al. (2015) studied pv cooling system accomplished by using a micro fin. it caused a decrease in the pv surface temperature and enhanced the performance by increasing the output power up to 50%. elnozahy et al. (2015) investigated cooling pv panels by employing an automatic cooling that controls the backside pv surface temperature to be close to the ambient temperature. they realized that the results showed a decrease in temperature by about 45.5% and an increase in efficiency to 11.7%. irwan et al. (2015) studied ways to enhance the work of pv panels based on the enhancing the electrical efficiency. to increase the electrical efficiency a halogen lamps as an extra daylight were used. the results revealed a decrease in temperature and an increase in the output power. rahimi et al. (2015) suggested the best way to cool pv panels by using a multi-header microchannel design. their result showed that 28% enhancement in electrical energy was obtained, while the temperature decreased by 19%. hussien et al. (2016) used the water as a coolant for the pv cells using it in a heat exchanger. water circulating in pipes located at the rear of panels where the heat loss by convection occurred between the water and panels. they noticed that the electrical and thermal efficiency were increased by 9.8% and 12.3%, respectively. chandrasekar et al. (2013) studied new way to cool pv systems under indian climate conditions by applying moist cotton wicks to pv panels for cooling. results showed a reduction in temperature by 12% and an increase of 14% in electrical efficiency. tao ma et al. (2015) used a pcm with solar thermal system to cool a pv cell. they realized that the electrical efficiency had increased by 9% and the thermal efficiency by 12%. sahay et al. (2015) studied a low-cost way of pv cooling system in india called ground central panel cooling system (gcpcs). smoking flow visualization technique was used and analyzed with anova to show the cooling effect on pv performance but no specific efficiency enhancement of pv was obtained. buker et al. (2015) used liquid desiccant-based dew point cooling to provide air to use it in heat exchanger as a heat source. their tests showed that power performance was increased by 10.7%. from the previous studies it can be seen that different methods were used for cooling the pv back surface to increase its power and efficiency but each method has its advantages and i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 315 issn: 2252-4940/©2023. the author(s). published by cbiore drawbacks. one of the drawbacks of some cooling techniques like spraying, they require lots of water and some techniques require huge and expensive facilities so they gain cooling effectiveness at high cost. moreover, one of the good cooling techniques uses nanofluid, but it is expensive. therefore, the reduction of the pv back surface temperature is the main target for many cooling processes, such that cooling using porous materials can be effectively better than cooling with air. it can be seen from the literature review that porous media was considered as one of better cooling strategies that can accelerate the cooling process to enhance the pv performance. also, previous studies used porous materials for cooling solar cells showed that there are needs for more practical and enumerative work that should be done in a comprehensive way. the objective of the present study is to find out an effective heat transfer condition using porous media to reduce the pv/t system's temperature and to increase the pv power output. different porosities were considered to find out the best cooling condition. also, different scenarios were compared with water alone cooling and how the cooling behaves at different volume flow rates (i.e. 1, 1.5, 2, 3, and 4 l/min) to improve the performance of the pv cell. these studies used gravels as porous media of different sizes that is globally available and have good conductivity. 2. methodology solar energy is one of the cheapest technologies for generating power in developing and developed nations. however, as previously mentioned, one of the primary shortcomings of photovoltaic cells is the influence of heated surface temperature, which leads to poorer power generation, particularly in hot climates such as in jordan during the summer. by absorbing the produced heat, the efficiency of the pv cells would be enhanced. in general, porous media was recognized as one of the most essential technologies employed for homogeneous cooling. the influence of porosity on the performance of a pv/t system will be explored experimentally and theoretically in this study. the pv cooling system shown in fig. 1 was used to produce electrical power as well as absorbing the ejected heat from pv back surface to flowing water. where lg is glass thickness, lc is solar cell thickness, lt is back surface of tedlar thickness and lin is insulation thickness. to determine the best possible performance from photovoltaic panel, various experiments were conducted with several cooling processes. a 5 cm thick cooling channel was attached on the back of the pv hot surface exposed to fixed indoor radiation intensity of 1600 w/m2 for all experiments (fig.1). in the channel, porous material with different porosities of 0.35, 0.4, 0.48 and 0.5 was used along with different volume flow rates of 1 l/min, 1.5 l/min, 2 l/min, 3 l/min, and 4 l/min. later the analytical study was used to validate the experimental results. fig 1. pv cooling channel with porous media (a) pv cooling channel (b) pv cooling channel porous media (gravel) i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 316 issn: 2252-4940/©2023. the author(s). published by cbiore 3. mathematical model formulation in order to investigate the effect of different elements on the performance of a pv/t hybrid system cooled by porous media, the electrical and thermal models can be modelled and analysed. 3.1. modeling of thermal system a mathematical model was created based on fig. 2 to calculate the hot surface temperature on the back of solar cells. the following assumptions were assumed for the model: (i) the system operates in a steady flow condition; (ii) there is no water leakage in the system because it was well sealed; (iii) the walls of the cooling channel insulated on all sides, except the upper surface on which the pv panel attached, so that there is no heat transfer to the surrounding and (iv) the change in the temperature is only along the length of the pv panel. 3.2. energy balance analysis of the pv cell photovoltaic cells (pv) do not convert all entire solar intensity received on its surface to the electrical power, because some of energy received is converted into heat, then the rear temperature of the pv cell can be calculated by eq. 1 (abu bakar et al. 2015) 𝐺. 𝛼𝑐 = 𝐸 + 𝑈𝑓𝑟𝑜𝑛𝑡 (𝑇𝑐 − 𝑇𝑎 ) + 𝑈𝑏𝑜𝑡𝑡𝑜𝑚 (𝑇𝑐 − 𝑇𝑇 ) (1) where tt is the pv cell's back surface temperature, the tc cell's surface temperature, the ta ambient temperature, g denotes solar radiation, e denotes electrical power, and 𝛼𝑐 denotes absorptivity. after rearranging eq. (1), the temperature of the pv cells can be calculated using eq. (2) as (eq 2); 𝑇𝑐 = 𝐺.𝛼𝑐 − 𝐸 + 𝑈𝑓𝑟𝑜𝑛𝑡𝑇𝑎 +𝑈𝑏𝑜𝑡𝑡𝑜𝑚 𝑇𝑇 𝑈𝑓𝑟𝑜𝑛𝑡+𝑈𝑏𝑜𝑡𝑡𝑜𝑚 (2) fig 2. thermal resistance circuit diagram ufront is the coefficient of heat transfer between the pv cells and the surrounding environment, and can be calculated by eq 3: 𝑈𝑓𝑟𝑜𝑛𝑡 = [ 1 ℎ𝑐+ℎ𝑟 + 𝐿𝑔 𝑘𝑔 ] −1 (3) where k denotes thermal conductivity, ℎ𝑐 is convective heat transfer coefficient, ℎ𝑟 is radiant heat transfer coefficient, the values of the coefficients can be found from eq. (4) and (5) respectively (irshad et al. 2014) ℎ𝑐 = 5.7 + 3.8v (4) where v is the velocity of the surrounding air around the pv cells. ℎ𝑟 = 𝜀 𝜎(𝑇𝑐 2 + 𝑇𝑎 2)(𝑇𝑐 + 𝑇𝑎 ) (5) emissivity (𝜀) and the stefan-boltzmann constant (𝜎). 𝑈𝑏𝑜𝑡𝑡𝑜𝑚 = [ 𝐿𝑇 𝑘𝑇 ] −1 (6) where ubottom is the overall heat transfer coefficient for tedlar, 𝐿𝑇, and 𝑘𝑇 are the thickness and thermal conductivity of tedlar, respectively. the temperature of tedlar can be evaluated by eq 7 to eq 9 (ramadan et al. 2007) 𝐺. 𝛼𝑐 = 𝑈𝑓𝑟𝑜𝑛𝑡2(𝑇𝑇 − 𝑇𝑎 ) + ℎ𝑐,𝑓 (𝑇𝑇 − 𝑇𝑓 ) + ℎ𝑟,𝑝𝑚 (𝑇𝑇 − 𝑇𝑝𝑚) (7) 𝑇𝑇 = 𝑈𝑓𝑟𝑜𝑛𝑡2𝑇𝑎+ℎ𝑐,𝑓𝑇𝑓+ℎ𝑟,𝑝𝑚𝑇𝑝𝑚+ 𝐺.𝛼𝑐 ℎ𝑐,𝑓+ℎ𝑟,𝑝𝑚+𝑈𝑓𝑟𝑜𝑛𝑡2 (8) 𝑈𝑓𝑟𝑜𝑛𝑡2 = [ 1 ℎ𝑐+ℎ𝑟 + 𝐿𝑔 𝑘𝑔 + 𝐿𝑇 𝑘𝑇 ] −1 (9) where ufront2 is the overall heat transfer coefficient between tedlar and ambient, ℎ𝑐,𝑓 is the convective heat transfer coefficient for water flowing through channel filled with porous medium and the tedlar which can be calculated by eq. 10 to eq 12 (ramadan et al. 2007) ℎ𝑐,𝑓 = [ 1 𝑁𝑢𝑝𝑚−𝑓.𝑘𝑓 𝐷ℎ𝑑 + 𝐷ℎ𝑑 𝛽𝑘𝑔 ] −1 (10) 𝐷ℎ𝑑 = 2𝜑𝑑𝑝𝑚 3(1−𝜑) (11) 𝑁𝑢𝑝𝑚−𝑓 = ( 0.255 𝜑 ) 𝑅𝑒𝑝𝑚 2 3 . 𝑃𝑟 𝑓 1 3⁄ (12) where 𝐷ℎ𝑑 is hydraulic diameter, 𝑑𝑝𝑚is the particle diameter, 𝜑 is porosity of the porous medium, 𝑁𝑢𝑝𝑚−𝑓 is nusselt number, 𝑃𝑟𝑓 is prandtl number, 𝑅𝑒𝑝𝑚 is reynolds number,∀𝑓is void volume and ∀𝑡 is total volume and can be computed as (eq 13) (masalha et al 2023); 𝜑 = ∀𝑓 ∀𝑡 (13) the ℎ𝑟,𝑝𝑚 is the radiation heat transfer coefficient from tedlar to the porous medium, which can be calculated based on the equation 14 (ramadan et al. 2007). ℎ𝑟,𝑝𝑚 = ℇ𝜎(𝑇𝑇 2 + 𝑇𝑝𝑚 2)(𝑇𝑇 2 + 𝑇𝑝𝑚 2) (14) i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 317 issn: 2252-4940/©2023. the author(s). published by cbiore fig 3. demonstration of a pv cell circuit (bahaidarah et al. 2013) 3.3. modeling of the electrical system the pv electrical system has been designed based on the following assumptions; the average temperature is calculated by taking the temperature in each stratum and averaging it out. in addition, the water flow in the channel and on the tedlar was supposed to be consistent. both sun light and the temperature have no effect on the resistances. radiation is the only factor that affects shunt resistance, and temperature is irrelevant. all parameters were estimated using an electrical model (vopen, ishort, power, and electrical efficiency). a pv cell's equivalent circuit is shown in fig. 3. in order to describe a pv module, we must look at its current-voltage curve (i-v). the parameters of the electrical model were evaluated similar to the parameter model presented by (bahaidarah et al. 2013) an ideality factor (a), light generated current (il), series resistance (rs), and shunt resistance (rs) are the primary parameters of the pv electrical model. table 1 shows the typical conditions specified at three i-v points on the pv curve. according to the conditions mentioned in table 1, the following characteristics are derived: 𝐼 = 𝐼𝐿 − 𝐼𝑂 [𝑒𝑥𝑝 ( 𝑉+𝐼𝑅𝑆 𝑎 ) − 1] − 𝑉+𝐼𝑅𝑆 𝑅𝑆ℎ (15) 𝐼𝑆𝐶,𝑟𝑒𝑓 = 𝐼𝐿,𝑟𝑒𝑓 − 𝐼𝑂,𝑟𝑒𝑓 [𝑒𝑥𝑝 ( 𝐼𝑠𝑐,𝑟𝑒𝑓𝑅𝑠,𝑟𝑒𝑓 𝑎𝑟𝑒𝑓 ) − 1] − ( 𝐼𝑠𝑐,𝑟𝑒𝑓𝑅𝑠,𝑟𝑒𝑓 𝑅𝑠ℎ,𝑟𝑒𝑓 ) (16) where 𝐼𝑆𝐶,𝑟𝑒𝑓 is the reference short circuit, this means short circuit at standard test conditions (stc), diode reverse saturation current (io), 𝐼𝐿,𝑟𝑒𝑓 is light generated current at stc (a). the reference maximum current can be calculated from eq. 17; 𝐼𝑚𝑝,𝑟𝑒𝑓 = 𝐼𝐿,𝑟𝑒𝑓 − 𝐼𝑂,𝑟𝑒𝑓 [𝑒𝑥𝑝 ( 𝑉𝑚𝑝,𝑟𝑒𝑓+𝐼𝑚𝑝,𝑟𝑒𝑓𝑅𝑠,𝑟𝑒𝑓 𝑎𝑟𝑒𝑓 ) − 1] − ( 𝑉𝑚𝑝,𝑟𝑒𝑓+𝐼𝑚𝑝,𝑟𝑒𝑓𝑅𝑠,𝑟𝑒𝑓 𝑅𝑠ℎ,𝑟𝑒𝑓 ) (17) 𝐼 = 𝐼𝐿 − 𝐼𝑂 [𝑒𝑥𝑝 ( 𝑉+𝐼𝑅𝑆 𝑎 ) − 1] − 𝑉+𝐼𝑅𝑆 𝑅𝑠ℎ (18) table 1 typical conditions specified at three i-v points on the pv curve. point conditions short circuit current 𝐼 = 𝐼𝑠ℎ , 𝑉 = 0 [𝑑𝐼/𝑑𝑉]𝑠𝑐 = −1 𝑅𝑠ℎ,𝑟𝑒𝑓 open circuit voltage 𝐼 = 𝐼𝑠ℎ = 0, 𝑉 = 𝑉𝑂𝐶 ,𝑟𝑒𝑓 maximum power 𝐼 = 𝐼𝑚𝑎𝑥, 𝑟𝑒𝑓 , 𝑉 = 𝑉𝑚𝑎𝑥, 𝑟𝑒𝑓 [𝑑(𝐼𝑉)/𝑑𝑉]𝑚𝑎𝑥 where series resistance at standard test conditions (𝑅𝑠,𝑟𝑒𝑓 ), 𝐼𝑚𝑝,𝑟𝑒𝑓 is reference maximum current, 𝑉𝑚𝑝,𝑟𝑒𝑓 is reference maximum voltage, 𝑅𝑠ℎ,𝑟𝑒𝑓 is reference shunt resistance, as shown in eq. 19 to 21; 𝐼𝑚𝑝,𝑟𝑒𝑓 𝑉𝑚𝑝,𝑟𝑒𝑓 = ( 𝐼𝑂,𝑟𝑒𝑓 𝑎𝑟𝑒𝑓 ) 𝑒𝑥𝑝( 𝑉𝑚𝑝,𝑟𝑒𝑓+𝐼𝑚𝑝,𝑟𝑒𝑓𝑅𝑠,𝑟𝑒𝑓 𝑎𝑟𝑒𝑓 )+ 1 𝑅𝑠ℎ,𝑟𝑒𝑓 1+( 𝐼𝑂,𝑟𝑒𝑓𝑅𝑠,𝑟𝑒𝑓 𝑎𝑟𝑒𝑓 ) 𝑒𝑥𝑝( 𝑉𝑚𝑝,𝑟𝑒𝑓+𝐼𝑚𝑝,𝑟𝑒𝑓𝑅𝑠,𝑟𝑒𝑓 𝑎𝑟𝑒𝑓 )+ 𝑅𝑠,𝑟𝑒𝑓 𝑅𝑠ℎ,𝑟𝑒𝑓 (19) 𝑎 𝑎𝑟𝑒𝑓 = 𝑇𝐶 𝑇𝑐,𝑟𝑒𝑓 (20) 𝐼𝐿 = 𝑆 𝑆𝑟𝑒𝑓 [𝐼𝐿,𝑟𝑒𝑓 + 𝜇𝐼𝑠𝑐 (𝑇𝐶 − 𝑇𝐶,𝑟𝑒𝑓 )] (21) where the tc is current cell temperature and tc,ref is reference cell temperature. the absorbed solar radiation can be estimated from eq. 22; 𝑆 𝑆𝑟𝑒𝑓 = 𝑀 ( 𝐺𝑏 𝐺𝑟𝑒𝑓 𝑅𝐾𝜏𝛼,𝑏 + 𝐺𝑑 𝐺𝑟𝑒𝑓 𝐾𝜏𝛼,𝑑 ( 1+𝑐𝑜𝑠(𝛽) 2 ) + 𝐺 𝐺𝑟𝑒𝑓 𝜌𝑔𝐾𝜏𝛼,𝑔 ( 1−𝑐𝑜𝑠(𝛽) 2 )) (22) where sref is absorbed solar radiation at stc (w/m2), s is absorbed solar radiation, 𝐺𝑟𝑒𝑓 is solar intensity (w/m 2). 𝐼𝑂 = 𝐼𝑜𝑟𝑒𝑓 ( 𝑇𝐶 𝑇𝐶,𝑟𝑒𝑓 ) 3 𝑒𝑥𝑝 ( 𝜀 𝐾𝑇𝐶,𝑟𝑒𝑓 − 𝜀 𝐾𝑇𝐶 ) (23) 𝑅𝑠ℎ 𝑅𝑠ℎ,𝑟𝑒𝑓 = 𝑆𝑟𝑒𝑓 𝑆 (24) solving eq. 20 to 24 we can get an optimum condition. next, the operational values 𝐼𝑚𝑝 and 𝑉𝑚𝑝can be found from the eq 25 to 28 (bahaidarah et al. 2013); 𝐼𝑚𝑝 𝑉𝑚𝑝 = [ 𝐼𝑂 𝑎 𝑒𝑥𝑝( 𝑉𝑚𝑝+𝐼𝑚𝑝𝑅𝑠 𝑎 )+ 1 𝑅𝑠ℎ 1+ 𝑅𝑠 𝑅𝑠ℎ + 𝐼𝑜𝑅𝑠 𝑎 𝑒𝑥𝑝( 𝑉𝑚𝑝+𝐼𝑚𝑝𝑅𝑠 𝑎 ) ] (25) 𝐼𝑚𝑝 = 𝐼𝐿 − 𝐼𝑂 [𝑒𝑥𝑝 ( 𝑉𝑚𝑝+𝐼𝑚𝑝𝑅𝑆 𝑎 ) − 1] − [ 𝑉𝑚𝑝+𝐼𝑚𝑝𝑅𝑆 𝑅𝑠ℎ ] (26) where the maximum power 𝑃𝑚𝑎𝑥 and efficiency 𝜂𝑐 can be found by eq 27 and 28 as follows: 𝑃𝑚𝑎𝑥 = 𝐼𝑚𝑝𝑉𝑚𝑝 (27) electrical efficiency is defined as the ratio of the power output produced by the pv cell to the solar intensity and the unit area of the pv cell (bahaidarah et al. 2013) ga vi c c maxmax= (28) 3.4. correlations 𝑃 = 𝐺𝐴∅𝑒 −𝑎(1+ 𝑣�̇� 𝑄 ) (29) where g solar radiation, a area of pv, ∅ porosity, �̇� volume flow rate, l length of pv and 𝜐 kinetic viscosity. in order to get the temperature correlation, eq 34 and 35 were used as following; i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 318 issn: 2252-4940/©2023. the author(s). published by cbiore 𝑇 = 𝐺𝛼𝑒𝑓𝑓∅ 𝐾 𝑎 (1+𝑙𝑛( �̇� 𝜐𝐿 )) + 𝑇𝑖𝑛 (30) where g solar radiation, k thermal conductivity, ∅ porosity, �̇� volume flow rate, l length of pv, 𝜐 kinetic viscosity, and 𝑇𝑖𝑛 inlet water temperature. 𝛼𝑒𝑓𝑓 = 𝜏𝐺 [𝛼𝐶 + 𝛽𝐶 + 𝛼𝑇 (1 − 𝛽𝐶 )] (31) 3.5. non-uniformity radiation and temperature radiation non-uniformity(%) = 𝐸max−𝐸min 𝐸max+𝐸min *100 (32) temperature non-uniformity(%) = 𝑇max−𝑇min 𝑇max+𝑇min *100% (33) where emax is the maximum radiation, 𝑇max is the max temperature above the pv cell, respectively. emin and 𝑇min are minimum radiation and temperature above the pv cell respectively. the rate of increase in temperature and energy output: %𝑇𝑚𝑎𝑥, 𝑖𝑛𝑐𝑟𝑒𝑎𝑠𝑒 = 𝑇𝑤𝑖𝑡ℎ𝑜𝑢𝑡𝑐𝑜𝑜𝑙𝑖𝑛𝑔−𝑇𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑇𝑤𝑖𝑡ℎ𝑜𝑢𝑡𝑐𝑜𝑜𝑙𝑖𝑛𝑔 ∗ 100% (34) %𝑃𝑚𝑎𝑥, 𝑖𝑛𝑐𝑟𝑒𝑎𝑠𝑒 = 𝑃𝑐𝑜𝑜𝑙𝑖𝑛𝑔−𝑃𝑤𝑖𝑡ℎ𝑜𝑢𝑡𝑐𝑜𝑜𝑙𝑖𝑛𝑔 𝑃𝑤𝑖𝑡ℎ𝑜𝑢𝑡𝑐𝑜𝑜𝑙𝑖𝑛𝑔 ∗ 100% (35) 4. experimental setup 4.1. solar simulator the solar simulator was designed and assembled at the workshops and installed in the innovation lab in the faculty of engineering technology at al-balqa applied university jordan. the solar simulator can provide the amount of radiation independent of time or weather conditions. throughout all the experimental work, the irradiance value remained constant at 1600 w/m2. fig.4 shows a solar simulator with 50 halogen bulbs each one of 50 w and 220 v. the height of the lamp was about 51 cm from the surface of the pv panel to obtain radiation of about 1600 w/m2. furthermore, the solar simulator has been calibrated by placing a wooden board instead of the photovoltaic panel to obtain temperatures that are more consistent. because of the poor heat conductivity of the wood, it reflects the uniform radiation possible. to produce uniform radiation, each bulb was individually adjusted to reach a maximum temperature non-uniformity of 1.7 %, which is class a. to provide the required radiation and to reduce the nonuniformity radiation, the solar simulator was built with an area larger than the area of the solar panel. however, radiation nonuniformity was determined to be 3.5 %, which corresponds to class b according to eq 32 and 33 (rajput et al. 2018 and meng et al. 2011). 4.2. cooling system setup as illustrated in fig. 5, the cooling experimental setup consisted of a pv panel (30w poly crystalline) and a thermal collector. table 2 lists the specifications of the pv unit used in the solar simulator. the cooling channel (collector) filled with porous media and connected to the inlet/outlet water flow apertures, and was attached to the rear side of the pv panel. the channel's dimensions were the same as the pv panel. five different flow rates of water were used through the cooling channel. the cooling water came from a tank that was linked to the pv/t system through plastic tubing. a water pump (0.37 kw) was utilized to circulate water through the cooling channel, and a flow meter was used to measure and control the volume flow rate of water via the cooling channel. fig 4. schematic view of a complete setup of pv simulator. fig 5. solar thermal cooling channel. i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 319 issn: 2252-4940/©2023. the author(s). published by cbiore table 2 specifications of the photovoltaic panel solar pv module parameter value at am 1.5 spectra stc1 value at halogen light2 maximum power, pmax 30 w 18.72 w open -circuit voltage, voc 22 v 23 v short -circuit current, isc 1.84 a 1.25 a voltage at pmax 18 v 19.3 v current at pmax 1.67a 0.97 a operating temperature, oc -40 to +85 -40 to +85 dimensions (545x345) cm (545x345) cm radiation intensity, g 1000 w/m2 1600 w/m2 1 values refer to (standard test conditions) with tcell = 25°c, 1000 w/m2 and 1.5 am spectra. 2values refer to the present study conditions with tcell = 27°c and zero m/s wind speed under halogen light. fig 6. schematic diagram of the pv simulator setup fig. 6 shows that when the water flows through the channel underneath the pv panel, the heat from the pv panel is dissipated to the porous media and the associated water flow. to prevent mixing hot and cold water, the heat rejected from the pv unit warms the water in the channel, and then the water flows out of the channel is stored in another tank. the water flows at a rate of one cubic meter per hour. hence, the temperatures of the flowing water were monitored at the channel's inflow and outflow. 4.3. pv electrical model setup the measurement instruments were used to measure the pv unit performance. it consists of a thermometer, voltmeter, ammeter and solar density meter. front and back temperatures were measured at three different locations, and the mean temperatures were calculated. variable resistance was used to obtain the highest output power of the photovoltaic panel by measuring the highest volts and the current. the accuracy/ sensitivity values of the measuring instruments used are given in table 3. table 3 the accurateness/ sensitivity of the used measuring devices device accurateness/ sensitivity solar power meter ±10% w/m2 voltmeter and ammeter ± 1mv and ± 1µa thermocouple ±0.1°c 5. results and discussions 5.1. pv performance using porous media porous media offer a large exterior surface area and a high fluid permeability, making them ideal for pv cells cooling (ahmed et al, 2019 and masalha et al 2019). a porous media cooling system is typically composed of a coolant fluid (water) flowing through the porous medium's in the channels. the coolant fluid and the porous media transports heat away from pv back surface through by conduction and convection, thereby cooling the pv system. the fluid movement of the coolant through the porous medium determines the porous media's efficacy of transporting heat away from the pv hot surfaces. based on this approach, multiple indoor tests were carried out to measure the average pv hot surface temperatures with varied porosities (0.35, 0.4, 0.48, and 0.5) and at volume flow rates (1 l/min, 1.5 l/min, 2 l/min, 3 l/min, and 4 l/min). the results were collected and displayed in fig. 7. it can be clearly seen that the minimum average pv back surface temperature occurred at porosity of 0.35 and flow rate of 2 l/min and it is considered the best case (optimum condition). this may be due to larger surface area of low porosity that makes high homogenous mixtures between porous media structure and flowing fluid, so that they provided higher heat transfer medium interactions at the moderate flow rate of 2 l/min, as can be seen in fig. 7. fig 7. measured results of a pv cell average temperature at different volume flow rates and porosities 34 38 42 46 50 54 58 1 1.5 2 3 4 p v s u rf a c e t e m p e ra tu re (o c ) volume flow rate(l/min) porosity 0.35 porosity 0.4 porosity 0.48 porosity 0.5 water i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 320 issn: 2252-4940/©2023. the author(s). published by cbiore 0 20 40 60 80 100 120 140 160 180 33 34 35 36 37 38 39 40 41 p v s u rf a c e t e m p e ra tu re ( o c ) time (min) 1l/min 1.5l/min 2l/min 3l/min 4l/min (a) average pv surface temperature 60 80 100 120 140 160 180 18.0 18.2 18.4 18.6 18.8 19.0 19.2 p o w e r( w ) time (min) 1l/min 1.5l/min 2l/min 3l/min 4l/min (b) power output fig 8. measured average pv surface temperature and power output with respect to different volume flow rates at a porosity of 0.35 an average pv surface temperatures and power output for different volume flow rates vs time are shown in fig. 8a and 8b respectively. the results of different volume flow rates and fixed porosity of 0.35 were carried out for three hours for each test. therefore, it can be seen clearly from fig. 8 that the lowest recorded temperature of 38.70oc and highest power of about 19 w occurred at volume flow rate of 2 l/min and porosity of 0.35, which has been considered as a best scenario (optimum condition) evolved from the best heat removal condition from pv surface to the cooling agents. the variations of the temperatures between various flow rates were around 7 to 8%. the results in this research are somewhat similar to the previous works (msoud et al 2014; nizetic et al 2016) such an experimental work can be applied in a bulky real pv stations, so that the channels can be made from plastic materials, while porous media is available globally and the circulated water can be used for irrigation after long use. 5.2. numerical analysis validation different factors affected the performance of the cooled pv/t hybrid system by porous media can be examined using electrical and thermal models analyses that can be linked together. equations 2 and 27 were used in the mathematical model to verify numerically the experimental results, through the determination of the average pv back surface temperature and power output at various porosity and volume flow rates, as shown in fig. 9a and 9b. it was found out that the deviation between the experimental and numerical findings ranged between 3.2 % to 5.6% for average pv surface temperature and 1% to 2.8% for power output respectively, according to the discrepancy’s analyses. therefore, the numerical analysis results matched the experimental results trend with small deviation. the deviation between both methods can be attributed to steadiness and idealization of the flow and heat transfer assumptions considered in the numerical results. the numerical simulation showed that it can be used to estimate the behaviour of the pv cooling processes in prior of commencing any laborious experimental work. experimental tests of the cases of the pv water alone cooling and pv without cooling over a test period of three continuous hours have been compared with optimum case. the purpose of these tests is to get an overall view of the pv performance under these conditions. the main performance parameters such as surface temperature, power output, efficiency, and i-v characteristic curve were investigated, as shown in fig. 10. fig.10a shows an average surface temperature, power, efficiency and current-voltage characteristic curve of the pv panel for three cases (without cooling, water alone cooling, and optimum case (porosity of 0.35 and volume flow rate of 2 l/min). it can be noticed that the pv surface temperature increased dramatically at the beginning of the pv operation without cooling and then stabilizes for the rest of time, while the cooling with porous media was stabilized from the beginning to the end of the test due to the homogeneity in thermal mass flow rate through porous media. also, it is better than water alone cooling. therefore, the 0.35 porosity had reduced pv panel surface temperature by 55.8% compared to uncooled process. figs 10b and c show the experimental findings for the time-dependent variation of the pv panel power production and the efficiency for all three scenarios. it was found that, as experiments progressed, the power output and efficiency of the pv panel without cooling dropped dramatically before stabilization after certain period of time, while for the optimum case it remained steady with high power and efficiency till the end of operation. therefore, the porous media cooling enhanced the pv power and efficiency more than 51.69% during its operation compared to uncooling; and 2.13% more than water alone cooling. fig. 10d shows the variation of i-v curve of pv panel versus the open voltage for all three cases at a volume rate of 2 l/min. it is clearly seen that the optimum case reached a higher voltage of 22.54 v than other cases. overall, the results showed that the presence of low porosity media of 0.35 in the pv cooling process was more effective than the other two scenarios, because the heat loss from pv surface through porous media layer have developed a homogenous heat diffusion removal and specially at moderate flow rate (2 l/min). the result was similar the result done by msalha et al (2019) i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 321 issn: 2252-4940/©2023. the author(s). published by cbiore 40 60 80 100 120 140 160 180 36 38 40 42 44 p v s u rf a c e t e m p e ra tu re ( o c ) time (min) numerical experimental (a) 1l/min 40 60 80 100 120 140 160 180 34 36 38 40 42 44 p v s u rf a c e t e m p e ra tu re ( o c ) time (min) numerical experimental (b) 1.5l/min 40 60 80 100 120 140 160 180 30 32 34 36 38 40 42 44 p v s u rf a c e t e m p e ra tu re ( o c ) time (min) numerical experimental (c) 2l/min 0 20 40 60 80 100 120 140 160 180 17.0 17.5 18.0 18.5 19.0 19.5 20.0 p o w e r( w ) time (min) numerical experimenta (d) 1l/min 0 20 40 60 80 100 120 140 160 180 18.0 18.5 19.0 19.5 20.0 p o w e r( w ) time (min) numerical experimenta (e) 1.5l/min 0 20 40 60 80 100 120 140 160 180 17.0 17.5 18.0 18.5 19.0 19.5 20.0 p o w e r( w ) time (min) numerical experimenta (f) 2l/min fig 9. comparison between numerical and measured results of pv output power and its surface temperature at various volume flow rates (1, 1.5 and 2 l/min i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 322 issn: 2252-4940/©2023. the author(s). published by cbiore table 4 reduction percentage in pv surface temperature at various volume flow rates and porosities in comparison with water alone cooling process. volume flow rate (l/min) pv cell surface temperature (oc) with different porosity reduction (%) in pv surface temperature with water alone cooling process 0.35 0.40 0.48 0.50 water alone 0.35 0.40 0.48 0.50 1.0 40.57 43.00 45.00 45.50 51.55 21.30 16.59 12.71 11.74 1.5 40.16 41.53 42.80 43.30 50.90 21.10 18.41 15.91 14.93 2.0 38.70 40.35 42.36 42.80 48.55 20.29 16.89 12.75 11.84 3.0 39.70 40.20 41.43 41.90 50.25 21.00 20.00 17.55 16.62 4.0 39.60 40.00 41.40 41.85 48.50 18.35 17.53 14.64 13.71 0 20 40 60 80 100 120 140 160 180 20 30 40 50 60 70 80 90 p v s u rf a c e t e m p e ra tu re ( o c ) time (min) without cooling water cooling porous media (a) pv surface temperature 0 20 40 60 80 100 120 140 160 180 12 13 14 15 16 17 18 19 20 p o w e r( w ) time (min) without cooling water cooling porous media (b) power output 0 20 40 60 80 100 120 140 160 180 4.0 4.5 5.0 5.5 6.0 6.5 e ff ic ie n c y % time (min) without cooling water cooling porous media (c) efficiency 0 5 10 15 20 25 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 c u rr e n t( a ) time (min) without cooling water cooling porous media (d) i-v curve fig 10. comparison of pv performance under three circumstances (without cooling, water cooling with a flow of 2 l/min, and cooling with a flow of 2 l/min through porous media) 5.3. photovoltaic panel power enhancement and surface temperature reduction to investigate the percentage increase in the output power and reduction in the surface temperature of the pv panel due to cooling and without cooling pv panel. karami's et al (2014) formulae were used to calculate the percentage decrease in average pv surface temperature and increase in power. tables 4 and 5 compare the reduction percentage in pv surface temperature calculated for different volume flow rates and porosities with the water alone cooling and uncooled process. it was noticed from the results in the tables 6 and 7 that the moderate flow rates and low porosity media removed pv heat loss much quicker than other cases, pv panel power output was measured, and the efficiency was estimated for various volume flow rates (1, 1.5, 2, 3, and 4 l/min) and porosities (0.35, 0.4, 0.48, and 0.5) as shown in tables 6 and 7. additionally, power production and pv efficiency improvement percentages are computed as well. it can be seen that optimum case proved to be the best scenario because it has the highest power production of 18.84 w and the highest efficiency 6.26 %. in addition, the data in the tables showed that porosity size has a greater impact on power production beside the flow rates. however, it is worth noting i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 323 issn: 2252-4940/©2023. the author(s). published by cbiore that the optimal case increased electricity production by 51.69 % and pv efficiency by 2.13 % compared to uncooled process. the pv power output and efficiency for the water alone case was found to be 39.29% and 0.39% respectively, compared with uncooled process. therefore, the optimum case enhanced the pv power and efficiency by 9% more than water alone process. 6. correlations the correlation process is one of the good methods of analysis, because it allows better understanding of relationships between different parameters affecting the pv performance, before conducting an experimental work that needs time, cost and efforts. so that correlations were developed to enable quick prediction of pv surface temperature and power due to cooling. 6.1. power correlation equation 29 was used to generate a power correlation as a function of several pv/t geometrical factors, such as solar radiation, pv area, porosity, volume flow rate, pv length, and kinetic viscosity. the comparison of experimental and correlation results for various pv panel power output with time was shown in fig. 11 for optimum case (i.e. porosity of 0.35 and volume flow rates of 2 l/min). the coupled power model was developed to examine several performance characteristics of the pv/t hybrid system cooled with porous media. correlation findings were calculated using eq. 29 to ensure the accuracy, and then these correlations have been checked against optimum case for optimization purposes. fig. 11 shows the experimental and correlations outcome for optimum case. both trends were nearly similar with a deviation of about 1.7% especially during the stable period of cooling process. table 5 reduction percentage in pv surface temperature at various volume flow rates and porosities in comparison with uncooled process. volume flow rate (l/min) pv cell surface temperature (oc) with different porosity reduction (%) in pv surface temperature compared with uncooled process 0.35 0.40 0.48 0.50 0.35 0.40 0.48 0.50 1.0 40.57 43.00 45.00 45.50 53.74 50.97 48.69 48.12 1.5 40.16 41.53 42.80 43.30 54.21 52.65 51.20 50.63 2.0 38.70 40.35 42.36 42.80 55.87 53.99 51.70 51.20 3.0 39.70 40.20 41.43 41.90 54.73 54.16 52.76 52.22 4.0 39.60 40.00 41.40 41.85 54.84 54.39 52.79 52.28 table 6 percentage increase in power output at various volume flow rates and porosities in comparison with uncooled process. volume flow rate output power (w) with different porosities uncooled power = 12.42w increase power (%) (l/min) 0.35 0.4 0.48 0.5 water alone 0.35 0.4 0.48 0.5 1.0 18.01 17.63 16.9 16.9 16.67 45.01 41.95 36.07 34.62 1.5 18.43 18 17.67 17.5 17.15 48.39 44.93 42.27 39.25 2.0 18.84 18.4 17.8 17.67 17.3 51.69 48.15 43.32 40.57 3.0 18.7 18.4 18.16 18 16.97 50.56 48.15 46.22 43.12 4.0 18.73 18.4 18.16 18.1 17.3 50.81 48.15 46.22 43.89 table 7 percentage increase in efficiency at various volume flow rates and porosities in comparison with uncooled process. volume flow rate efficiency (%) with different porosities uncooled efficiency = 4.13% enhancement (%) in efficiency (l/min) 0.35 0.4 0.48 0.5 water alone 0.35 0.4 0.48 0.5 1.0 5.99 5.86 5.62 5.62 5.54 1.86 1.73 1.49 1.49 1.5 6.13 5.98 5.87 5.82 5.70 2.00 1.85 1.74 1.69 2.0 6.26 6.12 5.92 5.87 5.75 2.13 1.99 1.79 1.74 3.0 6.22 6.12 6.04 5.98 5.64 2.09 1.99 1.91 1.85 4.0 6.23 6.12 6.04 6.02 5.75 2.1 1.99 1.91 1.89 i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 324 issn: 2252-4940/©2023. the author(s). published by cbiore 60 80 100 120 140 160 180 17 18 19 20 21 22 p o w e r( w ) time (min) experimental correlation fig 11. comparing experimental data and correlation data of power output at a volume flow rate of 2 l/min with porosity of φ = 0.35 0 20 40 60 80 100 120 140 160 180 26 28 30 32 34 36 38 40 42 44 p v p a n e l s u rf a c e t e m p e ra tu re ( o c ) time (min) experimental correlation fig 12. comparing experimental data and correlation data of pv surface temperature for a volume flow rate of (2 l/min) with porosity of φ = 0.35 6.2. correlation for temperature as a coupled model, the pv cell surface temperature model was developed to examine various performance characteristics of a pv/t hybrid system with optimum case. fig. 12 shows the experimental and correlation for optimum case. it can be seen that the maximum deviations between experimental and correlation results were found to be 5.2%. this means that the experimental data can be validated with the correlation data for validity. the experimental and correlations results showed that they are in good agreement and had same trend of the behaviour for the optimum case. therefore, the correlation model and experimental data gave clear indication that 0.35 porosity and 2 l/min flow rate is the best possible option for the pv panels cooling. table 8 presents many researchers found different findings on the reduction of pv temperatures and increase of power, this is due to different solar radiation measurements and different ambient conditions as well. it can also be noticed from table 8, that some researchers got nearly the reduction of temperature similar to current results such as msoud et al (2014), nizetic et al (2016), and hernandez et al (2013) but with lower power output than current study. therefore, current study showed that the cooling process used enhanced the power output and efficiency more efficiently than other cooling processes used by previous researchers. 7. conclusions the experimental and theoretical study on cooling the pv panel using porous media with various porosities and volume flow rates have been investigated. cooling pv panel using a porosity of 0.35 and flow rate of 2 l/min showed the highest increase in the power output, efficiency and i-v curve characteristics, correspondent to a higher reduction in the pv hot surface temperature of 55.87% more than uncooled process. the highest pv hot surface temperature of 87.7°c occurred for uncooled pv panel. using water-alone cooling, the temperature reduced to 48.55°c, while, using water cooling with porous media of porosity 0.35 and volume flow rate of 2 l/min (optimum case), the temperature reduced to 38.7°c. also, it is noticed that the optimum case registered a higher voltage of 22.54 v than other cases. table 8 a comparison between the current work and the previous work. the decrease in temperature and the increase in the output power percentage references reduction of temperature(oc) increase of power % current work 32 34-51 ahmad et al, 2016 10 6.6 bahaidarah et al, 2013 17 15.5 hernández et al, 2013 26 15 masalha et al, 2019 31 7.48 masalha et al, 2020 22 7.5 masalha et al, 2021 26 9.6 masoud et al, 2014 26 15 meysam et al, 2014 13 7.7 nižetić et al, 2016 26 14 shenyi et al, 2014 12.5 15 stefan et al, 2004 22 10.5 xiao et al, 2010 8 5.4 i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 325 issn: 2252-4940/©2023. the author(s). published by cbiore it can be concluded that optimum case proved to be the best scenario because it has the highest power production of 19 w and the highest efficiency 6.26 %. also, the optimum case enhanced the pv power and efficiency by 9% more than water alone process. the pv power output can be increased by changes in porosity and volume flow rate cooling. there is a clear correlation between the average pv hot surface temperature and porosity. when the porosity decreases the reduction in pv panel surface temperature increases, that is due to the homogeneous diffusivity of convective heat lose attained at lower porosities and higher flow rate. also, porous media offer a large exterior surface area and a high fluid permeability, making them ideal for pv cells cooling. the experimental and correlation results showed that they are in good agreement and had same trend of the behaviour for the optimum case. further studies shall be conducted in near future on the nano porous media to find out their effect on pv cooling process. references abu bakar, m.n, mahmod, o., mahathir, d., norain, m.a. (2013), development of an improved photovoltaic/thermal (pv/t) solar collector with bi-fluid configuration. international journal of chemical and environmental engineering, 4, 234. https://journaldatabase.info/articles/development_improved_p hotovoltaic.html ahmad, h., alnoman, h., hasan, s.a. (2016) energy efficiency enhancement of photovoltaics by phase change materials through thermal energy recovery. energies, 9, 782. https://doi.org/10.3390/en9100782 ahmed, o.k., hamada, k.i., salih, a.m. (2019). enhancement of the performance of photovoltaic/trombe wall system using the porous medium: experimental and theoretical study. energy, 171, 14-26 https://doi.org/10.1016/j.energy.2019.01.001 aldossary, a., mahmoud, s., al-dadah, r. (2016). technical feasibility study of passive and active cooling for concentrator pv in harsh environment. applied thermal engineering, 100, 490-500. https://doi.org/10.1016/j.applthermaleng.2016.02.023 alizadeh, h., ghasempour, r., shafii, m.b., ahmadi, m.h., yan, w.m., nazari, m.a. (2018). numerical simulation of pv cooling by using single turn pulsating heat pipe. international journal of heat and mass transfer, 127(a), 203-208; https://doi.org 10.1016/j.ijheatmasstransfer.2018.06.108 al-musawi, a.i.a., taheri, a., farzanehnia, a., sardarabadi, m., passandideh-fard, m. (2018). numerical study of the effects of nanofluids and phase-change materials in photovoltaic thermal (pvt) systems. journal of thermal analysis and calorimetry, 137(2), 623-636; https://doi.org/10.1007/s10973-018-7972-6 amelia, a.r., irwan, y.m., irwant, m., leow, w.z., gomesh, n., safwati, i., anuar, m.a.m. (2016). cooling on photovoltaic panel using forced air convection induced by dc fan. international journal of electrical and computer engineering, 6: 526-534. https://doi.org /10.11591/ijece.v6i2.9118 badran, o., abdulhadi, e., mamlook, r. (2010). evaluation of solar electric power technologies in jordan. jordan journal of mechanical and industrial engineering, 4, 121-128. https://www.researchgate.net/publication/228655407_evaluati on_of_solar_electric_power_technologies_in_jordan bahaidarah h, subhan abdul, rehman s. (2013). performance evaluation of a pv (photovoltaic) module by back surface water cooling for hot climatic conditions. energy, 59, 445–53 https://doi.org/10.1016/j.energy.2013.07.050 bahaidarah, h.m.s. (2016). experimental performance evaluation and modeling of jet impingement cooling for thermal management of photovoltaics. solar energy, 135: 605-617. https://doi.org/10.1016/j.solener.2016.06.015 bahaidarah, h.m.s., baloch, a.a.b., gandhidasan, p. (2016) uniform cooling of photovoltaic panels: a review. renewable and sustainable energy reviews, 57, 1520-1544. https://doi.org/10.1016/j.rser.2015.12.064 buker, m.s, mempouo, b., riffat, s.b. (2015). experimental investigation of a building integrated photovoltaic/thermal roof collector combined with a liquid desiccant enhanced indirect evaporative cooling system. energy conversion and management, 101, 239–254. https://doi.org/10.1016/j.enconman.2015.05.026 chandrasekara, m., suresh, s., senthilkumar, t., and karthikeyan, m.g. (2013). passive cooling of standalone flat pv module with cotton wick structures. energy conversion and management, 71, 43–50. https://doi.org/10.1016/j.enconman.2013.03.012 ebrahimi, m., rahimi, m., and rahimi, a. (2015). an experimental study on using natural vaporization for cooling of a photovoltaic solar cell, international communications in heat and mass transfer, 65: 22-30.https://doi.org/10.1016/j.icheatmasstransfer.2015.04.002 elminshawy, n.a.s., el-ghandour, m., elhenawy, y., bassyouni, m., eldamhogi, and addas, m.f, (2019). experimental investigation of a v-trough pv concentrator integrated with a buried water heat exchanger cooling system. solar energy, 193, 706-714. https://doi.org/10.1016/j.solener.2019.10.013 elminshawy, n.a.s., el-ghandour, m., gad, d.g., h.m. el-damhogi, elnahhas, k., and addas, m.f. (2019). the performance of a buried heat exchanger system for pv panel cooling under elevated air temperatures. geothermic, 82, 7-15. https://doi.org/10.1016/j.geothermics.2019.05.012 elnozahy, a., abdel rahman, a.k., ali, a.h.h., abdel-salam, m., and ookawara, s. (2015). performance of a pv module integrated with standalone building in hot arid areas as enhanced by surface cooling and cleaning. energy and buildings, 88, 100–109. https://doi.org/10.1016/j.enbuild.2014.12.012 golzari, s., kasaeian, a., amidpour, m., nasirivatan, s., and mousavi, s. (2018). experimental investigation of the effects of corona wind on the performance of an air-cooled pv/t. renewable energy, 127, 284-297. https://doi.org/10.1016/j.renene.2018.04.029 idoko, l., anaya-lara, o., and mcdonald, a. (2018). enhancing pv modules efficiency and power output using multi-concept cooling technique. energy reports, 4, 357-369. https://doi.org/10.1016/j.egyr.2018.05.004 irshad, k., habib, k., thirumalaiswamy, n., and elmahdi, a.e.a. (2014). performance analysis of photo voltaic trombe wall for tropical climate. appl mech mater, 1, 211-215. http://eprints.utp.edu.my/id/eprint/10577/ irwan, y.m., leow, w.z., irwanto, m., fareq. m, amelia, a.r., gomesh, n. and safwati, i. (2015). indoor test performance of pv panel through water cooling method, energy procedia, 79, 604611. https://doi.org/10.1016/j.egypro.2015.11.540 jamali, s., yari, m., and mahmoudi, s.m.s. (2018). enhanced power generation through cooling a semi-transparent pv power plant with a solar chimney. energy conversion and management, 175, 227-235. https://doi.org/10.1016/j.enconman.2018.09.004 józsef, b., balogh, a., gabnai, p., pályi, z., farkas, b., pintér, i., and zsiborács, g. (2016). technical and economic effects of cooling of monocrystalline photovoltaic modules under hungarian conditions. renewable and sustainable energy, 60, 1086-1099 https://doi.org/10.1016/j.rser.2016.02.003 kabeel, a.e., abdelgaied, m., and sathyamurthy, r. (2019). a comprehensive investigation of the optimization cooling technique for improving the performance of pv module with reflectors under egyptian conditions. solar energy, 186, 257-263. https://doi.org/10.1016/j.solener.2019.05.019 karami n., and rahimi, m. (2014). heat transfer enhancement in a pv cell using boehmite nanofluid. energy conversion and management, 86, 275–85. https://doi.org/10.1016/j.enconman.2014.05.037 krauter, s. (2004), increased electrical yield via water flow over the front of photovoltaic panels. sol energy mater sol cells, 82, 131–7. https://doi.org/10.1016/j.solmat.2004.01.011 ma, t., yang, h., zhang, y. lu, l., and wang, x. (2015). using phase change materials in photovoltaic systems for thermal regulation and electrical efficiency improvement. a review and outlook”, renewable and sustainable energy reviews, 43, 1273–1284. https://doi.org/10.1016/j.rser.2014.12.003 masalha,.i., abdullah,n., & rawashdeh, m. (2019) experimental and numerical investigation of pv module for better efficiency using porous media, international journal of mechanical and) https://journaldatabase.info/articles/development_improved_photovoltaic.html https://journaldatabase.info/articles/development_improved_photovoltaic.html https://doi.org/10.3390/en9100782 https://doi.org/10.1016/j.energy.2019.01.001 https://doi.org/10.1016/j.applthermaleng.2016.02.023 https://link.springer.com/journal/10973 https://doi.org/10.1007/s10973-018-7972-6 http://dx.doi.org/10.11591/ijece.v6i2.9118 http://dx.doi.org/10.11591/ijece.v6i2.9118 https://www.researchgate.net/publication/228655407_evaluation_of_solar_electric_power_technologies_in_jordan https://www.researchgate.net/publication/228655407_evaluation_of_solar_electric_power_technologies_in_jordan https://doi.org/10.1016/j.energy.2013.07.050 https://doi.org/10.1016/j.solener.2016.06.015 https://doi.org/10.1016/j.rser.2015.12.064 https://doi.org/10.1016/j.enconman.2015.05.026 https://doi.org/10.1016/j.enconman.2013.03.012 https://doi.org/10.1016/j.icheatmasstransfer.2015.04.002 https://doi.org/10.1016/j.solener.2019.10.013 http://dx.doi.org/10.1016/j.geothermics.2019.05.012 https://doi.org/10.1016/j.enbuild.2014.12.012 https://doi.org/10.1016/j.renene.2018.04.029 https://doi.org/10.1016/j.egyr.2018.05.004 http://eprints.utp.edu.my/id/eprint/10577/ https://www.sciencedirect.com/science/article/pii/s1876610215022729#! https://www.sciencedirect.com/science/article/pii/s1876610215022729#! https://www.sciencedirect.com/science/article/pii/s1876610215022729#! https://www.sciencedirect.com/science/article/pii/s1876610215022729#! https://www.sciencedirect.com/science/article/pii/s1876610215022729#! https://www.sciencedirect.com/science/article/pii/s1876610215022729#! https://www.sciencedirect.com/science/article/pii/s1876610215022729#! https://www.sciencedirect.com/science/article/pii/s1876610215022729#! https://doi.org/10.1016/j.egypro.2015.11.540 https://doi.org/10.1016/j.enconman.2018.09.004 https://doi.org/10.1016/j.rser.2016.02.003 https://doi.org/10.1016/j.solener.2019.05.019 https://doi.org/10.1016/j.enconman.2014.05.037 http://dx.doi.org/10.1016/j.solmat.2004.01.011 http://dx.doi.org/10.1016/j.solmat.2004.01.011 https://doi.org/10.1016/j.rser.2014.12.003 i. al-masalha et al. int. j. renew. energy dev 2023, 12(2), 313-327 | 326 issn: 2252-4940/©2023. the author(s). published by cbiore production engineering research and development, 9(4), 1283–1302 https://www.researchgate.net/publication/338101467_experi mental_and_numerical_investigation_of_pv_mo dule_for_better_efficiency_using_porous_media masalha, i., elayyan, m., alfaqs, f., & fayyad, s. (2020) experimental investigations for improving pv module efficiency using nanofluid, international journal of mechanical and production engineering research and development, 10(2), 1085–1098 https://www.researchgate.net/publication/340443456_experi mental_investigations_for_improving_pv_modul e_efficiency_using_nanofluid masalha, i., elayyan, m., al-jamea, d., badran, o., alsabagh, a., darweesh, n., (2021). an experimental and numerical study to improve the efficiency of pv modules by using nano-fluid cooling system, international review of mechanical engineering, 15(11), https://doi.org/10.15866/ireme.v15i11.21385 masoud, r. sheyda,v., parsamoghadam,p., amin, m, moein, m. m., abdulaziz.a.a., (2014) design of a self-adjusted jet impingement system for cooling of photovoltaic cells. energy convers management. 83, 48–57. https://doi.org/10.1016/j.enconman.2014.03.053 hernández, r., cascales j.r., garcía, f., káiser, a.s., zamora, b. (2013), improving the electrical parameters of a photovoltaic panel by means of an induced or forced air stream. int j photoenergy, 2013, article id 830968. https://doi.org/10.1155/2013/830968 meng, q., wang, y., and zhang, l. (2011). irradiance characteristics and optimization design of a large-scale solar simulator, solar energy, 85, 1758–1767. https://doi.org/10.1016/j.solener.2011.04.014 micheli, l., reddy, k.s., and mallick, t.k., (2015). plate micro-fins in natural convection: an opportunity for passive concentrating photovoltaic cooling, energy procedia, 82, 301–308. https://doi.org/10.1016/j.egypro.2015.12.037 masalha.i , masuri. s.u., badran. o.o., ariffin. m.k.a.m., abu talib. a.r., alfaqs f. (2023), outdoor experimental and numerical simulation of photovoltaic cooling using porous media. case studies in thermal engineering,42, 102748 https://doi.org/10.1016/j.csite.2023.102748 mojumder, j.c, chong, w.t, ong, h.c, leong, k.y., and al-mamoon, a. (2016). an experimental investigation on performance analysis of air type photovoltaic thermal collector system integrated with cooling fins design. energy and buildings, 130, 272-285. https://doi.org/10.1016/j.enbuild.2016.08.040 elayyan, m., al masalha, i., al alawin, a., maaitah, h., alsabagh, a., (2020), new design of a solar collector reflector, international review of mechanical engineering, 14 (3), 185-191. https://doi.org/10.15866/ireme.v14i3.18476 al-jamea. d. m. k., masalha. i., alsabagh. a. s., badran. o. o., maaitah. h., mashaqbeh. o. (2022), investigation on water immersing and spraying for cooling pv panel, international review of mechanical engineering, 16. https://doi.org/10.15866/ireme.v16i9.22680 moradgholi, m., mostafa,n., iman, a., (2014) application of heat pipe in an experimental investigation on a novel photovoltaic/thermal (pv/t) system. solar energy, 107, 82–8. https://doi.org/10.1016/j.solener.2014.05.018 nižetić, s., čoko, d., yadav, a., and grubišić-čabo, f. (2016). water spray cooling technique applied on a photovoltaic panel: the performance response. energy conversion and management, 105, 287-296. https://doi.org/10.1016/j.enconman.2015.10.079 özakin, a.n, and kaya, f. (2019). effect on the exergy of the pvt system of fins added to an air-cooled channel: a study on temperature and air velocity with ansys fluent. solar energy, 184, 561-569; https://doi.org/10.1016/j.solener.2019.03.100 masalha, i., elayyan, m., aldean bani issa, h., (2017) use of biogas energy in poultry farming heating, the international journal of engineering and science (ijes), 6, 3, 58-63. https://doi.org/10.9790/1813-0603025863 hammad, m., ebaid, m, and masalha.i, (2015), performance study of a domestic refrigerator designed to use butane asrefrigerant when replaced by r 134a powered by solar energy. conference: 5th. jordanian iir conference on refriugeration and air conditioning, jiircrac, 15 at: aqaba, jordan. https://www.researchgate.net/publication/269808404_perform ance_study_of_a_domestic_refrigerator_designed_to_use_but ane_as_refrigerant_when_replaced_by_r134a_powered_by_s olar_energy radwan, a., ahmed, m., ookawara, s. (2016). performance enhancement of concentrated photovoltaic systems using a microchannel heat sink with nanofluids. energy conversion and management,119,289–303. https://doi.org/10.1016/j.enconman.2016.04.045 rahimi, m., asadi, m., karami, n., and karimi, e. (2015). a comparative study on using single and multi-header microchannels in a hybrid pv cell cooling. energy conversion and management, 101,1-8. https://doi.org/10.1016/j.enconman.2015.05.034 rajput, u. and yang, j. (2018). comparison of heat sink and water type pv/t collector for polycrystalline photovoltaic panel cooling, renewable energy, 116, 479-491. https://ideas.repec.org/a/eee/renene/v116y2018ipap479491.html ramadan, m.r.i., el-sebaii, a.a., aboul-enein, s., and el-bialy, e. (2007). thermal performance of a packed bed double-pass solar air heater. energy, 32(8), 1524-1535. https://doi.org/10.1016/j.energy.2006.09.019 sahay, a., sethi, v.k., tiwari, a.c., pandey, m. (2015). a review of solar photovoltaic panel cooling systems with special reference to ground coupled central panel cooling system (gc-cpcs). renewable and sustainable energy reviews, 42, 306–312. https://doi.org/10.1016/j.rser.2014.10.009 shenyi, w., chenguang, x., (2014). passive cooling technology for photovoltaic panels for domestic houses. int j low-carbon technol 9,118–126. https://doi.org/10.1093/ijlct/ctu013 soliman, a.m.a. and hassan, h. (2019). effect of heat spreader size, microchannel configuration and nanoparticles on the performance of pv-heat spreader-microchannels system. solar energy, 182, 286–297. https://coek.info/pdf-effect-of-heatspreader-size-microchannel-configuration-and-nanoparticles-onthe.html stropnik, r. and stritih, u. (2016). increasing the efficiency of pv panel with the use of pcm. renewable energy, 97, 671-679. https://doi.org/10.1016/j.renene.2016.06.011 xiao, t., zhenhua, q., yaohua, z., (2010), experimental investigation of solar panel cooling by a novel micro heat pipe array. energy power eng; 2, 171–4. https://doi.org/10.1109/appeec.2010.5449518 tous, y., and abdelhafith, s. (2013). feasibility of residential grid connected pv system under the jordanian net metering renewable energy law. journal of energy technologies and policy, 3, 2224-3232. https://www.researchgate.net/publication/259137109_feasibili ty_of_residential_grid_connected_pv_system_under_the_jorda nian_net_metering_renewable_energy_law wu, s.y., chen, c., and xiao, l. (2018). heat transfer characteristics and performance evaluation of water-cooled pv/t system with cooling channel above pv panel. renewable energy, 125, 936-946. https://doi.org/10.1016/j.renene.2018.03.023 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://www.researchgate.net/publication/338101467_experimental_and_numerical_investigation_of_pv_module_for_better_efficiency_using_porous_media https://www.researchgate.net/publication/338101467_experimental_and_numerical_investigation_of_pv_module_for_better_efficiency_using_porous_media https://www.researchgate.net/publication/338101467_experimental_and_numerical_investigation_of_pv_module_for_better_efficiency_using_porous_media https://www.researchgate.net/publication/340443456_experimental_investigations_for_improving_pv_module_efficiency_using_nanofluid https://www.researchgate.net/publication/340443456_experimental_investigations_for_improving_pv_module_efficiency_using_nanofluid https://www.researchgate.net/publication/340443456_experimental_investigations_for_improving_pv_module_efficiency_using_nanofluid http://dx.doi.org/10.15866/ireme.v15i11.21385 https://doi.org/10.1016/j.enconman.2014.03.053 https://doi.org/10.1155/2013/830968 https://doi.org/10.1016/j.solener.2011.04.014 https://doi.org/10.1016/j.egypro.2015.12.037 https://www.sciencedirect.com/journal/case-studies-in-thermal-engineering https://www.sciencedirect.com/journal/case-studies-in-thermal-engineering https://www.sciencedirect.com/journal/case-studies-in-thermal-engineering/vol/42/suppl/c https://doi.org/10.1016/j.csite.2023.102748 https://doi.org/10.1016/j.enbuild.2016.08.040 https://doi.org/10.15866/ireme.v14i3.18476 https://doi.org/10.15866/ireme.v16i9.22680 https://doi.org/10.1016/j.solener.2014.05.018 https://www.sciencedirect.com/science/article/pii/s0196890415010018#! https://www.sciencedirect.com/science/article/pii/s0196890415010018#! https://www.sciencedirect.com/science/article/pii/s0196890415010018#! https://www.sciencedirect.com/science/article/pii/s0196890415010018#! https://doi.org/10.1016/j.enconman.2015.10.079 https://doi.org/10.1016/j.solener.2019.03.100 http://dx.doi.org/10.9790/1813-0603025863 https://doi.org/10.1016/j.enconman.2016.04.045 https://doi.org/10.1016/j.enconman.2015.05.034 https://ideas.repec.org/a/eee/renene/v116y2018ipap479-491.html https://ideas.repec.org/a/eee/renene/v116y2018ipap479-491.html https://doi.org/10.1016/j.energy.2006.09.019 https://doi.org/10.1016/j.rser.2014.10.009 https://doi.org/10.1093/ijlct/ctu013 https://coek.info/pdf-effect-of-heat-spreader-size-microchannel-configuration-and-nanoparticles-on-the.html https://coek.info/pdf-effect-of-heat-spreader-size-microchannel-configuration-and-nanoparticles-on-the.html https://coek.info/pdf-effect-of-heat-spreader-size-microchannel-configuration-and-nanoparticles-on-the.html http://dx.doi.org/10.1016/j.renene.2016.06.011 http://dx.doi.org/10.1016/j.renene.2016.06.011 https://doi.org/10.1109/appeec.2010.5449518 https://www.researchgate.net/publication/259137109_feasibility_of_residential_grid_connected_pv_system_under_the_jordanian_net_metering_renewable_energy_law https://www.researchgate.net/publication/259137109_feasibility_of_residential_grid_connected_pv_system_under_the_jordanian_net_metering_renewable_energy_law https://www.researchgate.net/publication/259137109_feasibility_of_residential_grid_connected_pv_system_under_the_jordanian_net_metering_renewable_energy_law https://doi.org/10.1016/j.renene.2018.03.023 international journal of renewable energy development int. j. renew. energy dev. 2023, 12(2), 419-429 | 419 https://doi.org/10.14710/ijred.2023.50361 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id sustainable long-term energy supply and demand: the gradual transition to a new and renewable energy system in indonesia by 2050 yudiartono yudiartonoa,b* , jaka windartaa , adiarso adiarsob a master program of energy, school of postgraduate studies, diponegoro university, indonesia bthe national research and innovation agency, indonesia abstract. the objective of this work is to evaluate long-term energy demand and supply decarbonization in indonesia. on the demand side, electric vehicles and biofuels for transportation and induction stoves and urban gas networks for households were considered. based on the national energy policy, primary energy supply projections optimized nre power plant use and increase nre's position in the national energy mix. a low emissions analysis platform (leap) model evaluates 2020–2050 energy demand predictions and low-carbon energy systems. this study's sustainable transition options require two basic technical advances. first, electric vehicles and induction stoves would reduce oil fuel usage by 228.34 million boe and lpg consumption by 24.65 million boe. second, power generation should be decarbonized using nre sources such as solar, hydro, biomass, geothermal, and nuclear. in 2050, solar power (40 gw), hydropower (38.47 gw), geothermal power (10 gw), and other nre (24.45 gw, 18.67 gw of which would be biomass power) would dominate nre electrical capacity. biomass co-firing for coal power plants would reach 36.35 million tons in 2050. in 2035, the java-bali or west kalimantan system will deploy 1 gw of nuclear power reactors, rising to 4 gw by 2050. under the transition energy (te) scenario, by 2025 and 2050, new and renewable energy would make up 23% and 31% of the primary energy mix, respectively, reducing ghg emissions per capita. according to predictions, annual ghg emissions per capita will decline from the bau scenario's 4.48 tonne co2eq/capita in 2050 to the te scenario's 4.1 tonne. keywords: sustainability; decarbonization; nre; ghg emissions; leap. @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 19th nov 2022; revised: 15th january 2023 accepted: 8th february 2023; available online: 22nd feb 2023 1. introduction sustainability is made up of three different parts: protecting the environment, keeping the economy going, and helping people (oyedepo, 2012). energy is connected to these three things in a roundabout way. almost every part of the economy, such as industry, transportation, household, commercial, and other sectors, depends on energy resources. from the energy supply side, both fossil fuels and renewable energy come from the environment, and most of the waste from energy processes (making, moving, storing, and using energy) goes into the environment (oyedepo, 2012). energy sustainability provides an opportunity to change society and grow the economy. it is also necessary to meet the growing demand for energy and reduce the world's carbon footprint (khan et al., 2022). the five primary objectives of a sustainable energy perspective are reduced carbon dioxide emissions, no detrimental ecological impacts, increased energy transformation reliability, lower costs of energy production, and enormous adoption of renewable energy technology (vidadili et al., 2017);(adewuyi et al., 2020). these goals for energy sustainability are called the "energy trilemma": making energy affordable and easy to get, making sure energy is secure, and * corresponding author email: yudiartono2013@gmail.com (y.yudiartono) making sure it doesn't hurt the environment. this will create a solid foundation for economic growth and competitiveness (khan et al., 2022). the field known as "energy security and transition" addresses problems pertaining to the accessibility of energy as well as the proportion of sources of clean energy in the total energy supply (dialga, 2021). the advancement of sustainable energies has a major influence on the entire nation, including the generation of energy from domestic sources, new local employment opportunities, and an increase in knowledge and level of expertise in specific green energy technologies. this is in addition to achieving independence, energy security, and reducing greenhouse gas emissions (dialga, 2021). further, the purpose of energy transition mechanisms is to reduce global warming by eliminating fossil fuels, especially coal in the near future, decarbonizing electricity generation, and electrifying end-use sectors as essential steps toward decarbonizing the entire energy system (vidinopoulos et al., 2020). thus, it is necessary to switch toward greener forms of energy generation, like renewables. the transition to renewable energy sources will be time-consuming, but it will have farreaching benefits for the planet and its inhabitants (lin et al., 2022). review/research article https://doi.org/10.14710/ijred.2023.50361 https://doi.org/10.14710/ijred.2023.50361 mailto:yudiartono2013@gmail.com https://orcid.org/0000-0001-6348-7838 https://orcid.org/0000-0003-1042-5064 https://orcid.org/0000-0002-0999-5556 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.50361&domain=pdf y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |420 issn: 2252-4940/© 2023. the author(s). published by cbiore the government of indonesia's policy regulation for energy development at the national level (ken, no. 79/2014) lays out the directions and actions needed to speed up the energy transition by promoting new and renewable energy (nre). by 2025, nre will make up at least 23% of the primary energy mix, and by 2050, it will make up at least 31% of the primary energy mix (goi, 2014). ken has been converted into law no. 22/2017 of the president on the planning scheme of national energy in order to make it operational (goi, 2017). without non-commercial biomass, the overall sum of primary energy produced in 2020 is 1,440.2 million boe. this is made up of 38.5% coal, 32.8% oil, 17.4% natural gas, and 11.2% nre. the goal set by the general plan of national energy (ruen), which is 14%(goi, 2017), has not been met for the nre share. in addition, primary energy rose at a moderate annual rate of 2.97 percent from 2010 to 2020 (kesdm, 2021). this fundamental energy is utilized to generate final energies such as electricity and fuel oil (refinery products). the national installed capacity and electricity production in 2020, including both on-grid and off-grid systems, are 72.81 gw and 299.96 twh, respectively (kesdm, 2021). the coal-fired power plants accounted for 47.53% (34.61 gw), followed by gas-fired power plants (29.53%, 21.72 gw), oil-fired power plants (8.20%, 5.97 gw), and the remainder are renewable energy-based power plants (14.4 percent, 10.52 gw) (kesdm, 2021), which is less than the general plan of national energy (ruen) objective of 16.2 gw(goi, 2017). it is evident that indonesia continues to rely largely on coal-fired power facilities. coal produces the highest greenhouse emissions and the least energy per thousand tons. it's the cheapest fuel, but renewables are growing (rahman et al., 2021). hydropower (8.4%, 6.14 gw), geothermal power (2.9%, 2.13 gw), biomass power (2.4%, 1.78 gw), solar power (0.3%, 0.185 gw), wind power (0.2%, 0.154 gw), and other renewables (0.2%, 0.124 gw) make up the 14.45% of power plants that use renewable energy (kesdm, 2021). hydropower and geothermal power make up most of the installed renewable energy capacity, while most of the other renewable energy sources are not very well developed. in 2020, biomass, solar, and wind made up less than 3% of the installed capacity, with off-grid applications making up the majority (sani et al., 2021). even though new policies were made to speed up the development of solar power, only 49 mw and 39.5 mw of capacity were added in 2019 and 2020, respectively. the first wind farm in indonesia opened in july 2018, and another one opened in 2019. together, they have a total installed capacity of 154.3 mw (sani et al., 2021). indonesia relies heavily on oil, coal, natural gas, lpg, and electricity for its final energy needs. in the years between 2010 and 2020, total energy consumption increased at a pace of 2.4% (excluding non-commercial biomass) (kesdm, 2021). the final energy consumption of the transportation and household sectors has a substantial impact on energy subsidies. the transportation sector uses a lot of oil-based fuels, so it's a big part of indonesia's energy mix. however, indonesia has been a net oil importer since 2004. over 60% of the oil that indonesia uses goes to this sector, and 70% of that goes to road transportation. the indonesian government's subsidization of gasoline and diesel fuels is a major issue for road transportation energy. if the policy doesn't change, rising demand for those fuels will lead to more subsidies, which will worsen as domestic oil supply declines (deendarlianto et al., 2017). so, it's important to look into other options, like electric vehicles (evs) and biofuel, for bringing in new types of energy or fuel for road transportation. indonesia's presidential regulation 55/2019 promotes evs adoption (goi, 2019). the evs regulation is an umbrella for future derivative legislation. norway, iceland, sweden, and the netherlands have above 5% ev penetration. ev market share exceeds 20% in some chinese and us cities. indonesia could apply lessons learned and best practices to its circumstances (adiatma & marciano, 2020). meanwhile, the residential lpg subsidy puts a strain on indonesia's budget. according to the 2019 state budget of indonesia, the lpg subsidy for 3 kg cylinders will cost more than $4.9 billion in 2019. indonesia, on the other hand, is building many new power plants with up to 35,000 mw of capacity (megawatt). due to the slowdown in economic growth, the added capacity will cause power overcapacity. thus, indonesia's induction stove program converts households to electric cooking to use excess power and reduce lpg subsidies (hakam et al., 2022). induction cooker use is encouraged in several nations. ecuador's national initiative replaced three million gas burners with induction stoves. energy imports account for 80% of ecuador's residential gas use. ecuador's lpg-to-induction stove program failed for many reasons. due to high start-up costs and the incorrect belief that induction stoves raise electricity bills, the 2018 target of 3.5 million homes adopting induction stoves was not attained (gould et al., 2020). moreover, in himachal pradesh, one of india's most electrified states, over 4000 rural households received induction stoves for clean cooking. electricity mostly displaced lpg, a supplementary cooking fuel, but not mud stoves. 5% of households switched to electric cooking (banerjee et al., 2016). this paper aims to analyze sustainable transition strategies by decarbonizing long-term energy demand and supply in indonesia. on the demand side, by taking into account the program of electric vehicles and biofuels for the transportation sector and induction stoves and urban gas networks for the household sector, with the base year 2020, this energy demand will be projected per type (fuel, electricity, gas, lpg, coal, biofuel, biomass). on the supply side, projections of primary energy supply (fossils and nre) will be carried out by optimizing the use of nre plants and increasing the share of nre in the national energy mix based on the provisions of the national energy policy. the resulting co2 emissions over the study's time horizon were also investigated. the analysis is carried out with the help of a low emissions analysis platform (leap), and the emission factors are based on ipcc tier 1, which is already outfitted with leap software (indrawan et al., 2017). 2. materials and methods 2.1 research framework the study sustainable long-term energy supply and demand: indonesia's gradual transition to a new and renewable energy system by 2050 requires an analytical framework, as shown in figure 1. so the study's goals can be realized. the evaluation of the final energy demand projection and the primary energy supply projection make up the analysis framework. economic growth grow (edi hilmawan et al., n.d.), population expansion (bps ri, 2018), policy, and a roadmap for energy development are used to calculate predicted energy demand. the use of different energy sources and the potential of energy resources, as well as the rules that are already in place and the progress of energy technology, are also taken into account in the analysis of the primary energy supply (national energy council, 2019). y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |421 issn: 2252-4940/© 2023. the author(s). published by cbiore macro assumptions of economic and population growth historical data on energy demand base year 2020 final energy intensity (energy consumption per unit of activity) transport sector : electric vehicles and biofuels household sector : induction stove commercial sector industrial sector projection of energy demand sectoral energy demand per type of fuel energy supply projection fossil new and renewable energy model leap energy supply to meet energy demand oil & gas, coal hydro pp, geothermal pp, solar pp, wind pp, biomass pp, cofiring coal pp, biogas pp, nuclear pp primary energy mix ghg emission projection technology options energy price energy transition renewable mix target government roadmap and policy historical data on energy supply base year 2020 sources : (edi hilmawan et al., n.d.),(national energy council, 2019) fig. 1 framework analysis on sustainable long-term energy supply and demand this study must consider roadmaps and government policies regarding the demand and supply of each type of energy, such as the energy transition roadmap towards carbon neutrality (ministry of energy and mineral resources, 2021), the business model for electricity provision (pt.pln (persero), 2021), the legally required biofuel development program mandated by memr regulation no. 12/2015 (kesdm, 2015), the development of vehicles powered solely by electricity in indonesia based on the quantitative target for the development of the national motor vehicle industry from the ministry of industry (direktur jenderal industri logam, mesin, alat transportasi, dan elektronika, 2021), and the mega conversion from lpg to induction stove (ministry of energy and mineral resources, 2021);(hakam et al., 2022). for figuring out how certain technologies affect energy demand and supply, both new and renewable energy (nre) technologies for the electrical sector and new and improved end-use technologies that need less energy are taken into account. to meet the target for the renewables mix, solar power, geothermal power, hydropower, pumped hydro storage, biomass power, biomass co-firing at coal-fired power plants, and nuclear power are all being studied. (pt.pln (persero), 2021). for this research, the energy sector in indonesia is modeled with an application software called leap (low emissions analysis platform), and the next energy modeling system for optimization (nemo), which is tightly integrated with leap, was used to calculate the least-cost optimization (heaps, 2021); (charles heaps, eric kemp-benedict, 2021);(ordonez et al., 2022). here are some of the leap studies undertaken in the asia area. sani et al. (sani et al., 2021) compared the role of modern renewables in the sumatran power sector to the indonesian government's plans. ordonez et al. (ordonez et al., 2022) examined cost-optimized methods for integrating greater proportions of renewable energy into the indonesian power grid. handayani et al. (handayani et al., 2019) studied long-term renewable energy scenarios with technological learning in mind for the java-bali electrical system development in indonesia. kachoee et al. (kachoee et al., 2018) assessed the effects of various energy policies on electricity demand and supply, the possibility of lowering greenhouse gas emissions, and the cost to iranian society. using renewable energy resources, kumar and madlener (kumar & madlener, 2016) examined india's future sustainable electricity supply. considering the baseline scenario, renewable sources, optimum clean coal, and energy efficiency and conservation cases, (mirjat et al., 2018);(tanoto et al., 2021) used leap to evaluate 2050 pakistan's long-term electricity distribution side cases. 2.2 scenario a scenario is a set of assumptions used to run the model, whose implications are assessed for the entire energy system. this paper compared the "business as usual" (bau) scenario to the "energy transition" (te) scenario. for both scenarios, the gdp growth was assumed to be the same. the bau scenario refers to the current state and predictions up to 2050 based on data from the past several years, such as additional city gas, electric stoves, electric automobiles, and others. additionally, the business plan for electricity provision (ruptl 2021–2030) guides power plant development. the te scenario refers to the adoption of electric vehicles (evs), induction cooktops, and urban gas networks (jargas) in accordance with the roadmaps developed by the ministry of industry for the development of electric vehicles (direktur jenderal industri logam, mesin, alat transportasi, dan elektronika, 2021) and the ministry of energy and mineral resources for the energy transition roadmap toward carbon neutrality (ministry of energy and mineral resources, 2021). when deciding how much the energy system needs to grow, the nre goal of the national energy policy for 2014 (ken 2014) is taken into account(goi, 2014). in both scenarios, a nuclear power plant is anticipated to be in operation by 2035. under the bau and te scenarios, nuclear pp capacity in 2050 would be 1 gw and 4 gw, respectively. 2.3 macro assumptions the following are some of the key assumptions that were used as input for the model: population growth (bps ri, 2018), gross domestic product (gdp) (edi hilmawan et al., n.d.), sales of evs under the bau scenario (edi hilmawan et al., n.d.),(pt.pln (persero), 2021) and te scenario (kementerian perindustrian, 2020), sales of induction stoves, and the development of gas networks in urban areas (jargas) (ministry of energy and mineral resources, 2021) for household demand. these assumptions are outlined in tables 1 and 2, respectively. table 1 assumptions on gdp and population growth macro assumptions year 2020 2025 2030 2040 2050 population (million) 270.2 283.1 294.8 313.3 324.6 population growth (%/year) 1.16 0.94 0.81 0.54 0.3 gdp growth (%/year) -2.07 5.17 5.5 5.5 4.5 source : (yudiartono et al., 2022) y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |422 issn: 2252-4940/© 2023. the author(s). published by cbiore table 2 programs considered in each scenario programs scenario year 2020 2025 2040 2050 electric car sales (thsnd) bau 0.26 42.28 420 885 te 0.26 400 1500 2800 electric motorcycle sales (thsnd) bau 2.01 311 1944 3042 te 2.01 1760 4375 7875 induction stove (mill. household) bau 0 0.3 1.2 2.2 te 0 8.2 38.2 48.2 jargas (mill. household) bau 0.267 2.5 6.7 10.3 te 0.267 5.2 20.3 23.4 note: jargas is an urban gas network source: (yudiartono et al., 2022) all scenarios assume the same population growth and economic growth rates between 2020 and 2050. in the bau scenario, the number of electric vehicles (evs) will reach approximately 42.3 thousand for electric cars and 311 thousand for electric motor cycles in 2025, according to outlook energy indonesia 2021 and ruptl's program. in the te scenario, on the other hand, the domestic automotive industry is expected to produce 400 thousand electric cars and 1,760 thousand motorcycles by 2025, according to a goal set by the ministry of industry for the development of electric vehicles (see table 2). the ministry of industry's electric vehicle production goal only extends to 2035. consequently, the number of evs manufactured between 2036 and 2050 is based on the prior growth history of electric vehicles. indonesia's induction stove initiative converts homes to electric cooking to utilize excess power and lower lpg subsidies. the scheme will replace lpg stoves in 8.2 million households by 2025 and 48.2 million by 2050 with induction stoves (ministry of energy and mineral resources, 2021). 3. empirical results and discussions 3.1 final energy demand projection because of ongoing economic and demographic factors, energy demand is predicted to keep going up. in the time frame under consideration, both the economy and the population would expand at an annualized rate of 5.03% and 0.633%, respectively. table 3 indicates that overall final energy demand (excluding non-commercial biomass) would ascend from 845.26 million boe in 2020 to 2,846.19 million boe and 2,593.20 million boe in 2050 under the bau and te scenarios. the variations in fossil fuel consumption by fuel type from 2020 to 2050 are presented in table 3. in the bau scenario, oil fuels and lpg consumption climbed from 222.79 million boe and 69.72 million boe, respectively, to 659.93 million boe and 115.78 million boe. in 2050, oil fuels and lpg consumption under the te scenario are expected to be 340.37 million boe and 29.5 million boe, respectively, a reduction of 319.56 million boe and 86.28 million boe as compared with the bau scenario. these decreases would be substituted by an increase in electricity consumption; by 2050, its consumption would reach 809.03 million boe (1,337.14 twh) under the te scenario compared to 674.46 million boe (1,114.73 twh) for the bau scenario. such an increase is a consequence of electric vehicle deployment in the transportation sector and the planned switchover from liquefied petroleum gas to induction stoves in the household sector. table 3 projection of final energy demand by type (million boe) energy type scenario year 2020 2025 2050 electricity bau 162.16 206.33 674.46 te 162.16 221.54 809.03 oil fuels bau 222.79 270.31 659.93 te 222.79 241.29 340.37 gas bau 97.48 110.32 345.87 te 97.48 113.44 364.15 coal bau 113.60 125.45 381.47 te 113.60 125.45 381.47 lpg bau 69.72 82.02 115.78 te 69.72 69.29 29.50 biofuel bau 179.32 222.3 667.65 te 179.32 222.3 667.65 biogas bau 0.18 0.39 1.03 te 0.18 0.39 1.03 total bau 845.26 1017.11 2846.19 te 845.26 993.69 2593.20 on the other hand, coal is used in manufacturing (cement, paper, textiles, etc.) and base-load power generation. its use is growing at an average rate of 4.1% per year, but it is still used much less than electricity. in the last 10 years, the government has made cpo-based biofuels, which are mixed with diesel fuel to make b30 (30% biodiesel and 70% diesel). biodiesel is a renewable alternative to diesel fuel with lower co2 emissions because it is carbon neutral. this is good for the environment and long-term growth (deendarlianto et al., 2020). under the bau and te scenarios, it is expected that the final energy demand for biofuel would go from 179.32 million boe in 2020 to 667.65 million boe in 2050, which is almost 4 times what it was in 2020. there are many different applications for biofuels, including in transport, industries, commerce, and even power plants. lack of access to sufficient raw materials prevents widespread usage of bioethanol. 3.1.1 the effect of the electric vehicle on energy demand the transportation sector comprises road, rail, air, and water modes, which utilize the final energy of oil fuels. gas, or cng, is only used for road transportation, and electricity is for electric rail transportation and electric vehicles. electric vehicles under the bau scenario refer to current conditions, where the sales percentage of electric vehicles in the future will not be as great as it is in the te scenario. the 2020 ministry of industry regulation no. 27 outlines a strategy for the development of electric vehicles. this is the subject of the te scenario. figure 3 demonstrates that the mass-scale deployment of electric vehicles in the te scenario can reduce the overall final energy demand during the whole study period. in 2050, overall energy usage is projected to be 1,306.77 million boe under the bau scenario, compared to 1,078.44 million boe under the te scenario, a decrease of 228.34 million boe. this is because the percentage of electric vehicle sales in the bau scenario is only 29.5%, much lower than the te scenario, which reaches more than 90%. besides that, electric vehicles use less energy than cars that run on gasoline (internal combustion engines). more than 77% of the grid's electricity is transformed into mechanical energy by the motors of evs (us department of energy 2020). y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |423 issn: 2252-4940/© 2023. the author(s). published by cbiore note: other oil fuels include avtur, avgas, fuel oil, and kerosene fig. 3 projection of final energy demand by type for the transportation sector most gasoline-powered cars turn between 12 and 30% of the energy in the gasoline into power at the wheels. electric vehicles, on the other hand, could be three times more energy efficient than cars with internal combustion engines (us department of energy 2020). as shown in figure 3, the higher ownership rates of electric vehicles under the te scenario than in the bau scenario, would boost the transportation sector's electricity demand. the demand for electricity would increase from 10.54 million boe (17.42 twh) in 2025 to 137.36 million boe (227.02 twh) in 2050 at a pace of about 11% annually. meanwhile, according to the general plan of national energy (ruen), the transportation sector will use 2.3 twh and 31.6 twh of electricity in 2025 and 2050, respectively. this is accomplished with a 2025 goal of 2,200 units of four-wheeled electric vehicles, which is far lower than the government's goal of 400,000 units. unfortunately, the 2050 goal for ruen electric vehicles is not specified. optimizing the electrification of road transportation also reduces gasoline and diesel oil consumption: in the te scenario, gasoline and diesel oil consumption in 2050 are 232.77 million boe and 19.71 million boe, respectively, compared to 544.09 million boe and 27.95 million boe in the bau scenario. the fall in the transportation sector's demand for gasoline and diesel oil would significantly reduce its reliance on imported fossil fuels. as a result, the greatest impact on oil consumption is from the adoption of innovative energy-efficient devices, such as electric vehicles in the transportation sector. fig. 4 projection of final energy demand by type for the household sector (excluding non-commercial biomass) 3.1.2 the effect of migration program for lpg to induction stoves and city gas networks on energy demand under the te scenario, household energy demand (excluding non-commercial biomass) is predicted to increase from 142.13 million boe in the base year to 316.96 million boe in 2050, a drop of 24.65 million boe from the bau scenario. this is because induction stoves are more efficient (81.78%) than lpg stoves (45.06%) (hakam et al., 2022). the lpg to electricity conversion program would change indonesian households' fuel composition in cooking activities. figure 4 depicts how, in 2050, lpg consumption would fall from 103.26 million boe (30.23%) in the bau scenario to 16.98 million boe (5.36%) in the te scenario. these decreases would be offset by increased electricity consumption, from 222.85 million boe (65.24%) to 266.2 million boe (83.99%) under bau and te scenarios, respectively. the demand for natural gas also continues to increase; by 2050, its consumption will reach 14.47 million boe (4.23%) under the bau scenario. due to the development of gas pipeline networks in urban areas, the consumption of such natural gas would rise considerably to 32.75 million boe (10.33%) in the case of the te scenario. 3.2 projection of primary energy supply indonesia's principal energy sources include crude oil, coal, natural gas, and new and renewable energies (hydropower, geothermal, solar, wind, biomass, bioenergy, and nuclear). in industry, coal is also used as a final source of energy, and natural gas is used as a final source of energy in industry, residential, and commercial sectors (minister for environment and forestry, 2021). the primary energy supply is projected to increase by 4.0% per year under the bau scenario and 4.2% per year under the te scenario between 2020 and 2050. figure 5 displays the forecasted distributions of primary energy sources. in both cases, coal is the primary fossil fuel used to power utilities. as a result of the transportation industry, oil's importance remains high. as more people switch to electric vehicles, demand for primary energy sources, particularly coal in the te scenario, will rise. 3.3 comparison of primary energy mix between bau scenario and te scenario in 2020, under the bau scenario, the nre mix only accounted for 11.2% of total primary energy supply. by 2025, the share of nre is estimated to be only 18.13%, and by 2050, it will be 20.79%, which is still quite far from the target of national energy policy regulation 79/2014 (ken). fig. 5 projection of primary energy supply 364.33 452.95 433.02 1,306.77 1,078.44 0 200 400 600 800 1000 1200 1400 base year bau te bau te 2020 2025 2050 m il li o n b o e gas other oil fuels biofuel diesel oil gasoline electricity total 142.13 177.98 174.48 341.60 316.96 0 50 100 150 200 250 300 350 400 base year bau te bau te 2020 2025 2050 m il li on b o e electricity natural gas kerosene lpg biogas total 1,448 1,742 1,778 2,268 2,358 2,787 2,916 3,392 3,548 4,025 4,215 4,705 4,933 500 1,000 1,500 2,000 2,500 3,000 3,500 4,000 4,500 5,000 bau te bau te bau te bau te bau te bau te 2020 2025 2030 2035 2040 2045 2050 m il li o n b o e oil gas coal nre total y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |424 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 6 primary energy mix projection using optimal electric vehicles and applying induction stoves for the household sector in the te scenario will increase nre use in the power plant sector and reduce oil and lpg usage in the transportation and household sectors. hence, the role of nre will increase sharply to 23.09% in 2025 and 31.33% in 2050, as shown in figure 6. this is the same as the provisions in nep article 9, which states that the role of nre will be at least 23% in 2025 and at least 31% in 2050, provided that they are economically fulfilled. increased use of renewable energy sources would have numerous positive effects, including stimulating the economy and creating new job opportunities, decreasing reliance on energy imports (ulrike lehr, 2017), fostering sustainable growth in rural areas (clausen & rudolph, 2020);(potrč et al., 2021), improving people's health (ferroukhi et al., 2016);(potrč et al., 2021) and lowering greenhouse gas emissions and air pollution (potrč et al., 2021). transitioning to a greener and more efficient energy system in indonesia would also improve energy supply security (azzuni et al., 2020);(potrč et al., 2021). 3.4 power sector profile to put it another way, the government is making a concerted effort to reduce its reliance on coal by increasing the proportion of nre in the electrical sector. in indonesia, there is an abundance of choice when it comes to the generation of power from renewable sources of energy. solar photovoltaics, hydropower, geothermal energy, and biomass are the most prominent potentials that have the potential to be further developed in indonesia. in the te scenario, the renewable capacity grows to 28.14 gw in order to meet the nre goal of 23% by 2025, which is part of the national energy policy. included are hydropower (11.17 gw), solar power (7 gw), geothermal power (3.58 gw), and other nre (6.39 gw), which includes biomass, wind, biogas, and a waste-to-energy plant. in 2050, the system's capacity will reach 318.38 gw, with 116.93 gw coming from nre (shown in figure 7). this means that the national energy policy's nre goal of 31% would be met. solar power (40 gw) and hydropower (38.47 gw) would make up most of the new and renewable electricity capacity. geothermal power (10 gw) would be next, and the rest, 24.45 gw, would come from other nre, with 18.67 gw coming from biomass power. co-firing biomass with coal in power plants would use 10.88 million tons of biomass in 2025 and 36.35 million tons in 2050. in the java-bali or west kalimantan systems, nuclear power plants with a capacity of 1 gw are expected to be built in 2035. by 2050, a total of 4 gw of nuclear power will be in place. by 2025, 23% of the general plan of national energy (ruen)'s electricity will be renewable. renewable energy power plants will account for 45.2 gw of the 135.5 gw total capacity in 2025. with 167.6 gw from nre, the ruen's 2050 target is 443.1 gw. note : other nre include biomass, wind, biogas, and waste to energy plant fig. 7 power sector profile the national energy program's 31% nre objective will be met. the target of the ruen power plant's installed capacity is higher than our study results. due to ruen's strong economic growth forecasts, which average 7.36 percent from 2020 to 2050, the covid-19 pandemic has been considered in our research, resulting in a 3.72 percent average yearly growth rate between 2020 and 2050. it is further noticed from figure 7 that, under the te scenario, the total amount of electricity produced in 2050 would amount to 1,494.76 twh. the generation mix would consist of fossil fuels (62.07%) and nre sources (37.93%), such as hydro (14.26%), other nre sources (10.25%), solar (5.86%), geothermal (5.55%), and nuclear (1.95%). indonesia possesses forty percent of the world's geothermal resources, with infinite resources and reserves of 23,766 mwe and reserves of at least 14,422 mwe (kesdm, 2021). geothermal development has been hindered by high-risk exploration, forestry permission requirements, and tariff disputes. only a minor portion of the entire proven reserves, or 2,130.7 mw, had been consumed by 2020 (kesdm, 2021). furthermore, geothermal power plants take several years to develop during the discovery phase. it takes a minimum of seven years for a conventional geothermal power station in indonesia to begin generating electricity (maulidia et al., 2019), which is why only 10 gw of geothermal power will be developed in 2050—much less than the 40 gw of solar power. wind and solar power have gotten cheaper to the point where they are increasingly being used in new power plant construction (burke et al., 2019). recent studies by irena show that the global weighted average lcoe (levelized cost of electricity) for photovoltaics on a large scale dropped from usd 0.381/kwh in 2010 to usd 0.057/kwh in 2020, a drop of 85%. utility-scale solar pv is now competitive with the least-priced new fossil fuel-fired capacity thanks to a considerable drop in costs. onshore wind farms have contributed to a 56% drop in the global weighted average price of power over the same time period, from usd 0.089/kwh to usd 0.039/kwh (irena, 2021). the indonesian memr (ministry of energy and mineral resources) estimates a 208 gw solar energy potential, of which only about 0.07% has been exploited thus far (barran & setiawan, 2022). solar power is one of the most effective means to reach the goal of using a larger share of green energy sources. indonesia is constructing pv roofs, solar farms, and floating solar with an installed capacity target of 37.15 gw (barran & setiawan, 2022). in addition, according to the findings of our research, solar power results in a rapid capacity expansion, 0% 20% 40% 60% 80% 100% bau te bau te bau te bau te bau te bau te 2020 2025 2030 2035 2040 2045 2050 oil gas coal nre 200 400 600 800 1,000 1,200 1,400 1,600 50 100 150 200 250 300 350 base year bau te bau te 2020 2025 2050 e le ct ri ci ty p ro d . (t w h ) c a p a ci ty ( g w ) fossil pp hydro pp geothermal pp solar pv pp other nre pp nuclear pp total elect. prod. y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |425 issn: 2252-4940/© 2023. the author(s). published by cbiore reaching 40 gw by 2050, which is more than 30 gw higher than the bau scenario. meanwhile, in the te scenario, wind power is expected to have a small role, growing to 2.5 gw by 2050. on average, around the globe, indonesia is one of the countries with the least amount of wind. as a result, wind power is not now at the forefront of indonesia's efforts to change its energy system (langer et al., 2022). in addition, under the te scenario, hydropower reaches 38.47 gw, with mini-hydropower contributing 30%, pumped hydro energy storage providing 18%, and large hydropower accounting for the remaining 7%. delivering power when demand is high and storing it when generation is high, pumped storage hydropower can guarantee the stabilization of the electrical grid and the energy time-shifting of renewable energy intermittency (ali et al., 2021). the relocation of many villages is a necessary social cost for large hydropower projects. in light of this, the development of large-scale hydro projects will require additional time and effort to adhere to tighter social and environmentally friendly requirements (maulidia et al., 2019). expanding biomass-based power facilities is met with similar and more overt obstacles. to generate power through biomass combustion, a large area of land will be required for the cultivation of biomass raw materials (maulidia et al., 2019). on the other side, small-scale power plants that run off of local biomass can be constructed, and these plants can then be positioned close to demand centers on indonesia's many islands. biomass resources include things like residual forest waste, cultivation of woody and non-woody biomass, animal waste, wastewater, and urban solid waste (wahono et al., 2022). in this study, biomass power reaches 18.67 gw in the te scenario and 16.35 gw in the bau scenario. this is a 12% increase in installed capacity. in particular, nuclear power is seen as a low priority by the indonesian government, which has labeled it an "option of last resort." accordingly, the energy market in indonesia is not currently receiving any nuclear power supplies. indonesia continues to exhibit interest in nuclear power despite the fact that it is not a primary policy choice. this is due to nuclear power being a carbon-free electrical option (cho et al., 2021). based on the results, the first 1 gw of nuclear power would be put in place in 2035, and by 2050, the total would reach 4 gw. a small portion of the national energy council (den) seems to be against building nuclear power plants, but the majority of the council has agreed to start nuclear development soon because it will help reduce emissions and make sure there is enough energy for everyone (maulidia et al., 2019). due to the high ownership rates of electric vehicles and induction stoves, additional power plant capacity is required to meet the increased electrical demand. it is anticipated that there will be an extra 89.05 gw of power plant capacity in 2050 as a result of these programs, with 55.29 gw coming from nre and the remainder from fossil fuel power plants. the direct emissions of carbon dioxide and other pollutants into the air from road transportation can be greatly diminished if electric vehicles are widely adopted. in indonesia, where coal-fired power plants are prevalent, the demand for electric vehicles could increase co2 emissions. in such cases, the environmental advantages of electric vehicles will not be completely achieved (european environment agency, 2016). coal-fired power plants are still the most popular way to generate electricity, mostly due to their low cost and ability to provide a continuous supply. the outcomes of this study indicate that throughout the transition period, fossil fuels are still used, but they are gradually replaced by a more sustainable renewable energy system. nonetheless, this transitional period must be meticulously planned in order to meet energy demand while ensuring environmental sustainability (maulidia et al., 2019). 3.5 greenhouse gas emission ghg emissions will result from the use of fossil fuels and commercial biomass to meet the bau and te scenarios. the power, transportation, and industrial sectors are the top three contributors to greenhouse gas emissions by sector in 2050. the percentage of the power plant sector will exceed fifty percent, followed by the industrial and transportation sectors ( twenty percent each), the residential sector and oil refinery ( two percent each), and the commercial and other sectors ( one percent each). due to the electric vehicle and induction stove programs on the demand side and nre deployment on the supply side, total ghg emissions for the energy sector in 2050 decrease from 1,428.47 million tonnes co2eq under the bau scenario to 1,332.4 million tonnes co2eq under the te scenario, as shown in figure 8. carbon dioxide is released into the atmosphere during the burning of fossil fuels and the making of cement. the combustion of solid, liquid, and gaseous fuels, in addition to gas flaring, are all major contributors to this category of emissions (the world bank, 2020). numerous factors, including economic and demographic factors, technical advancements, institutional frameworks, ways of life, and international trade, influence carbon dioxide emissions (kachoee et al., 2018). according to the world bank, carbon dioxide emissions per capita in the world in 2020 were 4.5 metric tonnes, with a 0.85% increase from 1999 to 2019 (the world bank, 2020). figure 9 displays projected ghg emissions per capita from energy activities, showing that this number will decrease from 4.48 tonnes co2eq /capita in 2050 under the bau scenario to 4.1 tonnes co2eq /capita in the case of the te scenario, where high effect mitigation action such as nre has been largely implemented. as can be seen in figure 9, the two scenarios' emission projections are nearly identical up until the year 2023. beginning in 2023, when more aggressive mitigation steps are put into place, the trajectories begin to diverge from the bau trajectory. in 2025, under the te scenario, the proportion of new and renewable energy in the entire national energy mix would rise to 23%; by 2050, it would rise to 31%; this shift would lead to lower ghg emissions per capita. note : not considering fugitive oil & gas and fugitive mining fig. 8 ghg emission in the energy sector 0 200 400 600 800 1,000 1,200 1,400 base year bau te bau te 2020 2025 2050 m il li o n t o n n es c o 2 eq industry transportation household commercial other sector power plants oil refinery y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |426 issn: 2252-4940/© 2023. the author(s). published by cbiore fig. 9 ghg emissions per capita and per gdp moreover, despite a 6% increase in global carbon dioxide emissions and gdp in 2021, the average emissions intensity of global economic output remained unchanged at 0.26 tonnes of co2eq per $1000 of gdp (iea, 2022). meanwhile, ghg emissions per gdp in indonesia would continue to fall under the bau and te scenarios, from 0.67 tonnes of co2eq per $ 1000 in 2020 to 0.42 tonnes of co2eq per $ 1000 and 0.39 tonnes of co2eq per $ 1000 in 2050, respectively. 4. policy implications this paper analyzes the impact of indonesia's energy transition pathway through the electrification of end-use sectors and the decarbonization of power generation. under the energy transition (te) scenario, the nre mix in 2025 and 2050 would reach 23% and 31%, respectively, of the total national energy mix, with significant energy savings in the transportation and household sectors. solar power, hydropower, biomass power, geothermal power, and nuclear power would dominate indonesia's electrical mix in the foreseeable future. in addition, a rise in electricity production from nre would have a beneficial effect on the environment. either ghg emissions per capita or ghg emissions per gdp might be lowered by 7% under the te scenario compared to the bau scenario. the results of this research can be used as a basis for creating policies that will make indonesia's nre sector more sustainable. these policies should be able to address issues relating to social factors, the economy, technology, and government that get in the way of nre implementation. (moorthy et al., 2019). social barriers to renewable energy include not having enough skilled workers and buying up land that could have been used for farming, tourism, etc. a big problem with solar farms is that they need a huge amount of land to make the same amount of energy that a small coal-fired power plant can make (moorthy et al., 2019). the average amount of land needed for one solar module is about 2 m2, so it is expected that it will be hard to find the land needed in the future (barran & setiawan, 2022). indonesia has come up with floating solar on hydropower dams, solar power on land that used to be mined, and solar power on roofs as ways to deal with land problems. the high initial investment required, the paucity of financial institutions, the rivalry with fossil fuels, and the lower level of subsidies available in comparison to conventional fuels are all factors that contribute to the economic and financial challenges faced (moorthy et al., 2019). an agency for the management of environmental funds, known as bpdlh (badan pengelola dana lingkungan hidup), was set up by the indonesian government to break budget limits and ensure financial support for protecting and managing the environment, including taking steps to stop and adapt to climate change. carbon trading, soft loans, traditional grants, environmental subsidies, and government viability gap funding (vgf) support are the sources of money controlled by bpdlh (wahono et al., 2022). technology has also impeded the adoption of nres. problems include not having enough of the right technology, not knowing how to use and maintain it properly, not being able to do research and development, and the exorbitant expense of the technology (sambodo et al., 2022);(moorthy et al., 2019). as a result, the government must place a greater emphasis on renewables research and development in order to improve the technological capabilities of local renewables producers and, ultimately, lower renewable energy prices (iesr, 2018). in addition, the absence of a smart grid that integrates conventional and renewable energy sources and the restricted role of research and development in energy storage technologies are regarded as technological obstacles (sambodo et al., 2022). this prospective smart grid technology will promote access to healthier, cleaner, and more long-lasting resources (elrahmani et al., 2021), whereas the development of an energy storage system appears to be one option for addressing the intermittent nature of solar energy (wattana & aungyut, 2022). regarding regulatory impediments, inefficient government policies, inadequate economic incentives, and administrative and bureaucratic difficulties have hampered the rapid expansion of renewable energy (moorthy et al., 2019). therefore, for a nation's sustainable development, effective regulatory rules within the energy business are essential, and cooperation between different agencies should be strengthened. thus, indonesia needs national leadership that consistently exhibits a strong political commitment to encourage and support renewable energy. as a reference price, the government published a regulation based on the regional and national main production cost of generating power (bpp). the rule doesn't take into account tariff increases, which are demanded by private investors to offset future inflation (maulidia et al., 2019). by developing feed-intariff (fit) policies that favor renewable energy and are also adaptable to changes in technology and market conditions, it is possible to dismantle this unhelpful legislation (iesr, 2018). the period of 2021–2022 can be considered an important milestone for the development of nre. to date, the government has issued several essential regulations: the energy ministry’s regulation no. 26/2021 in terms of rooftop solar power systems; the business plan for the national electricity supply from 2021– 2030 (putriastuti & fitriyanti, 2022); the president of the republic of indonesia's regulatory order 112/2022 on the fast development of renewable energy sources for electricity production (goi, 2022); and the harmonization of tax regulation. in law no. 7 of 2021 concerning the harmonization of tax regulations, the government imposed a carbon tax (uu hpp). article 13 of the hpp law governs, among other things, the carbon tax imposed on carbon emissions that harm the environment (suryani, 2022). initially, the implementation of carbon trading, including the imposition of a carbon tax, was targeted for april 1, 2022, with a cap and trade and tax scheme. however, the policy's implementation was postponed due to inadequate infrastructure. the study implies that nres like solar power, hydropower, biomass power, geothermal power, and nuclear power are all potentially feasible possibilities for the generation of electricity. it is possible that the nre will be able to have a big effect on the economic, social, and environmental sustainability of the energy 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 t o n n es c o 2 e/ c a p it a t o n n es c o 2 e/ u s d 1 0 0 0 bau te bau te y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |427 issn: 2252-4940/© 2023. the author(s). published by cbiore industry. solar energy, hydropower, biomass, geothermal energy, and nuclear power are just a few of the alternative energy sources that will play a significant role in the continuing change. solar energy is inexpensive because, after the initial investment, the ongoing costs of producing solar energy are typically unnoticeable. because of its accessibility on a local level, solar power is an essential component of reliable energy sources. then, hydropower can be paired with wind, solar, and other sources to provide affordable, reliable grid electricity. in addition to irrigation, providing water for industry and homes, preventing flooding, and creating jobs, hydropower provides energy security by making less use of fossil fuels (kabeyi & olanrewaju, 2022). due to the huge potential of feedstocks in rural regions, where there is also a need to enhance access to inexpensive, dependable electricity and boost rural development, biomass power plants, like the ones on the mentawai islands, are a particularly appealing choice for indonesia (wong, 2021). biomass power plants can have a steady supply of feedstock for a long time if they use planting strategies that take into account both the community and nature. this is a way to encourage the construction of biomass power plants across the country (wahono et al., 2022), which according to this study, are projected to reach 18.67 gw by 2050. in addition, the most significant obstacle that geothermal electricity generation must overcome is the lengthy period of time required for project development, the high risks involved in the beginning of the process, and enormous expenditures for conventional systems, which also have poor efficiency in the process of converting electricity. geothermal power plants could be built faster and with less risk if they were wellhead power plants (kabeyi & olanrewaju, 2022). the use of nuclear power has a number of environmental benefits, the most important of which is that its greenhouse gas emissions during its entire lifecycle, including those associated with uranium mining and processing, are nearly equal to those of sustainable power plants. because of this, using nuclear power more will lead to less pollution and less damage to the environment (kabeyi & olanrewaju, 2022). 5. conclusion indonesia's nre goals are important because they show that indonesia is aware of nre as a possible way to get more people on the power grid, make sure there is enough power, and reduce environmental damage. in line with rule no. 79 of 2014 from the government on the national energy policy, under the te scenario, the renewable capacity grows to 28.14 gw to meet the national energy policy's nre target of 23% by 2025. this includes 11.17 gw of hydropower, 7 gw of solar power, 3.58 gw of geothermal power, and 6.39 gw of other nre (biomass, wind, biogas, waste-to-energy plants). in 2050, the system's capacity will reach 318.38 gw, with 116.93 gw coming from renewable and nuclear energy. this means that the national energy policy's nre goal of 31% would be met. by the end of the study, the capacity of solar power would have grown from almost nothing to more than 216 times its current level. hydropower's capacity would have increased by more than six times from its current level of 6.14 gw in 2020, while geothermal and biomass power capacities would have increased by over 5 times and more than 10 times, respectively. some renewable power plants, like solar and wind power, can only supply power for short periods of time. for a steady supply, these plants would need to be connected to a baseload power supply system, like hydropower plants or nuclear power plants. also, energy storage is a key part of a sustainable energy system because it helps solve the problem of intermittent renewable energy sources. pumped-storage hydroelectricity and bess (battery energy storage system) are two ways to store energy. also, ghg emissions per capita are expected to drop from 4.48 tonnes co2eq/capita in 2050 under the bau scenario to 4.1 tonnes co2eq/capita under the te scenario if high-impact mitigation actions like nre have been mostly put into place. in the meantime, ghg emissions per $1000 of gdp would fall from 0.42 tonnes co2eq per $1000 under the bau scenario to 0.39 tonnes co2eq per $1000 under the te scenario in 2050. solar power, hydropower, geothermal power, biomass, and nuclear power could all be important parts of indonesia's future power mix. nevertheless, in the short to medium term, coalbased transformations may be required before moving on to a more sustainable low-carbon energy source. this phase of change needs to be carefully planned so that energy needs are met while environmental impacts are kept to a minimum. the implementation of low-carbon development is hindered by the social, economic, technological, and regulatory obstacles that are crucial to the acceleration of nre deployment. for the deployment of new and renewable energy to be successful, big problems like a lack of skilled workers, land issues, expensive initial capital, not enough money for r&d, and ineffective government regulations should be dealt with as soon as possible. future research might concentrate on removing coal from the electricity sector as a way to get global carbon dioxide (co2) emissions to zero by 2050. when it comes to new coal power plants that have a long life ahead of them, thinking about a ccs (carbon capture and storage) retrofit in the future could be a good way to avoid the power plant's almost full depreciation. acknowledgments the authors would like to thank energy planning members at the national research and innovation agency, indonesia, for their assistance in modeling sustainable long-term energy supply demand with leap; in particular, mr. joko santosa, who has provided a lot of guidance and very useful discussions. conflicts of interest: the authors state that there is no conflict of interest. references adewuyi, o. b., kiptoo, m. k., afolayan, a. f., amara, t., alawode, o. i., & senjyu, t. (2020). challenges and prospects of nigeria’s sustainable energy transition with lessons from other countries’ experiences. energy reports, 6, 993–1009. https://doi.org/10.1016/j.egyr.2020.04.022 adiatma, j. c., & marciano, i. (2020). the role of electric vehicles in decarbonizing indonesia’s road transport sector (f. tumiwa (ed.)). institute for essential services reform (iesr). https://iesr.or.id/en/pustaka/the-role-of-electric-vehicles-indecarbonizing-indonesias-road-transport-sector ali, s., stewart, r. a., & sahin, o. (2021). drivers and barriers to the deployment of pumped hydro energy storage applications: systematic literature review. cleaner engineering and technology, 5, 100281. https://doi.org/10.1016/j.clet.2021.100281 azzuni, a., aghahosseini, a., ram, m., bogdanov, d., caldera, u., & breyer, c. (2020). energy security analysis for a 100% renewable energy transition in jordan by 2050. sustainability (switzerland), 12(12), 4921. https://doi.org/10.3390/su12124921 banerjee, m., prasad, r., rehman, i. h., & gill, b. (2016). induction stoves as an option for clean cooking in rural india. energy policy, 88, 159–167. https://doi.org/10.1016/j.enpol.2015.10.021 barran, i., & setiawan, e. a. (2022). pv rooftop repowering potential in indonesia: study case. iop conference series: earth and https://doi.org/10.1016/j.egyr.2020.04.022 https://iesr.or.id/en/pustaka/the-role-of-electric-vehicles-in-decarbonizing-indonesias-road-transport-sector https://iesr.or.id/en/pustaka/the-role-of-electric-vehicles-in-decarbonizing-indonesias-road-transport-sector https://doi.org/10.1016/j.clet.2021.100281 https://doi.org/10.3390/su12124921 https://doi.org/10.1016/j.enpol.2015.10.021 y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |428 issn: 2252-4940/© 2023. the author(s). published by cbiore environmental science, 997(1). https://doi.org/10.1088/17551315/997/1/012016 bps ri. (2018). proyeksi penduduk indonesia 2015-2045 hasil supas 2015. https://www.bps.go.id/publication/2018/10/19/78d24d90200 26ad95c6b5965/proyeksi-penduduk-indonesia-2015-2045hasil-supas-2015.html burke, p. j., widnyana, j., anjum, z., aisbett, e., resosudarmo, b., & baldwin, k. g. h. (2019). overcoming barriers to solar and wind energy adoption in two asian giants: india and indonesia. energy policy, 132, 1216–1228. https://doi.org/10.1016/j.enpol.2019.05.055 charles heaps, eric kemp-benedict, j. v. (2021). next energy modelling system for optimization (nemo). stockholm environment institute. https://www.sei.org/projects-and-tools/tools/nemothe-next-energy-modeling-system-for-optimization/ cho, i., oh, s., kim, s., ardin, f., & heo, e. (2021). determinants of nuclear power expansion in indonesia. nuclear engineering and technology, 53(1), 314–321. https://doi.org/10.1016/j.net.2020.06.008 clausen, l. t., & rudolph, d. (2020). renewable energy for sustainable rural development: synergies and mismatches. energy policy, 138, 111289. https://doi.org/10.1016/j.enpol.2020.111289 deendarlianto, widyaparaga, a., sopha, b. m., budiman, a., muthohar, i., setiawan, i. c., lindasista, a., soemardjito, j., & oka, k. (2017). scenarios analysis of energy mix for road transportation sector in indonesia. renewable and sustainable energy reviews, 70(2), 13–23. https://doi.org/10.1016/j.rser.2016.11.206 deendarlianto, widyaparaga, a., widodo, t., handika, i., chandra setiawan, i., & lindasista, a. (2020). modelling of indonesian road transport energy sector in order to fulfill the national energy and oil reduction targets. renewable energy, 146, 504–518. https://doi.org/10.1016/j.renene.2019.06.169 dialga, i. (2021). evaluating normandy ’ s sustainable development and energy transition policies. journal of cleaner production, 305, 127096. https://doi.org/10.1016/j.jclepro.2021.127096 direktur jenderal industri logam, mesin, alat transportasi, dan elektronika, k. (2021). pengembangan industri kendaraan bermotor listrik berbasis baterai (kblbb). kementerian perindustrian. edi hilmawan, fitriana, i., sugiyono, a., & adiarso (eds.). (n.d.). outlook energi indonesia 2021. pusat pengkajian industri proses dan energi (ppipe) , bppt. https://www.bppt.go.id/dokumen/file/865/download elrahmani, a., hannun, j., eljack, f., & kazi, m. k. (2021). status of renewable energy in the gcc region and future opportunities. current opinion in chemical engineering, 31, 100664. https://doi.org/10.1016/j.coche.2020.100664 european environment agency. (2016). electric vehicles and the energy sector impacts on europe’s future emissions. https://www.eea.europa.eu/publications/electric-vehicles-andthe-energy ferroukhi, r., lopez-peña, a., kieffer, g., nagpal, d., hawila, d., khalid, a., el-katiri, l., vinci, s., & fernandez, a. (2016). renewable energy benefits: measuring the economics. irena international renewable energy agency, 92. https://www.irena.org//media/files/irena/agency/publication/2016/irena_mea suring-the-economics_2016.pdf goi. (2014). government regulation no. 79/2014 on national energy policy-kebijakan energi nasional (ken). government of indonesia. https://jdih.esdm.go.id/peraturan/pp%20no.%2079%20thn% 202014.pdf goi. (2017). presidential decree no. 22/2017 regarding national energy general plan-rencana umum energi nasional (ruen). government of indonesia. https://www.esdm.go.id/assets/media/content/contentrencana-umum-energi-nasional-ruen.pdf goi. (2019). presidential decree no. 55/2019 regarding the acceleration of the battery-based electric motor vehicle (kblbb) program for road transportation-percepatan program kblbb untuk transportasi jalan. government of indonesia. https://jdih.esdm.go.id/storage/document/perpres%20nomor %2055%20tahun%202019.pdf goi. (2022). presidential decree no. 112/2022 on acceleration of renewable energy development for electricity power supply-percepatan pengembangan energi terbarukan untuk penyediaan tenaga listrik. government of indonesia. https://drive.esdm.go.id/wl/?id=o8wdm5f2axpp9awt2y4cf nvb3t2jdoaf gould, c. f., schlesinger, s. b., molina, e., bejarano, m. l., valarezo, a., & jack, d. w. (2020). household fuel mixes in peri-urban and rural ecuador: explaining the context of lpg, patterns of continued firewood use, and the challenges of induction cooking. energy policy, 136, 111053. https://doi.org/10.1016/j.enpol.2019.111053 hakam, d. f., nugraha, h., wicaksono, a., rahadi, r. a., & kanugrahan, s. p. (2022). mega conversion from lpg to induction stove to achieve indonesia’s clean energy transition. energy strategy reviews, 41, 100856. https://doi.org/10.1016/j.esr.2022.100856 handayani, k., krozer, y., & filatova, t. (2019). from fossil fuels to renewables: an analysis of long-term scenarios considering technological learning. energy policy, 127(september 2018), 134– 146. https://doi.org/10.1016/j.enpol.2018.11.045 heaps, c. g. (2021). leap: the low emissions analysis platform. [software version: 2020.1.47]. stockholm environment institute. somerville, ma, usa. https://leap.sei.org iea. (2022). co2 emissions intensity of gdp, 1990-2021. iea paris. https://www.iea.org/data-and-statistics/charts/co2-emissionsintensity-of-gdp-1990-2021 iesr. (2018). igniting a rapid deployment of renewable energy in indonesia: lessons learned from three countries. 1–92. https://iesr.or.id/wpcontent/uploads/2019/05/iesr_research_igniting-a-rapiddeployment-of-re-in-indonesia.pdf indrawan, n., thapa, s., rahman, s. f., park, j. h., park, s. h., wijaya, m. e., gobikrishnan, s., purwanto, w. w., & park, d. h. (2017). palm biodiesel prospect in the indonesian power sector. environmental technology and innovation, 7, 110–127. https://doi.org/10.1016/j.eti.2017.01.001 irena. (2021). world energy transitions outlook. in irena. https://irena.org/publications/2021/march/world-energytransitions-outlook kabeyi, m. j. b., & olanrewaju, o. a. (2022). sustainable energy transition for renewable and low carbon grid electricity generation and supply. frontiers in energy research, 9, 1–45. https://doi.org/10.3389/fenrg.2021.743114 kachoee, m. s., salimi, m., & amidpour, m. (2018). the long-term scenario and greenhouse gas effects cost-benefit analysis of iran’s electricity sector. energy, 143, 585–596. https://doi.org/10.1016/j.energy.2017.11.049 kementerian perindustrian. (2020). permenperin no. 27/2020 ttg spesifikasi, peta jalan pengembangan, dan ketentuan penghitungan nilai tkdn kblbb (battery electric vehicle). https://www.gaikindo.or.id/wpcontent/uploads/2020/11/permenperin_27_2020-tkdn-kblberbasis-baterai.pdf kesdm. (2015). permen esdm no. 12/2015 ttg penyediaan, pemanfaatan, dan tata niaga bbn (biofuel) sbg bahan bakar lain. https://jdih.esdm.go.id/peraturan/permen%20esdm%2012% 20thn%202015.pdf kesdm. (2021). handbook of energy and economic statistics of indonesia 2020 (a. c. adi & f. lasnawatin (eds.)). ministry of energy and mineral resources republic of indonesia. https://www.esdm.go.id/assets/media/content/contenthandbook-of-energy-and-economic-statistics-of-indonesia2020.pdf khan, i., zakari, a., zhang, j., dagar, v., & singh, s. (2022). a study of trilemma energy balance, clean energy transitions, and economic expansion in the midst of environmental sustainability: new insights from three trilemma leadership. energy, 248, 123619. https://doi.org/10.1016/j.energy.2022.123619 kumar, s., & madlener, r. (2016). co2 emission reduction potential assessment using renewable energy in india. energy, 97, 273– 282. https://doi.org/10.1016/j.energy.2015.12.131 langer, j., simanjuntak, s., pfenninger, s., laguna, a. j., lavidas, g., polinder, h., quist, j., rahayu, h. p., & blok, k. (2022). how offshore wind could become economically attractive in lowresource regions like indonesia. iscience, 25(9), 104945. https://doi.org/10.1016/j.isci.2022.104945 lin, c., yin, k., moslehpour, m., viet, h., dang, k., & quang, t. (2022). https://doi.org/10.1088/1755-1315/997/1/012016 https://doi.org/10.1088/1755-1315/997/1/012016 https://www.bps.go.id/publication/2018/10/19/78d24d9020026ad95c6b5965/proyeksi-penduduk-indonesia-2015-2045-hasil-supas-2015.html https://www.bps.go.id/publication/2018/10/19/78d24d9020026ad95c6b5965/proyeksi-penduduk-indonesia-2015-2045-hasil-supas-2015.html https://www.bps.go.id/publication/2018/10/19/78d24d9020026ad95c6b5965/proyeksi-penduduk-indonesia-2015-2045-hasil-supas-2015.html https://doi.org/10.1016/j.enpol.2019.05.055 https://www.sei.org/projects-and-tools/tools/nemo-the-next-energy-modeling-system-for-optimization/ https://www.sei.org/projects-and-tools/tools/nemo-the-next-energy-modeling-system-for-optimization/ https://doi.org/10.1016/j.net.2020.06.008 https://doi.org/10.1016/j.enpol.2020.111289 https://doi.org/10.1016/j.rser.2016.11.206 https://doi.org/10.1016/j.renene.2019.06.169 https://doi.org/10.1016/j.jclepro.2021.127096 https://www.bppt.go.id/dokumen/file/865/download https://doi.org/10.1016/j.coche.2020.100664 https://www.eea.europa.eu/publications/electric-vehicles-and-the-energy https://www.eea.europa.eu/publications/electric-vehicles-and-the-energy https://www.irena.org/-/media/files/irena/agency/publication/2016/irena_measuring-the-economics_2016.pdf https://www.irena.org/-/media/files/irena/agency/publication/2016/irena_measuring-the-economics_2016.pdf https://www.irena.org/-/media/files/irena/agency/publication/2016/irena_measuring-the-economics_2016.pdf https://jdih.esdm.go.id/peraturan/pp%20no.%2079%20thn%202014.pdf https://jdih.esdm.go.id/peraturan/pp%20no.%2079%20thn%202014.pdf https://www.esdm.go.id/assets/media/content/content-rencana-umum-energi-nasional-ruen.pdf https://www.esdm.go.id/assets/media/content/content-rencana-umum-energi-nasional-ruen.pdf https://jdih.esdm.go.id/storage/document/perpres%20nomor%2055%20tahun%202019.pdf https://jdih.esdm.go.id/storage/document/perpres%20nomor%2055%20tahun%202019.pdf https://drive.esdm.go.id/wl/?id=o8wdm5f2axpp9awt2y4cfnvb3t2jdoaf https://drive.esdm.go.id/wl/?id=o8wdm5f2axpp9awt2y4cfnvb3t2jdoaf https://doi.org/10.1016/j.enpol.2019.111053 https://doi.org/10.1016/j.esr.2022.100856 https://doi.org/10.1016/j.enpol.2018.11.045 https://leap.sei.org/ https://www.iea.org/data-and-statistics/charts/co2-emissions-intensity-of-gdp-1990-2021 https://www.iea.org/data-and-statistics/charts/co2-emissions-intensity-of-gdp-1990-2021 https://iesr.or.id/wp-content/uploads/2019/05/iesr_research_igniting-a-rapid-deployment-of-re-in-indonesia.pdf https://iesr.or.id/wp-content/uploads/2019/05/iesr_research_igniting-a-rapid-deployment-of-re-in-indonesia.pdf https://iesr.or.id/wp-content/uploads/2019/05/iesr_research_igniting-a-rapid-deployment-of-re-in-indonesia.pdf https://doi.org/10.1016/j.eti.2017.01.001 https://irena.org/publications/2021/march/world-energy-transitions-outlook https://irena.org/publications/2021/march/world-energy-transitions-outlook https://doi.org/10.3389/fenrg.2021.743114 https://doi.org/10.1016/j.energy.2017.11.049 https://www.gaikindo.or.id/wp-content/uploads/2020/11/permenperin_27_2020-tkdn-kbl-berbasis-baterai.pdf https://www.gaikindo.or.id/wp-content/uploads/2020/11/permenperin_27_2020-tkdn-kbl-berbasis-baterai.pdf https://www.gaikindo.or.id/wp-content/uploads/2020/11/permenperin_27_2020-tkdn-kbl-berbasis-baterai.pdf https://jdih.esdm.go.id/peraturan/permen%20esdm%2012%20thn%202015.pdf https://jdih.esdm.go.id/peraturan/permen%20esdm%2012%20thn%202015.pdf https://www.esdm.go.id/assets/media/content/content-handbook-of-energy-and-economic-statistics-of-indonesia-2020.pdf https://www.esdm.go.id/assets/media/content/content-handbook-of-energy-and-economic-statistics-of-indonesia-2020.pdf https://www.esdm.go.id/assets/media/content/content-handbook-of-energy-and-economic-statistics-of-indonesia-2020.pdf https://doi.org/10.1016/j.energy.2022.123619 https://doi.org/10.1016/j.energy.2015.12.131 https://doi.org/10.1016/j.isci.2022.104945 y. yudiartono et al int. j. renew. energy dev 2023, 12(2), 419-429 |429 issn: 2252-4940/© 2023. the author(s). published by cbiore factors influencing the sustainable energy technologies adaptation in asean countries. sustainable energy technologies and assessments, 53(pc), 102668. https://doi.org/10.1016/j.seta.2022.102668 maulidia, m., dargusch, p., ashworth, p., & ardiansyah, f. (2019). rethinking renewable energy targets and electricity sector reform in indonesia: a private sector perspective. renewable and sustainable energy reviews, 101, 231–247. https://doi.org/10.1016/j.rser.2018.11.005 minister for environment and forestry. (2021). indonesia long-term strategy for low carbon and climate resilience 2050 (indonesia lts-lccr 2050). ministry of environment and forestry. https://unfccc.int/sites/default/files/resource/indonesia_lts -lccr_2021.pdf ministry of energy and mineral resources. (2021). energy transition roadmap towards carbon neutral. https://www.esdm.go.id/en/media-center/newsarchives/road-map-to-carbon-neutrality-is-being-preparedsays-energy-minister mirjat, n. h., uqaili, m. a., harijan, k., walasai, g. das, mondal, m. a. h., & sahin, h. (2018). long-term electricity demand forecast and supply side scenarios for pakistan (2015–2050): a leap model application for policy analysis. energy, 165, 512–526. https://doi.org/10.1016/j.energy.2018.10.012 moorthy, k., patwa, n., & gupta, y. (2019). breaking barriers in deployment of renewable energy. heliyon, september 2018, e01166. https://doi.org/10.1016/j.heliyon.2019.e01166 national energy council. (2019). indonesia energy outlook 2019 (s. abdurrahman, m. pertiwi, & walujanto (eds.)). secretary general of the national energy council. https://www.esdm.go.id/assets/media/content/contentindonesia-energy-outlook-2019-english-version.pdf ordonez, j. a., fritz, m., & eckstein, j. (2022). coal vs. renewables: least-cost optimization of the indonesian power sector. energy for sustainable development, 68, 350–363. https://doi.org/10.1016/j.esd.2022.04.017 oyedepo, s. o. (2012). on energy for sustainable development in nigeria. renewable and sustainable energy reviews, 16(5), 2583– 2598. https://doi.org/10.1016/j.rser.2012.02.010 potrč, s., čuček, l., martin, m., & kravanja, z. (2021). sustainable renewable energy supply networks optimization – the gradual transition to a renewable energy system within the european union by 2050. renewable and sustainable energy reviews, 146. https://doi.org/10.1016/j.rser.2021.111186 pt.pln (persero). (2021). rencana usaha penyediaan tenaga listrik (ruptl) business plan for electricity provision 2021-2030. https://web.pln.co.id/statics/uploads/2021/10/ruptl-20212030.pdf putriastuti, m. a. c., & fitriyanti, v. (2022). a turning point for renewable energy in indonesia? https://www.purnomoyusgiantorocenter.org/a-turning-point-forrenewable-energy-in-indonesia/1–6. rahman, a., dargusch, p., & wadley, d. (2021). the political economy of oil supply in indonesia and the implications for renewable energy development. renewable and sustainable energy reviews, 144(march), 111027. https://doi.org/10.1016/j.rser.2021.111027 sambodo, m. t., yuliana, c. i., hidayat, s., novandra, r., handoyo, f. w., farandy, a. r., inayah, i., & yuniarti, p. i. (2022). breaking barriers to low-carbon development in indonesia: deployment of renewable energy. heliyon, 8(4), e09304. https://doi.org/10.1016/j.heliyon.2022.e09304 sani, l., khatiwada, d., harahap, f., & silveira, s. (2021). decarbonization pathways for the power sector in sumatra, indonesia. renewable and sustainable energy reviews, 150(july 2020), 111507. https://doi.org/10.1016/j.rser.2021.111507 suryani, a. s. (2022). preparation for carbon tax implementation in indonesia. in a brief study of actual and strategic issues: vol. vol.xiv. tanoto, y., haghdadi, n., bruce, a., & macgill, i. (2021). reliability-cost trade-offs for electricity industry planning with high variable renewable energy penetrations in emerging economies: a case study of indonesia’s java-bali grid. energy, 227, 120474. https://doi.org/10.1016/j.energy.2021.120474 the world bank. (2020). co2 emissions (metric tons per capita). climate watch. 2020. ghg emissions. washington, dc: world resources institute. https://data.worldbank.org/indicator/en.atm.co2e.pc ulrike lehr, p. u. (2017). economic impacts of renewable energy increase in germany. in towards 100% renewable energy (pp. 263–272). http://theargeo.org/november/files/economic%20impacts%2 0of%20renewable%20energy%20promotion%20in%20german y.pdf vidadili, n., suleymanov, e., bulut, c., & mahmudlu, c. (2017). transition to renewable energy and sustainable energy development in azerbaijan. renewable and sustainable energy reviews, 80(april), 1153–1161. https://doi.org/10.1016/j.rser.2017.05.168 vidinopoulos, a., whale, j., & fuentes hutfilter, u. (2020). assessing the technical potential of asean countries to achieve 100% renewable energy supply. sustainable energy technologies and assessments, 42(october), 100878. https://doi.org/10.1016/j.seta.2020.100878 wahono, j., brady, m., & baral, h. (2022). ensuring monetary, human capital and natural capital returns in biomass production: lessons from the mentawai biomass gasification power plant project. in bioenergy for landscape restoration and livelihoods: recreating energy-smart ecosystems on degraded landscape. cifor. https://doi.org/10.17528/cifor/008500 wattana, b., & aungyut, p. (2022). impacts of solar electricity generation on the thai electricity industry. international journal of renewable energy development, 11(1), 157–163. https://doi.org/10.14710/ijred.2022.41059 wong, j. (2021). a combined niche transition and energy justice study of biomass gasification in indonesia. thesis. tu delft. http://resolver.tudelft.nl/uuid:be0ed7f3-b696-4756-b31f59b4b17ecb86 yudiartono, y., windarta, j., & adiarso, a. (2022). analisis prakiraan kebutuhan energi nasional jangka panjang untuk mendukung program peta jalan transisi energi menuju karbon netral. jurnal energi baru dan terbarukan, 3, 201 – 217. https://doi.org/10.14710/jebt.2022.14264 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1016/j.seta.2022.102668 https://doi.org/10.1016/j.rser.2018.11.005 https://unfccc.int/sites/default/files/resource/indonesia_lts-lccr_2021.pdf https://unfccc.int/sites/default/files/resource/indonesia_lts-lccr_2021.pdf https://www.esdm.go.id/en/media-center/news-archives/road-map-to-carbon-neutrality-is-being-prepared-says-energy-minister https://www.esdm.go.id/en/media-center/news-archives/road-map-to-carbon-neutrality-is-being-prepared-says-energy-minister https://www.esdm.go.id/en/media-center/news-archives/road-map-to-carbon-neutrality-is-being-prepared-says-energy-minister https://doi.org/10.1016/j.energy.2018.10.012 https://doi.org/10.1016/j.heliyon.2019.e01166 https://www.esdm.go.id/assets/media/content/content-indonesia-energy-outlook-2019-english-version.pdf https://www.esdm.go.id/assets/media/content/content-indonesia-energy-outlook-2019-english-version.pdf https://doi.org/10.1016/j.esd.2022.04.017 https://doi.org/10.1016/j.rser.2012.02.010 https://doi.org/10.1016/j.rser.2021.111186 https://web.pln.co.id/statics/uploads/2021/10/ruptl-2021-2030.pdf https://web.pln.co.id/statics/uploads/2021/10/ruptl-2021-2030.pdf https://www.purnomoyusgiantorocenter.org/a-turning-point-for-renewable-energy-in-indonesia/1%e2%80%936 https://www.purnomoyusgiantorocenter.org/a-turning-point-for-renewable-energy-in-indonesia/1%e2%80%936 https://doi.org/10.1016/j.rser.2021.111027 https://doi.org/10.1016/j.heliyon.2022.e09304 https://doi.org/10.1016/j.rser.2021.111507 https://doi.org/10.1016/j.energy.2021.120474 https://data.worldbank.org/indicator/en.atm.co2e.pc http://theargeo.org/november/files/economic%20impacts%20of%20renewable%20energy%20promotion%20in%20germany.pdf http://theargeo.org/november/files/economic%20impacts%20of%20renewable%20energy%20promotion%20in%20germany.pdf http://theargeo.org/november/files/economic%20impacts%20of%20renewable%20energy%20promotion%20in%20germany.pdf https://doi.org/10.1016/j.rser.2017.05.168 https://doi.org/10.1016/j.seta.2020.100878 https://doi.org/10.17528/cifor/008500 https://doi.org/10.14710/ijred.2022.41059 http://resolver.tudelft.nl/uuid:be0ed7f3-b696-4756-b31f-59b4b17ecb86 http://resolver.tudelft.nl/uuid:be0ed7f3-b696-4756-b31f-59b4b17ecb86 https://doi.org/10.14710/jebt.2022.14264 sustainable long-term energy supply and demand: the gradual transition to a new and renewable energy system in indonesia by 2050 yudiartono yudiartonoa,b0f , jaka windartaa, adiarso adiarsob 1. introduction international journal of renewable energy development int. j. renew. energy dev. 2023, 12(5), 816-831 | 816 https://doi.org/10.14710/ijred.2023.53249 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id enhancing transient stability and dynamic response of windpenetrated power systems through pss and statcom cooperation khaled kouider* and abdelkader bekri the control, analysis and optimization of the electro-energetic systems (caosee) laboratory, tahri mohamed university, bechar, algeria abstract. the large-scale integration of doubly-fed induction generator (dfig) based wind power plants poses stability challenges for power system operation. this study investigates the transient stability and dynamic performance of a modified 3-machine, 9-bus western system coordinating council (wscc) system. the investigation was conducted by connecting the dfig wind farm to the sixth bus via a low-impedance transmission line and installing power system stabilizers (psss) on all automatic voltage regulators (avrs). a three-phase fault simulation was carried out to test the system, with and without power system stabilizers and a static synchronous compensator (statcom) device. time-domain simulations demonstrate improved transient response with pss-statcom control. a 50% reduction in settling time and 70% decrease in power angle undershoots at the slack bus are achieved following disturbances, even at minimum wind penetration levels. load flow analysis shows the coordinated controllers maintain voltages within 0.5% of nominal at 60% wind penetration, while voltages at load buses can deviate up to 15% without control. eigenvalue analysis indicates the pss-statcom boosts damping ratios of critical oscillatory modes from nearly 0% to over 30% under high wind injection. together, the present findings provide significant evidence that pss and statcom cooperation enhances dynamic voltage regulation, angle stability, and damping across operating ranges, thereby maintaining secure operation in systems with high renewable integration. keywords: doubly-fed induction generator, power system stabilizer, transient stability, power system oscillations, static synchronous compensator, western system coordinating council @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 15th march 2023; revised: 10th june 2023; accepted: 18th july 2023; available online: 25th july 2023 1. introduction renewable energy has become a new area of study due to growing concerns about climate change and the limitation of fossil fuels. renewable energy sources such as wind and solar can help reduce the reliance on traditional fossil fuels. wind turbines' reliability and durability increase with the wind farm penetration level. wind farms must supply the appropriate amount of electricity based on the grid's wind speed and energy demand (simani & farsoni, 2018). dfig wind technologies are becoming increasingly popular in the renewable energy sector due to their high level of efficiency in relation to their cost (miller, 2010; yunus et al., 2019). with the broad penetration of these intermittent energy sources, power system stability can be affected in the case of insufficient control and poor damping. conceptually and from a heuristic standpoint, a system is considered stable if it can maintain equilibrium under normal conditions and return to an acceptable equilibrium after being disturbed (prabha, 1994). loss of synchronism between rotating inertias, low voltages, natural disturbances, or protection system malfunctions can all potentially cause power system instability (eremia & shahidehpour, 2013). power system stability can be categorized into three main types: rotor angle, voltage, and frequency stability. in (asija et al., 2015), matlab was used to perform contingency analysis and power flow studies on a wscc 9-bus test system to * corresponding author email: khouiledkhaled@gmail.com (k. kouider) investigate its stability and dynamics. recently, the impact of dfig wind turbines on power system stability has gained considerable attention due to advances in wind power conversion technology. integrating dfig wind turbines into power systems raises particular concerns regarding small-signal stability, which involves the ability to recover and maintain equilibrium after minor disturbances. nkosi et al., (2023) analyze small signal stability in power systems with doubly fed induction generators. they review the latest advancements in modeling dfig-based wind farms and examine control techniques that improve the damping properties of the power system. squirrel cage induction generator (scig) and dfig implementations in a 14-bus ieee system investigate small signal stability in (chandra et al., 2014), and simulation results are reported. as stated in (bagchi et al., 2016), analogous functions of static var compensator (svc) and statcom are embedded in dfigs for small signal stability requirements. in (pérez-londoño et al., 2012), the impact of dfigs on power system voltage stability was studied, and deduce that only small-scale wind power penetration preserves stability. large power systems face numerous challenges related to transient stability. a stable state after a disturbance is called post-fault equilibrium, and it indicates the ability of the system to return to a pre-fault state following significant disturbances, such as faults, overloads, or generator unit failures research article https://doi.org/10.14710/ijred.2023.53249 https://doi.org/10.14710/ijred.2023.53249 mailto:khouiledkhaled@gmail.com https://orcid.org/0000-0003-3556-946x https://orcid.org/0000-0003-4477-8132 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.53249&domain=pdf k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |817 issn: 2252-4940/© 2023. the author(s). published by cbiore (hatziargyriou et al., 2021; ramanujam, 2010; xu et al., 2023). there has been extensive research aimed at assessing transient stability in wind power systems. (reza et al., 2003) examines how the transient stability of the transmission system is affected by distributed generation (dg) and analyzes various dg technologies and their levels of penetration in the power system. in (morshed, 2020), a coordination strategy using the zerodynamics technique to ameliorate the transient stability of ieee 39-bus power systems, including doubly-fed induction generators (dfigs), under varying operating conditions. (xia et al., 2018) analyzes the impact of wind power generation on power system transient stability during changes in energy flow. specifically examining the effects of wind farms' location, speed, and capacity on the stability of the ieee 14-bus system. the study (pillai et al., 2013) examines the transient stability of wind power systems with storage utilizing a central zone controller. (edrah et al., 2015) analyze insightfully the impact of controlling and operating the dfig on the rotor angle stability of power systems. they propose addressing this issue by implementing a pss on the reactive control loop for the rotor-side converter (rsc) and reconfiguring the grid-side converter (gsc) to function as a statcom. which aims to minimize the influence of the dfig on the overall stability of the power system. the transient stability of wind farms in multi-machine power systems can be enhanced using fuzzy controllers based on statcoms and svcs, as demonstrated in (m. hemeida, h. rezk, and m. hamada, 2018). meanwhile, simulation results in (fdaili et al., 2021) show that the proposed uncontrolled fault current limiter (ncfcl) with reactive power back-up is a better approach than crowbar protection for enhancing fault ride through (frt) capabilities. a method for stabilizing rotor angle is proposed in (zheng et al., 2019), which is based on the phaseamplitude characteristics of grid transient voltage. the impact of dfigs on the system is assessed using an index that reflects variations in the acceleration area of synchronous generators. the simulation outcomes confirm that the strategy works well. (cai, dong, & liao, 2021) examines the dynamic performance of wind turbines under extreme grid disturbances, particularly through zero-voltage ride-through (zvrt). during zvrt, the reactive response of a doubly-fed (type-3) wind turbine is influenced by the converter's control technique and the steadystate and transient-state performance of the dfig. reference (alsakati et al., 2022) studies adjusting the transient stability of a two-zone, four-machine wind energy system using the multiple-band power system stabilizer (ombpss4c). (eshkaftaki et al., 2020) propose two methods to improve the transient and dynamic stability of the local synchronous generator (sg). the first method is a genetically tuned electromagnetic torque band damping controller (etbdc), and the second method is a reactive power band damping controller (rpbdc) utilizing the suggested transient controller (tc) for transient stability. under temporary disturbances, the tc transitions the dfig operation from generator to motor speed, while the etbdc and rpbdc significantly improve the dynamic performance of the sg. the study (agarala et al., 2022) presents a new control technique called automatic reactive power support (ars) that enhances multi-machine power system stability by injecting available reactive power during faults through converters. the comparative analysis across different test cases, including scenarios with different types of renewable sources, such as wind generators like permanent magnet synchronous generators (pmsg) and doubly fed induction generators, as well as solar pv, demonstrates the effectiveness of the proposed control technique in improving system stability and critical clearing time. the impacts of a closed-loop dfig model on the transient stability of a power system with high penetration of dfig wind energy are thoroughly examined (shabani, kalantar, & hajizadeh, 2021a). an innovative method for real-time transient instability (tid) detection is described in (shabani & kalantar, 2021b). utilizing a transient energy function, this approach is deployed in a power system with a high penetration of dfig wind farms. tid in a shorter time is achieved using the proposed strategy, according to simulation findings. the issue of power system oscillations poses significant concerns in the field of power system stability. several factors may contribute to these oscillations, including torque imbalances, insufficient damping, inadequate controller tuning, and including interactions between controllers and transmission lines compensated by series capacitors (eremia & shahidehpour, 2013; prabha, 1994;). moreover, the frequency ranges (0.1-0.8 hz) and (1-2 hz) enable the identification of two distinct oscillation modes in the power system, namely local and inter-area modes. these electro-mechanical oscillation modes were investigated in previous works as cited in references (avdakovic et al., 2009; klein, rogers, & prabha, 1991; yang et al., 2011). other studies analyzed and compared various aspects of power system stability, including oscillation damping (edrah et al., 2016; falehi et al., 2012; li et al., 2022; thanpisit & ngamroo, 2017; zhang et al., 2018). the findings of (edrah et al., 2016) indicate that incorporating conventional fixed parameter pss into the reactive power control loop of the dfig rotor-side converter has a favorable damping effect across various operating conditions. in addition, the study demonstrates that dfig-based wind farms equipped with the suggested farm-level pss exhibit superior effectiveness in attenuating power system oscillations compared to pss used in synchronous generators. moreover, it is possible to enhance power system stability by optimizing and coordinating additional controllers based on svcs and psss, using a genetic algorithm (ga) as proposed by (falehi et al., 2012). furthermore, the optimization model for power oscillation dampers (pod) parameters described by (li et al., 2022) incorporates the constraint of the dfig to mitigate the oscillations of both the dfig and the power grid. per smallsignal and transient stability measurements, the pss and pod recommended in the study by (thanpisit & ngamroo, 2017) show noticeably better damping performance than traditional pss and pod controllers in various operating conditions and fault scenarios. per (zhang et al., 2018), nter-area modes can be effectively suppressed by employing both pss and tcs in interconnected power systems of new england and new york, as shown by the implementation of a specific design approach. in contrast, based on the author's review, very few publications are available in the literature that address the issue of transient stability in parallel with damping power oscillation types (gurung & kamalasadan, 2020; he et al., 2022; morshed & fekih, 2019; yu et al., 2018). a model-based reduced-order optimal oscillation damping controller (oodc) for a large-scale wind farm with a dual-fed induction generator was demonstrated in (gurung & kamalasadan, 2020). simulation results using a matched ieee 68 bus network demonstrate that the suggested oodc effectively improves the inter-area mode damping of the system. (he et al., 2022) proposes a combined approach of statcom, pod controller, and psss to enhance power system stability. an intelligent optimization algorithm, integrating ga and particle swarm algorithm (pso), is implemented to surmount local convergence challenges. numerical simulations on ieee systems demonstrate the effectiveness of the method in suppressing low-frequency k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |818 issn: 2252-4940/© 2023. the author(s). published by cbiore power oscillation in the wind-pv-thermal-bundled (wptb) systems. the study in (morshed & fekih, 2019) proposes a modified imperialist competitive algorithm (mica) and a stochastic eigenvalue approach to combine the power system stabilizers settings of the synchronous generator, the power oscillation dampers of the dfig, and the controllers of the static synchronous compensator to improve the dynamic stability of the power system. the proposed solution tests on a modified 39-bus new england power system under different wind conditions. the performance of the system evaluates using time-domain analysis, eigenvalue mapping, and robustness analysis. reference (yu et al., 2018) introduce an innovative strategy called improved adaptive phasor power oscillation damping (eappod) to adjust for time-varying communication latencies and mitigate low-frequency oscillations in inter-area signals due to external disturbances in a complex power system with dfig wind farm. the current paper examines the impact of a dfig wind farm on the transient stability and dynamic performance of a modified nine-bus wscc power system. despite encouraging advancements in previous studies (gurung & kamalasadan, 2020; he et al., 2022; morshed & fekih, 2019; yu et al., 2018), detailed investigations into voltage stability and reactive power support are still needed. the paper begins with discussing the influence of high penetration of wind sources on power system reliability and stability, then provides a brief statement on power system stability classifications. the topic of transient stability of large power systems, including dfig wind farms, then analyzes in depth, emphasizing the notion and nature of oscillation in power systems. the paper explains the primary purpose of power system stabilizers, which is to maintain rotor angle stability after a large disturbance and effectively handle the two types of oscillations. further, the main objective of the statcom equipment is providing enough reactive power to recover the voltage profile safely under harsh operating conditions. finally, the paper suggests a combination of psss/statcom to address both the rotor angle stability problem and provide strong reactive power support in the event of a three-phase grid fault and significant wind power penetration. the study uses the matlab-based power system analysis toolbox (psat) to measure the system performance. 2. system modeling the dynamic models of synchronous machines that are suitable for stability investigations have been thoroughly discussed in (prabha, 1994; shabani et al 2021a). this section focuses primarily on dfig modeling. dfig wind turbines connect properly to the power grid through back-to-back converters, as depicted in figure 1. 2.1 dfig dynamic modeling details of the dfig-based wind turbine’s mathematical modelling as well as its electrical equations can be found in (tang et al., 2018; yang et al., 2016). as a result, utilizing the extensive set of equations in (muñoz & cañizares, 2011) that describe the dfig model, certain assumptions can be made. specifically, the stator and rotor flux dynamics are faster compared to grid dynamics. additionally, the converter controls decouple the generator and grid interactions. based on these assumptions, the following outcomes can be derived: 𝑣𝑑𝑠 = −𝑅𝑠 𝑖𝑑𝑠 + (𝑥𝑠 + 𝑥𝑚 )𝑖𝑞𝑠 + 𝑥𝑚 𝑖𝑞𝑟 (1) 𝑣𝑞𝑠 = −𝑅𝑠 𝑖𝑞𝑠 − (𝑥𝑠 + 𝑥𝑚 )𝑖𝑑𝑠 + 𝑥𝑚 𝑖𝑑𝑟 (2) for the rotor circuit: 𝑣𝑑𝑟 = −𝑅𝑟 𝑖𝑑𝑟 − (1 + 𝜔)((𝑥𝑟 + 𝑥𝑚 )𝑖𝑞𝑟 + 𝑥𝑚 𝑖𝑞𝑠) (3) 𝑣𝑞𝑟 = −𝑅𝑟 𝑖𝑞𝑟 − (1 − 𝜔)((𝑥𝑟 + 𝑥𝑚 )𝑖𝑑𝑟 + 𝑥𝑚 𝑖𝑑𝑠) (4) where 𝑣𝑑𝑠, 𝑣𝑞𝑠, 𝑣𝑑𝑟 , 𝑣𝑞𝑟 are dq components of the stator and rotor voltages. 𝑖𝑑𝑠, 𝑖𝑞𝑠, 𝑖𝑑𝑟 , 𝑖𝑞𝑟 are the d and q axes stator and rotor currents. 𝑅𝑠 , 𝑅𝑟 are the resistances of both the stator and the rotor. 𝑥𝑠 , 𝑥𝑟, 𝑥𝑚 stator, rotor, and mutual inductances. 𝜔 is the rotor speed. the electromagnetic torque may therefore be expressed as: 𝑇𝑒 = 𝑥𝑚 (𝑖𝑞𝑟 𝑖𝑞𝑠 − 𝑖𝑑𝑠𝑖𝑑𝑟 ) (5) additionally, the active and reactive powers for both the stator and rotor can be expressed as follows: 𝑃𝑠 = 𝑣𝑑𝑠𝑖𝑑𝑠 + 𝑣𝑞𝑠𝑖𝑞𝑠 (6) 𝑄𝑠 = 𝑣𝑞𝑠𝑖𝑑𝑠 − 𝑣𝑑𝑠𝑖𝑞𝑠 (7) 𝑃𝑟 = 𝑣𝑑𝑟 𝑖𝑑𝑟 + 𝑣𝑞𝑟 𝑖𝑞𝑟 (8) 𝑄𝑟 = 𝑣𝑞𝑟 𝑖𝑑𝑟 − 𝑣𝑑𝑟 𝑖𝑞𝑟 (9) with 𝑃𝑠 , 𝑄𝑠 are the active and reactive powers, while the rotor active and reactive powers are presented by 𝑃𝑟 , 𝑄𝑟 in that order. finally, the total active and reactive powers 𝑃𝑡𝑜𝑡 , 𝑄𝑡𝑜𝑡 exchanged with the grid: 𝑃𝑡𝑜𝑡 = 𝑃𝑠 + 𝑃𝑟 (10) 𝑄𝑡𝑜𝑡 = 𝑂𝑠 + 𝑄𝑟 (11) with vector control of the dfig, the active and reactive powers are decoupled by orienting the rotor reference frame. the q-axis current component governs active power, while the d-axis component controls reactive power. to ensure precise dynamic and steady-state operation of the dfig, the rotor 𝑑𝑞 current limits can be accurately calculated: 𝑖𝑞𝑟𝑚𝑎𝑥 ≈ − 𝑥𝑠+𝑥𝑚 𝑥𝑚 𝑃𝑚𝑖𝑛 (12) fig. 1 a grid-connected dfig wind system's typical configuration. fig. 2 power system stabilizer type ii block diagram. k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |819 issn: 2252-4940/© 2023. the author(s). published by cbiore 𝑖𝑞𝑟𝑚𝑖𝑛 ≈ − 𝑥𝑠+𝑥𝑚 𝑥𝑚 𝑃𝑚𝑎𝑥 (13) 𝑖𝑑𝑟𝑚𝑎𝑥 ≈ − 𝑥𝑠+𝑥𝑚 𝑥𝑚 𝑄𝑚𝑖𝑛 − 𝑥𝑠+𝑥𝑚 𝑥𝑚 2 (14) 𝑖𝑑𝑟𝑚𝑖𝑛 ≈ − 𝑥𝑠+𝑥𝑚 𝑥𝑚 𝑄𝑚𝑎𝑥 − 𝑥𝑠+𝑥𝑚 𝑥𝑚 2 (15) with 𝑃𝑚𝑎𝑥 , 𝑃𝑚𝑖𝑛, 𝑄𝑚𝑎𝑥 , and 𝑄𝑚𝑖𝑛 are the dfig’s max-min active and reactive powers. 2.2 power system stabilizer (pss) negative damping in synchronous machines is primarily caused by delays in the field winding excitation. to mitigate this issue, power system stabilizers (pss) are designed to counteract the destabilizing effect of these delays at critical frequencies between 1.0-2.0 hz (ramanujam, 2010). a typical configuration of a pss is shown in figure 2. the phase compensation block provides the necessary phase lead to the speed deviation signal. practical pss designs may utilize multiple lead-lag blocks for phase compensation. the washout block, with time constant 𝑇𝑤 typically 0.1 to 20 seconds, prevents the pss signal from introducing dc bias into the voltage regulator setpoint. the damping factor 𝐾𝑝𝑠𝑠 is a good measure of the damping amount. the anti-windup limiter controls the output signal 𝒗𝒔 dynamics, which has a small time constant 𝑇∈ = 0.001s (milano, 2008). note that the input signal in our study 𝑣𝑠𝑖 is the rotor speed variation δω . the pss type ii differential-algebraic equations are given as follows: �̇�1 = −(𝐾𝑤 𝑣𝑠𝑖 + 𝑣1)/𝑇𝑤 (16) �̇�2 = ((1 − 𝑇1 𝑇2 )(𝐾𝑤 𝑣𝑠𝑖 + 𝑣1) − 𝑣2)/𝑇2 �̇�3 = ((1 − 𝑇3 𝑇4 )(𝑣2 + ( 𝑇1 𝑇2 (𝐾𝑤 𝑣𝑠𝑖 + 𝑣1))) − 𝑣3)/𝑇4 fig. 3 statcom’s basic model. (a) (b) (c) (d) fig. 4 wscc 9 bus test system in psat. (a) original test system. (b) modified test system (case1). (c) modified test system (case2). (d) modified test system (case3). k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |820 issn: 2252-4940/© 2023. the author(s). published by cbiore �̇�𝑠 = (𝑣3 + 𝑇3 𝑇4 (𝑣2 + ( 𝑇1 𝑇2 (𝐾𝑤 𝑣𝑠𝑖 + 𝑣1)) − 𝑣𝑠)/𝑇∈ with 𝑇1, 𝑇2, 𝑇3, 𝑇4 are the time constants for the 1 𝑠𝑡,2 𝑛𝑑 , 3 𝑟𝑑 and the 4 𝑡ℎ stabilizers. 𝑣1, 𝑣2, 𝑣3 are the state variables. 2.3 static synchronous compensator (statcom) a static synchronous compensator (statcom) provides reactive power compensation by connecting in parallel with the power system. it utilizes solid-state switching converters supplied by capacitors or other energy storage elements (hemeida et al 2018). similar to a static var compensator (svc), the statcom stabilizes transmission voltage through reactive shunt compensation. however, it can operate as either a voltage-sourced or current-sourced converter (kothari & nagrath, 2019). the statcom is versatile, capable of harmonic filtering, stability enhancement, and preventing voltage collapse. integrating a statcom with wind generation can significantly improve system performance and stability (hemeida et al 2018; qiao et al., 2009). figure 3 illustrates the model of a statcom device. to transmit reactive power from a statcom (𝑄𝑐 ) to a power system, the device's voltage magnitude can be controlled using equation (17): 𝑄𝑐 = 𝑉(𝑉𝑐−𝑉) 𝑋𝑡 (17) with v is the grid voltage, 𝑉𝑐 representing statcom output voltage, and 𝑋𝑡 the leakage reactance. 3. the simulation model description figure 4 summarizes the tested model configurations. the work utilizes the standard wscc 9-bus power system model with a 100 mva, 60 hz base. three synchronous generators are connected to buses 1, 2, and 3, with bus 1 as the slack bus and buses 2 and 3 as pv buses. additionally, there are three pq load buses interconnected at buses 5, 6, and 8, which connect to the rest of the system through six transmission lines, as manifested in figure 4a. the specific system data used in the simulations is provided in detail in the appendix. the simulations were performed using the psat toolbox, a freely available matlab-based software for power system analysis (milano, 2008). psat provides a comprehensive graphical interface and simulink-based network editor, enabling convenient evaluation of system dynamics. the toolkit also features graphical user interfaces for constructing electrical schematics and multi-machine networks using pictorial blocks (kumar et al., 2020). to evaluate the performance of the wscc test system, a dfig wind farm was integrated at bus 6 through a low im pedance transmission line. bus 6 was strategically chosen for the wind farm connection to effectively assess the overall system stability. the first test scenario subjected the reconfigured system to a three-phase fault at bus 6 without pss and statcom controls (figure 4b). the fault persisted for five cycles between t = 3 s and t = 3.083 s. the second scenario added pss controllers to the avr systems of all synchronous generators (figure 4c). finally, the third case incorporated a statcom device at bus 6. the last case configuration is illustrated in figure 4d. 4. simulation results and discussion the trapezoidal integration method, widely recognized as a reliable and stable approach for diverse test scenarios, was utilized for the time-domain simulations. notably, the upcoming subsection (4.1) will elaborate on the findings and results presented in figures 5-9 and table 1. 4.1 time-domain simulation results this section evaluates the rotor angle 𝛿, angular velocity 𝜔 of the synchronous generators, the dfig’s produced power 𝑃𝐷𝐹𝐼𝐺 , and the 𝑑𝑞 component of the rotor currents 𝑖𝑑𝑟 and 𝑖𝑞𝑟. the three test cases are compared under wind power penetration levels of 0.2, 0.4, and 0.6 p.u. table 1 summarizes the observed overshoots and undershoots for the simulated parameters. a. comparison with 𝑃𝐷𝐹𝐼𝐺 = 0.2 𝑝. 𝑢. (all cases) large-disturbance rotor angle stability is a critical aspect of transient stability in power systems. it refers to the ability of synchronous generators to maintain synchronism and steady rotor angles when subjected to significant disturbances such as faults, sudden load changes, or transmission line outages. rotor angle stability analysis focuses on controlling and minimizing rotor angle deviations during and after major system disturbances. loss of synchronism between generators leads to unstable angular oscillations and system collapse (prabha, 1994; xu et al., 2023). the system as a whole appears stable, as plotted in figs 5(a & b) and figs 6(a & b). figure 5a shows the power angles of the three generating units. for 𝛿1, the no-control case has an over shoot of (+0.6546 rad, +37.5°) and undershoot of (-1.2213 rad, -70°) due to the slack bus response after the fault. the pss, with high gain (𝐾𝑤,=50) and fast time constant (𝑇𝑤,=0.1s), rapidly applies corrective damping but initially overshoots (+0.73625 rad, +42.16°). however, it significantly reduces the undershoot by nearly 50% to (-0.68035 rad, -38.98°). with pss (a) (b) fig. 5 time-domain simulation results with (𝑷𝑫𝑭𝑰𝑮 = 0.2 p.u.). (a) power angles, (b) rotor angular speeds. k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |821 issn: 2252-4940/© 2023. the author(s). published by cbiore and statcom, the overshoot further drops from +42.16° to +38.29°, while the undershoot is halved again from -38.98° to 18.44° with fastest 2% settling time of 10.9s. regarding 𝜹𝟐, in the first case has a maximum overshoot of +43.8° and undershoot of -73.4° due to the high power output. the pss slightly increases the overshoot to +44.06° but reduces the undershoot by 45.6% to -39.93°. with pss and statcom, the overshoot decreases to +38.98° and undershoot drops further by 53.46% to -18.55°. as for 𝜹𝟑, the first case exhibits a maximum peak of (+0.627 rad, +35.9°), and the lowest value reaches (−1.2248 rad, −70.1°). these large angle deviations are attributed to the third generator having the lowest inertia of 6.3 seconds. in spite of this, with the implementation of the pss, it counterbalances the fast response of the low-inertia generator, resulting in a larger power angle deviation of +29.42% (46.04°). conversely, the lower band experiences a significant reduction of 44.62% to (-38.07°). subsequently, both peaks are further lowered to 42.16° and -20.23°, respectively, following the integration of the statcom. figure 5b shows the rotor angular velocities 𝜔1, 𝜔2, and 𝜔3in case 1, the initial overshoot was +0.4% (+0.24 hz) in 𝜔1 and +0.6% (+0.36 hz) in 𝜔2. however, in both the 2nd and the 3rd cases, overshoots noticeably reduced to +0.3% (+0.18 hz) in 𝜔1 and +0.5% (+0.30 hz) in 𝜔2. for the minimum points, 𝜔1 undershoot declined by 32.3% to -0.23% (-0.138 hz), while 𝜔2 declined by 37.04% to -0.34% (-0.204 hz). generator 1, with an inertia of 47s, had a slower, reduced response to disturbances, resulting in decreased overshoots. in contrast, generator 2, with an inertia of 12.8s, exhibited a faster response with larger peaks. regarding 𝜔3, all cases had identical overshoots of +0.7% (increase of 0.42hz). the fast response of the 3rd generator with low inertia (6.2s) to faults provides insufficient time for the pss/statcom to fully handle the initial overshoot. yet, the undershoot experienced a considerable reduction of 40.74% from (-0.54%, 0.334hz) in the 1st case to (-0.32%, 0.192hz) in both subsequent cases. the three curves settle in around 32 seconds. the three generators have inertias of 47.2, 12.8 and 6.2 seconds, respectively. higher inertia provides for greater energy storage, damping, time constants, and synchronizing power. these factors reduce the sensitivity of rotor speed to disturbances, resulting in improved system stability and resilience against fluctuations in mechanical torque (denholm et al., 2020, eriksson et al., 2018). moreover, table 1 and figs 5(a & b) demonstrate that even with +20mw wind penetration, the system can exhibit instability characterized by low frequency oscillations (lfo) in both 𝛿 and 𝜔. however, stability is significantly enhanced by the pss/statcom combination, as evidenced by the reduced damped oscillations and peak values. the plots 6(a & b) illustrate the dfig’s produced power 𝑃𝐷𝐹𝐼𝐺 and the 𝑑𝑞 rotor currents, correspondingly. case 2 has the highest power peak at +27.4%, corresponding to the maximum power angle deviations. conversely, case 1 has the lowest undershoot of -16.75%. with pss and statcom in case 3, the overshoot reduces to +25.65% while the undershoot substantially decreases to -8.8%. the power signal shows effective damping characteristics over time. notably, case 3 has the fastest 2% settling time of 15.4 seconds, indicating rapid dynamic response. based on equations (12-15), minimum undershoots of -0.7 p.u. and -0.93 p.u. are observed for 𝑖𝑑𝑟_𝑚𝑖𝑛 and 𝑖𝑞𝑟_𝑚𝑖𝑛 respectively in all cases at fault inception. in case 1, 𝑖𝑑𝑟 peaks at +17.01% (𝑄𝑚𝑖𝑛 of 0.0761 p.u.), while 𝑖𝑞𝑟_𝑚𝑎𝑥 reaches +56.21%, increasing 𝑃𝐷𝐹𝐼𝐺 by +3.75%. with pss, 𝑖𝑑𝑟 drops by -42.05% (𝑄𝑚𝑖𝑛 of 0.0107 p.u.), confirming its indirect voltage control contribution. meanwhile, 𝑖𝑞𝑟_𝑚𝑎𝑥 peaks at +56.63%, reflecting the +3.75% 𝑃𝐷𝐹𝐼𝐺 rise. the peaks occur at t=4.54s for 𝑖𝑑𝑟_𝑚𝑎𝑥 and t=3.121s for 𝑖𝑞𝑟_𝑚𝑎𝑥. in case 3, 𝑖𝑑𝑟 substantially decreases by 76.38% to 2.33% (𝑄𝑚𝑖𝑛 of -0.0105 p.u.), indicating reactive power absorption. this significant reduction results from the additional reactive support by the statcom, further evidenced by the decreased over-voltage at bus 10 in table 2. conversely, 𝑖𝑞𝑟 increases by 38.41%, representing a 4% rise in wind farm output. alongside the power boost, case 3 has the shortest 2% settling times for both currents at t=4.16s and t=8.88s for 𝑖𝑑𝑟 and 𝑖𝑞𝑟 respectively. (a) (b) (c) (d) fig. 6 time-domain simulation results (a) dfig power with (𝑷𝑫𝑭𝑰𝑮 = 0.2 p.u.), (b) rotor dq currents (𝑷𝑫𝑭𝑰𝑮= 0.2 p.u.), (c) power angles (𝑷𝑫𝑭𝑰𝑮 = 0.4, 0.6 p.u.), (d) rotor angular speeds (𝑷𝑫𝑭𝑰𝑮 = 0.4, 0.6 p.u.). k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |822 issn: 2252-4940/© 2023. the author(s). published by cbiore b. comparison with 𝑃𝐷𝐹𝐼𝐺 = 0.4 𝑝. 𝑢., 0.6 p.u. (case1) figures 6(c & d) and figs 7(a & b) show that case 1 becomes unstable at t = 22.4s and t = 12.8s for 0.4 p.u. and 0.6 p.u. wind penetration, in turn. specifically, at 0.6 p.u., voltages on buses 5, 6, 8, 10, and 11 undergo catastrophic failure. the dfig rotor currents reach minimum limits, triggering the speed and voltage controller anti-windup mechanisms to prevent over-currents in the converters and maintain powers and currents within limits. however, the generator power angles still diverge and rotor speeds display completely undamped behavior with very slow dynamic divergence. c-comparison with 𝑃𝐷𝐹𝐼𝐺 = 0.4 𝑝. 𝑢. (case 2 and 3) figures 7(c, d, & e), and 8a depict demonstrate system stability under the specified conditions. compared to 0.2 p.u., adding pss in case 2 decreases the power angle overshoots by 14.5%, 9.2%, and 16.86% for 𝛿1, 𝛿2, and 𝛿3 respectively, as shown in fig. 7c and table 1. this is attributed to the extra 0.53s inertia from the dfig. with pss and statcom damping in case 3, the angles further reduce to 30.97°, 33.85°, and 36.14°, changed by -14.12%, -15.38%, and -5.57% correspondingly. regarding undershoots, case 2 has lower values versus case 1: -13.87% for 𝛿1, -13.75% for 𝛿2, and -24.6% for 𝛿3, unlike at 20 mw. moreover, case 3 optimizes these minimums substantially with declines of -52.7%, -53.4%, and -42.9% for the angles respectively. case 3 also exhibits the fastest 2% settling time of 13.05 seconds. for rotor speeds 𝜔1, 𝜔2, and 𝜔3, cases 2 and 3 have similar overshoots of +0.3%, +0.5%, and +0.7% and undershoots of 0.21%, -0.37%, and -0.3% respectively. however, compared to 0.2 p.u., the undershoots in case 2 decrease by -8.70% for 𝜔1 and -6.25% for 𝜔3, while 𝜔2 increases by +8.82% due to the dfig's added 0.53s inertia influencing system dynamics. the three units resynchronize in approximately 14.2 seconds, as shown (fig 7d). the generated power and the 𝑑𝑞 rotor currents are pictured in figures 7e and 8a, in that order. adding the statcom in case 3 reduces the power peaks from (+12.75%, -5.6%) in case 2 to (+10.92%, -3.475%), decreasing them by 14.35% and -37.95% respectively, with a 2% settling time of 12.05s. as stated before, 𝑖𝑑𝑟_𝑚𝑎𝑥 and 𝑖𝑞𝑟_𝑚𝑖𝑛 are the minimum points. in case 2, 𝑖𝑑𝑟 peaks at +7.9% (𝑄𝑚𝑖𝑛 of -0.0511 p.u.), while max 𝑖𝑞𝑟 reaches +19.3%, increasing turbine power by +3.225%. versus case 1, 𝑖𝑑𝑟 shows a smaller +0.37% rise (𝑄𝑚𝑖𝑛 of -0.0236 p.u.), while 𝑖𝑞𝑟 exhibits a larger +32% growth corresponding to a +2.85% power increase. the peaks occur at t=4.5s for 𝑖𝑑𝑟 and t=3.128s for 𝑖𝑞𝑟. the currents settle to 2% pre-fault value at t=4.54s and t=6.04s. d-comparison with 𝑃𝐷𝐹𝐼𝐺 = 0.6 𝑝. 𝑢. (case 2 and 3) figure 8b plots the power angles of the three generators. compared to 0.4 p.u., case 2 shows reduced overshoot peaks of -15.59%, -23.20%, and -5.58% for 𝛿1, 𝛿2, and 𝛿3 in turn, while the minimums decrease by -10.63%, 2.82%, and 13.2% respectively. in case 3, overshoots are further lowered by 17.28%, -3.39%, and -14.56%, and minimums are optimized with reductions of -47.47%, -55.16%, and -37.94% for the angles, reaching steady state after 27s. these results validate the enhanced system stability provided by additional damping support from the wind generation and controllers. prior work by (edrah et al., 2015; he et al., 2022; shahgholian & izadpanahi, 2016) supports utilizing wind farms for stability improvement. table 1 and figure 8c show identical rotor velocity overshoots for both cases 2 and 3. however, the minimums table 1 overshoots and undershoots of 𝛿𝑠𝑦𝑛, 𝜔𝑠𝑦𝑛, 𝑃𝐷𝐹𝐼𝐺 , 𝑖𝑑𝑟 , 𝑖𝑞𝑟, 𝐼𝑠𝑡𝑎𝑡𝑐𝑜𝑚 for all cases. case 1 (no pss/no statcom) case 2 (only pss) case 3 (pss + statcom) 0.2 p.u. 0.4 p.u./ 0.6 p.u. instable instable instable instable instable instable instable instable instable instable 0.2 p.u. 0.4 p.u. 0.6 p.u. 0.2 p.u. 0.4 p.u. 0.6 p.u. 𝛿1 (rad) 0.6546 (+37.5 °) 0.73625 (+42.16 °) 0.6574 (+36.06 °) 0.5314 (+30.44 °) 0.66815 (+38.29 °) 0.5406 (+30.97 °) 0.4399 (+25.18 °) −1.2213 (-70°) -0.68035 (-38.98 °) 0.5858 (-33.56 °) -0.5235 (-29.98 °) -0.32185 (-18.44 °) −0.2767 (-15.85 °) -0.2748 (15.75°) 𝛿2 (rad) 0.764 (+ 43.8°) 0.769 (+44.06 °) 0.698 (+40 °) 0.624 (+35.72 °) 0.68 (+38.98 °) 0.59 (+33.85 °) 0.518 (+29.68 °) −1.2826 (-73.4 °) -0.6974 (-39.93 °) -0.6011 (-34.44 °) -0.5845 (-33.47 °) -0.3238 (-18.55 °) -0.2798 (-16.03 °) -0.2645 (-15.17 °) 𝛿3 (rad) 0.627 (+35.9 °) 0.803 (+46.04 °) 0.668 (+38.27 °) +0.631 (+36.14 °) 0.737 (+42.16 °) 0.63 (+36.14 °) 0.539 (+30.9 °) −1.2248 (70.1°) -0.6641 (-38.07 °) -0.501 (-28.7 °) -0.4346 (-24.89 °) -0.3531 (20.23°) -0.2856 (−16.36 °) -0. 2701 (-15.47 °) 𝜔1 (rad/s) +0.4%(+0.24hz) +0.3%(+0.18hz) +0.3%(+0.18hz) +0.3%(+0.18hz) +0.3%(+0.18hz) +0.3%(+0.18hz) +0.3%(+0.18hz) -0.34%(-0.204hz) -0.23%(-0.138hz) -0.21%(-0.126hz) -0.22%(-0.132hz) -0.21%(-0.126hz) -0.21%(-0.126hz) -0.18%(-0.108hz) 𝜔2 (rad/s) +0.6%(+0.36hz) +0.5%(+0.36hz) +0.5%(+0.3hz) +0.5%(+0.3hz) +0.5%(+0.3hz) +0.5%(+0.3hz) +0.5%(+0.3hz) -0.54%(-0.324hz) -0.34%(-0204hz) -0.37%(+0.222hz) -0.39%(-0.234hz) -0.33%(-0.198hz) -0.37%(-0.222hz) -0.35%(-0.21hz) 𝜔3 (rad/s) +0.7%(+0.42hz) +0.7%(+0.42hz) +0.7%(+0.42hz) +0.7%(+0.42hz) +0.7%(+0.42hz) +0.7%(+0.42hz) +0.7%(+0.42hz) -0.54%(-0.324hz) -0.32%(-0.192hz) -0.3%(-0.18hz) -0.31%(-0.186hz) -0.31%(-0.186hz) -0.3%(-0.18hz) -0.21%(-0.126hz) 𝑃dfig (p.u.) +21. 5% +27.4% +12.75% +7.83% +25.65% +10.92% +6.32% -16.75% -16.05% -5.6% -3.78% -8.8% -3.475% -2.32% 𝑖dr (p.u.) +17.01% +9.86% +7.9% +5% +2.33% +0.37% +0.42% 𝑖drmin 𝑖drmin 𝑖drmin 𝑖drmin 𝑖drmin 𝑖drmin 𝑖drmin 𝑖qr (p.u.) +56.21% +56.63% +19.3% +8.53% +78.38% +32% +16.81% 𝑖qrmin 𝑖qrmin 𝑖qrmin 𝑖qrmin 𝑖qrmin 𝑖qrmin 𝑖qrmin 𝐼statcom (p.u.) — — — — — — 𝑖statcom_max -0.3143 𝑖statcom_max -0.3105 𝑖statcom_max -0.2932 k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |823 issn: 2252-4940/© 2023. the author(s). published by cbiore decrease by -18.18%, -10.26%, and -32.26% with the statcom addition in case 3. compared to 0.4 p.u., the pss and statcom combination reduces the power peaks from (+7.83%, -5.6%) in case 2 to (+6.32%, -3.475%) in case 3, decreasing them by (a) (b) (c) (d) (e) fig. 7 time-domain simulation results (a) dfig power with (𝑷𝑫𝑭𝑰𝑮 = 0.4, 0.6 p.u.), (b) rotor dq currents (𝑷𝑫𝑭𝑰𝑮= 0.4, 0.6 p.u.), (c) power angles (𝑷𝑫𝑭𝑰𝑮 = 0.4 p.u.), (d) rotor angular speeds (𝑷𝑫𝑭𝑰𝑮 = 0.4 p.u.), (e) dfig power (𝑷𝑫𝑭𝑰𝑮 = 0.4 p.u.). (a) (b) (c) fig. 8 time-domain simulation results (a) rotor dq currents (𝑷𝑫𝑭𝑰𝑮 = 0.4 p.u.), (b) power angles (𝑷𝑫𝑭𝑰𝑮 = 0.6 p.u.), (c) rotor angular speeds (𝑷𝑫𝑭𝑰𝑮 = 0.6 p.u.). k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |824 issn: 2252-4940/© 2023. the author(s). published by cbiore 19.28% and -38.62% respectively. this also represents respective declines of -42.01% and -49.92% versus case 2. additionally, the 2% settling time for power shortens to 8.17s faster, as depicted in plot 9a. likewise, figure 9b conveys the dfig’s rotor currents. in case 2, 𝑖𝑑𝑟_𝑚𝑎𝑥 reaches +5% (𝑄𝑚𝑖𝑛 ≈ 0.0573 p.u.), while max 𝑖𝑞𝑟 is +8.53%, increasing p_dfig by +3.1%. however, in case 3, 𝑖𝑑𝑟 peaks at +0.42% (𝑄𝑚𝑖𝑛 ≈ -0.0115 p.u.), while 𝑖𝑞𝑟_𝑚𝑎𝑥 rises to +16.81%, reflecting a +2.93% 𝑃𝐷𝐹𝐼𝐺 increment per equation 12. the peaks occur at t=4.55s and t=3.128s for 𝑖𝑑𝑟 and 𝑖𝑞𝑟 consecutively. notably, case 3 has the fastest settling times at t=4.2s an d t=5.67s for the 𝑑𝑞 currents. figure 9c exhibits the statcom currents. despite being a current source, the statcom controls voltage while providing reactive power. during the t=3s fault, 𝑉6 drops -98.5% to 0.01154 p.u. the statcom responds by generating max 𝑄𝑠𝑡𝑎𝑡𝑐𝑜𝑚=2.7 mvar at maximum current to restore 𝑉6. after fault clearance at t=3.128s, it absorbs (-28.4 mvar), (-30.1 mvar), and (-30.4 mvar) for 60 mw, 40 mw, and 0.2 p.u. 𝑃𝐷𝐹𝐼𝐺 levels respectively to regulate 𝑉6. thus, higher wind penetration leads to less reactive power absorption post-fault. 𝑉6 is maintained at 1.027 p.u. at steady state regardless of wind level. the currents settle in approximately 10.5 seconds. table 2 bus voltage 𝑉1, 𝑉2, 𝑉3, 𝑉6, 𝑉10 (at t=3.5s) case 1 (no pss/no statcom) case 2 (only pss) case 3 (pss + statcom) 0.2 p.u. 0.4 p.u. 0.6 p.u. 0.2 p.u. 0.4 p.u. 0.6 p.u. 0.2 p.u. 0.4 p.u. 0.6 p.u. 𝑉1 (p.u.) +1.44%(1.04) +1.54%(1.04) +1.73%(1.04) +1.92%(1.04) +1.44%(1.04) +0.87%(1.04) +1.54%(1.04) +1.06%(1.04) +0.48%(1.04) 𝑉𝟐 (p.u.) +1.36%(1.025) +1.46%(1.025) +1.56%(1.025) +0.78%(1.025) +0.49%(1.025) +0.29%(1.025) +0.59%(1.025) +0.49%(1.025) +0.10%(1.025) 𝑉𝟑 (p.u.) +1.07%(1.025) +1.17%(1.025) +1.27%(1.025) -0.78%(1.025) -0.98%(1.025) -1.07%(1.025) -1.17%(1.025) -1.27%(1.025) -1.46%(1.025) 𝑉𝟔 (p.u.) +1.95%(1.027) +2.24(1.027) +2.43%(1.027) +1.56%(1.027) +1.46%(1.027) +1.17%(1.027) +0.19%(1.027) +0.19%(1.027) +0.1%(1.027) 𝑉𝟏𝟎 (p.u.) +3.01%(1.03) +3.01%(1.03) +3.11%(1.03) +2.43%(1.03) +2.43%(1.03) +2.33%(1.03) +1.17%(1.03) +1.17%(1.03) +1.17%(1.03) table 3 bus voltage 𝑉4, 𝑉2, 𝑉5, 𝑉7, 𝑉8 , 𝑉9, 𝑉11 (at t=3.5s) case 1 (no pss/no statcom) case 2 (only pss) case 3 (pss + statcom) 0.2 p.u. 0.4 p.u. 0.6 p.u. 0.2 p.u. 0.4 p.u. 0.6 p.u. 0.2 p.u. 0.4 p.u. 0.6 p.u. 𝑉𝟒 (p.u.) +2.52%(1.03) +1.94%(1.03) +2.04%(1.03) +2.14%(1.03) +2.14%(1.03) +2.04%(1.03) +0.78%(1.03) +0.78%(1.03) +0.87%(1.03) 𝑉𝟓 (p.u.) +5.05%(0.9988) +5.12%(0.9988) +5.23%(0.9988) +5.16%(0.9988) +4.64%(0.9988) +4.13%(0.9988) +4.32%(0.9988) +3.92%(0.9988) +3.53%(0.9988) 𝑉𝟕 (p.u.) -1%(1.027) -0.78%(1.027) -0.68%(1.027) -1.07%(1.027) -1.46%(1.027) -1.85%(1.027) -1.66%(1.027) -1.95%(1.027) -2.34%(1.027) 𝑉𝟖 (p.u.) +2.36%(1.018) +2.46%(1.018) +2.55%(1.018) +1.77%(1.018) +1.47%(1.018) +1.28%(1.018) +1.38%(1.018) +1.18%(1.018) +0.88%(1.018) 𝑉𝟗 (p.u.) -0.29%(1.035) -0.19%(1.035) -0.1%(1.035) -1.16%(1.035) -1.45%(1.035) -1.64%(1.035) -1.64%(1.035) -1.74%(1.035) -2.03%(1.035) 𝑉𝟏𝟏 (p.u.) +1.65%(1.032) +2.62%(1.032) +2.81%(1.032) +0.29%(1.032) +0.1%(1.032) +0%(1.032) -0.52%(1.032) -0.38%(1.032) -0.58%(1.032) (a) (b) (c) fig. 9 time-domain simulation results (a) dfig power (𝑷𝑫𝑭𝑰𝑮 = 0.6 p.u.), (b) rotor dq currents (𝑷𝑫𝑭𝑰𝑮 = 0.6 p.u.), (c) statcom’s current (p_dfig =0.2, 0.4 and 0.6 p.u). k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |825 issn: 2252-4940/© 2023. the author(s). published by cbiore 4.2 load flow results assessing voltage stability is crucial for power system management, as it ensures optimal voltage levels are maintained during normal operation and disturbances. this is vital for reliable functioning of the grid, requiring continuous monitoring (carson et al., 1994). considering the three test cases, tables 2 and 3 comprehensively summarize the load flow results, specifically showcasing the voltage values at buses 1 through 11. these provide insights into the voltage profiles within the system. the 0.2 p.u. wind case without controllers (bracketed values in tables 2 & 3) is used as a reference to examine the impact of increased wind injection and controller cooperation on dynamic voltage response. with similar pss and statcom time constants (𝑇𝑤 =𝑇𝑟=0.1s), t=3.5s is selected to evaluate steady-state bus voltages, allowing ~0.5s settling time. a ±10% tolerance in voltages is permitted. specifically, the postfault cleared voltages offer visibility into short-term dynamic response. key insights include (adebayo & sun, 2017; balasubramanian & singh, 2011; ma et al., 2017): • detecting dangerous electromechanical oscillations through sustained voltage swings at certain buses, indicating poorly damped modes requiring improved controls. • identifying areas slow to recover voltage, guiding placement of fast dynamic reactive reserves to support atrisk regions. • revealing dangerously depressed voltage levels right after fault clearance that could precipitate delayed instability without rapid correction. • verifying performance of voltage regulators, pss, statcoms, and generator reactive power in quickly restoring voltages, with deficiencies corrected • validating short-term transient stability margins by checking for dangerously low voltages that could signify insufficient margins. a. case 1 based on tables 2&3, with a wind integration of 0.2 p.u., the bus voltages (1, 2, 3, 6, and 10) exhibit transient overvoltage in a range of +1% to +3% due to the system's natural response to the sudden change in load. furthermore, at 0.2 p.u., bus 5 (heaviest load) has the highest voltage surge of +5.05%, while 𝑉4 reduces by -1%. with 0.4 p.u. wind, bus 10 maintains a +3% rise, 𝑉4 drops 0.58%, and buses 7 and 9 increase by 0.22% and 0.1% respectively. the other buses show smaller 0.10-0.98% changes. however, at t=22.41s, voltage instability occurs with failures predominantly at load buses 5, 6, 8 and dfig buses 10, 11, due to their low impedance connections to the faulty location. similarly, with 0.6 p.u. wind, voltages surge a further 0.10.2% at t=3.5 s across all buses. the largest increases of +0.19% occur at 𝑉1, 𝑉6, and 𝑉11. however, at t=12.8s, severe voltage instability occurs with catastrophic violations at load buses 5, 6, 8 and dfig buses 10, 11. this result from the large reactive current rush during the fault leading to sustained low frequency oscillations and complete transient voltage collapse. the results highlight the critical need to prioritize voltage regulation as wind integration increases, which is crucial for maintaining power system stability and reliability. (a) (b) (c) fig. 10 tested model eigenvalues plot with 𝑷𝑫𝑭𝑰𝑮 = 𝟎. 𝟐 𝒑. 𝒖 (a) case1, (b) case2, (c) case3. k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |826 issn: 2252-4940/© 2023. the author(s). published by cbiore b. comparison between case 2 and 3 1with 𝑃𝐷𝐹𝐼𝐺 = 0.2 𝑝. 𝑢 in case 3, adding a statcom provided enhanced voltage regulation compared to just pss control in case 2, with improvements of 1.36%, 0.84%, 1.37% and 1.26% at buses 4, 5, 6 and 10, respectively. however, minor voltage drops between 0.29-0.59% occurred at buses 3, 7, 9 and 11. the coordinated pss and statcom yielded overall positive, though not universal, voltage profile enhancements at 0.2 p.u. wind. the most substantial regulation improvements occurred near the statcom location, demonstrating its local area voltage support capabilities. 2with 𝑃𝐷𝐹𝐼𝐺 = 0.4 𝑝. 𝑢 under 40% of wind invasion, the coordinated pss and statcom controls in case 3 provided appreciable voltage regulation improvements at several buses versus sole pss control in case 2. however, select buses exhibited minor degradation versus case 2, denoting opportunities for further optimization. specifically, adding the statcom yielded notable gains of 1.36% at bus 4, 0.72% at bus 5, 1.27% at bus 6, and 1.26% at bus 10 over pss alone. conversely, regulation dropped 0.29% at bus 3, 0.49% at bus 7, 0.29% at bus 9, and 0.14% at bus 11 compared to case 2. nonetheless, the multidevice strategy facilitated considerable 0.29-1.36% enhancements at buses 1, 4, 5, 6, 8, and 10 at 0.4 p.u. wind. additional assessment of techniques minimizing deviations at specific buses remains vital for optimizing voltage profiles. 3with 𝑃𝐷𝐹𝐼𝐺 = 0.6 𝑝. 𝑢 at 0.6 p.u. wind, adding a statcom in case 3 provided appreciable voltage regulation improvements at several buses compared to just pss control in case 2. notable gains occurred at buses 1, 2, 4, 5, 6, 8 and 10. however, degradation materialized at buses 3, 7, 9, and 11. while beneficial overall, further refinement and coordination of the pss and statcom controllers remains vital for achieving consistent voltage profile enhancements across all buses, given the variability observed under 0.6 p.u. renewable integration. table 5 linear value analysis of the tested system. cases eig.n° most associated states real part img part freq damping ratio% mode/remarks case 1: 𝑃𝐷𝐹𝐼𝐺 = 0.2𝑝. 𝑢. 5-6 omega_syn_3, delta_syn_3 -0.66556 ±12.8017 2.0402 +5.2 local mode 3 7-8 delta_syn_2, omega_syn_2 -0.14151 ±8.2949 1.3204 +1.7 inter area mode (gen2, gen1) 19-20 omega_syn_1, delta_syn_1 -0.13735 ±0.83001 0.1339 +16.5 local mode 1 case 2: 𝑃𝐷𝐹𝐼𝐺 = 0.2𝑝. 𝑢. 30-31 delta_syn_1,omega_syn_1 -0.0901 ±0.64637 0.10387 +13.93 local mode 1 case 3: 𝑃𝐷𝐹𝐼𝐺 = 0.2𝑝. 𝑢. 31-32 delta_syn_1,omega_syn_1 -0.13512 ±0.67969 0.11029 +19.8 local mode 1 case 1: 𝑃𝐷𝐹𝐼𝐺 = 0.4𝑝. 𝑢. 4-5 omega_syn_3,delta_syn_3 -0.88353 ±11.7908 1.8818 +7.49 local mode 3 6-7 omega_syn_2,delta_syn_2 -0.18133 ±7.6103 1.2116 +2.38 inter area mode (gen2, gen1) 16 delta_syn_1,omega_syn_1 0.00011 0.1465 0.02332 / marginal instability 17 omega_syn_1,delta_syn_1 0.00011 -0.1465 0.02332 / marginal instability case 2: 𝑃𝐷𝐹𝐼𝐺 = 0.4𝑝. 𝑢. 30-31 delta_syn_1,omega_syn_1 -0.08991 ±0.64547 0.11909 +13.93 local mode 1 case 3: 𝑃𝐷𝐹𝐼𝐺 = 0.4𝑝. 𝑢. 31-32 delta_syn_1,omega_syn_1 -0.16356 ±0.78322 0.12734 +20.88 local mode 1 case 1: 𝑃𝐷𝐹𝐼𝐺 = 0.6𝑝. 𝑢. 4-5 delta_syn_3,omega_syn_3 -0.86609 ±11.7848 1.8807 +7.43 local mode 3 6-7 omega_syn_2,delta_syn_2 -0.17429 ±7.5708 1.2052 +2.3 inter area mode (gen2, gen1) 16 delta_syn_1,omega_syn_1 5e-05 0.09393 0.01495 / marginal instability 17 delta_syn_1,omega_syn_1 5e-05 0.09393 0.01495 / marginal instability case 2: 𝑃𝐷𝐹𝐼𝐺 = 0.6𝑝. 𝑢. 30-31 delta_syn_1,omega_syn_1 -0.13807 ±0.85751 0.13823 +16.12 local mode 1 case 3: 𝑃𝐷𝐹𝐼𝐺 = 0.6𝑝. 𝑢. 28-29 omega_syn_1,delta_syn_1 -0.40367 ±1.1875 0.19962 +34 local mode 1 table 4 overall power statistics. case 1 (no pss/no statcom) case 2 (only pss) case 3 (pss + statcom) 0.2 p.u. 0.4 p.u. 0.6 p.u. 0.2 p.u. 0.4 p.u. 0.6 p.u. 0.2 p.u. 0.4 p.u. 0.6 p.u. p 𝑔𝑒𝑛 (p.u.) 3.1953 3.1961 3.1986 3.1953 3.1961 3.1986 3.1953 3.1961 3.1986 q 𝑔𝑒𝑛 (p.u.) 0.0402 0.05017 0.08459 0.0402 0.05017 0.08459 0.03657 0.0471 0.08133 p 𝑙𝑜𝑎𝑑 (p.u.) 3.15 3.15 3.15 3.15 3.15 3.15 3.15 3.15 3.15 q 𝑙𝑜𝑎𝑑 (p.u.) 1.15 1.15 1.15 1.15 1.15 1.15 1.15 1.15 1.15 p 𝑙𝑜𝑠𝑠 (p.u.) 0.04526 0.04605 0.0486 0.04526 0.04605 0.0486 0.04527 0.04605 0.0486 q 𝑙𝑜𝑠𝑠 (p.u.) -1.1098 -1.0998 -1.0654 -1.1098 -1.0998 -1.0654 -1.1096 -1.0988 -1.0645 k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |827 issn: 2252-4940/© 2023. the author(s). published by cbiore table 4 provides a clear and concise presentation of the power data, including total active and reactive generated power, load, and losses. it offers a comprehensive overview of the power flow within the system, enabling a detailed analysis of the different components involved table 4 clearly and concisely summarizes the power flow data, including total active and reactive generation, load, and losses. it provides a comprehensive overview enabling detailed analysis. for power losses, cases 1 and 2 are identical since psss do not directly impact them. in both, at 0.2 p.u. wind, reactive power loss is highest at -1.1098 p.u. due to the low 𝑄𝑔𝑒𝑛 0.22885 p.u. from units 1 and 2, while the dfig and unit 3 absorb 0.18865 p.u. as wind increases to 0.4 p.u., the reactive power loss shrinks to -1.0998 p.u., due to higher 0.24311 p.u. 𝑄𝑔𝑒𝑛 and 0.19224 p.u. absorption. at 60 mw dfig output, 𝑄𝑙𝑜𝑠𝑠 further reduces to -1.0654 p.u. with 0.26845 p.u. 𝑄𝑔𝑒𝑛 and 0.18385 p.u. absorption. furthermore, from 0.2 to 0.6 p.u. 𝑃𝐷𝐹𝐼𝐺 in case 3, reactive power loss drops from -1.1096 p.u. to -1.0645 p.u., a 4.06% reduction. higher dfig output leads to increased reactive absorption, as the dfig operates at a lagging power factor as output rises. however, the statcom can increase its output to compensate. this additional statcom contribution helps maintain low reactive power losses as wind penetration increases (qiao et al., 2009, shahgholian & izadpanahi, 2016). 4.3 eigenvalue analysis this study employed the eigenvalue analysis tool that is incorporated in the psat toolbox. a total of 39 states were initiated and analyzed, which comprise 13 complex pairs. figures 10-12 present the eigenvalues plots regarding the tested system. based on the results of the eigenvalue analysis, table 5 presents further details on the dominant oscillation modes. participation factors were utilized to identify the most weaklydamped buses susceptible to instability. specifically, the rotor angles and velocities of the three generating units were identified as the weakest states. the associated generator states indicate the presence of both local and inter-area oscillation modes in the system. thus, the investigation will incrementally analyze the 3 cases at wind penetration levels of 0.2 p.u., 0.4 p.u., and 0.6 p.u. to assess system stability and oscillatory response. the following sections detail the results at each penetration level: 10.2 p.u. the system was determined to be stable, as evidenced by all eigenvalues residing in the left half of the complex plane. however, as depicted in figure 10a, case 1 appeared the most susceptible to potential instability, with eigenvalues positioned closest to the imaginary axis. in contrast, cases 2 and 3, as shown in figures 10b and 10c, demonstrated more robust stability margins, with all eigenvalues farther into the left halfplane (no positive eigenvalues were present). hence, final observations from 0.2 p.u. wind penetration: case 1 exhibited multiple modes, including both local generator oscillations and an inter-area oscillation: • a local mode of generator 3 at 2.0402 hz (modes 5-6) with 5.2% damping. (a) (b) (c) fig. 11 tested model eigenvalues plot with 𝑷𝑫𝑭𝑰𝑮 = 𝟎. 𝟒 𝒑. 𝒖 (a) case1, (b) case2, (c) case3. k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |828 issn: 2252-4940/© 2023. the author(s). published by cbiore • an inter-area mode between generators 1 & 2 at 1.3204 hz (modes 7-8) with very light damping ratio of 1.7%. • a local under-damped mode of generator 1 at 0.1339 hz (modes 19-20). cases 2 and 3 only showed localized oscillation modes of generator 1: • case 2 had a generator 1 local mode at 0.10387 hz (modes 30-31) with 13.94% damping. • case 3 had a generator 1 local mode at 0.11029 hz (modes 31-32) with damping of 19.9%. in summary, case 1 displayed multiple modes with lighter damping, including risky inter-area oscillations. cases 2 and 3 had better damped local generator modes indicating more stable operation. 20.4 p.u. figure 11a and table 5 reveal that when the wind power integration level increased to 0.4 p.u., case 1 became marginally unstable because of the emergence of 2 positive eigenvalue in states n°16/17 (i.e., (delta_syn_1, omega_syn_1) and (omega_syn_1, delta_syn_1) with a very low real part of +1.1e4). with the application of pss and statcom, as portrayed in figures 11b and 11c, the damping of oscillatory modes was increased, restoring stability margins and stabilizing the system. furthermore, we conclude: a-for case 1 at 0.4 p.u. case 1 at 0.4 p.u. loading exhibits multiple concerning oscillatory modes: the eigenvalues investigation pointed several suboptimal system dynamics needing further review. a 1.8818 hz under-damped local mode of generator 3 was found, having only 7.5% damping. also identified was a concerning 1.2116 hz inter-area mode between generators 1 & 2 with marginal damping of 2.4%. most significantly, modes 16-17 exhibited sustained 0.02332 hz oscillations with zero damping and a slightly positive real part of +1.1e-04. this corresponds to very low frequency oscillations at 0.02332 hz, or a 42.84 second period. the problematic oscillations involve states delta_syn_1 and omega_syn_1 of generator 1. while these oscillations are extremely lightly damped, the system remains practically stable over reasonable timeframes. however, such sustained oscillations are atypical and suggest potential modeling or control issues that should be investigated further. in summary, while still stable, this investigation indicated multiple suboptimal system modes and oscillatory dynamics that warrant deeper review and mitigation efforts. b-for cases 2 and 3 at 0.4 p.u. local generator modes were observed, reasonably damped in case 2 at 13.93% and well-damped in case 3 at 20.88%. to recap, targeted implementation of supplemental damping controllers enabled stable operation of case 1 at increased 0.4 p.u. wind integration, despite marginally unstable oscillatory modes, by eliminating positive eigenvalues and restoring stability margins. 30.6 p.u. finally, the eigenvalue plots for 𝑷𝑫𝑭𝑰𝑮=0.6 p.u. are shown in figure 12. similarly to case 1 with 0.4 p.u., 2 positive (a) (b) (c) fig. 12 tested model eigenvalues plot with 𝑷𝑫𝑭𝑰𝑮 = 𝟎. 𝟔 𝒑. 𝒖. (a) case1, (b) case2, (c) case3. k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |829 issn: 2252-4940/© 2023. the author(s). published by cbiore eigenvalues emerge, causing instability in the slack bus synchronous generator, as seen in fig 12a. states n°16 (delta_syn_1, omega_syn_1) and n°17 (omega_syn_1, delta_syn_1) with extremely small positive real parts of 5e-5 and imaginary parts of +/0.09393, as revealed in table 5, are the primary causes of this marginal instability. in spite of this, plots 12b and 12c demonstrate how the system remains stable despite significant wind penetration, thanks to the controllers' effects. key findings from the analysis at 0.6 p.u. wind penetration include: a-for case 1 at 0.6 p.u. case 1 at 0.6 p.u. wind integration displays multiple problematic oscillatory modes: the modal analysis revealed several concerning dynamics that warrant further investigation. a risky 1.8807 hz under-damped local mode of generator 3 was identified, with only 7.35% damping. additionally, a hazardous 1.2052 hz inter-area mode between generators 1 & 2 exhibited marginal damping of just 2.3%. most critically, modes 16-17 showed sustained 0.01495 hz oscillations with near-zero damping and a slightly positive real part of +5e-05. this points to extremely low frequency oscillations at 0.01495 hz, corresponding to a 66.8 second period. these problematic oscillations involve states delta_syn_1 and omega_syn_1 of generator 1. while technically stable from a positive real part, such dynamics are practically unstable and indicate potential modeling or control issues that should be looked into. overall, this analysis surfaced multiple areas of the system displaying unfavorable dynamics and oscillations that require mitigation. b-for cases 2 and 3 at 0.6 p.u. cases 2 and 3 maintain adequately damped local generator modes. in a nutshell, at 0.6 p.u. wind power, case 1 displayed persistent, lightly-damped local and inter-area oscillations, requiring damping control. case 1 showed a 7.35% damped generator 3 local mode and concerning 2.3% damped inter-area mode. in contrast, cases 2 and 3 maintained stability using coordinated pss-statcom damping, achieving 34% damped oscillations in case 3. this demonstrates the value of complementary damping control to enable higher renewable integration through targeted oscillation mitigation. 5. conclusion this study analyzes the dynamic response and transient stability of a modified wscc 9-bus system integrated with a dfig wind farm under different control configurations. the results demonstrate that a coordinated pss and statcom control strategy provides the best damping performance and stability for the system. three operating scenarios were simulated, including no supplementary controls, pss control only, and combined pss+statcom control. the eigenvalue analysis reveals that as wind penetration increases, the system may exceed safe operating limits without proper controls, leading to instability. the pss+statcom configuration offers significantly improved damping ratios and time-domain responses compared to the other cases. while the complexity of the power system model constrained the extent of this study, the findings highlight the importance of robust control strategies to maintain stability with high wind power integration. in particular, coordinated pss and statcom controls are shown to be effective in damping oscillations and transient responses caused by disturbances. future research should focus on analyzing the impacts of increased renewable penetration on large, complex power grids. detailed small-signal stability and transient stability studies can identify critical oscillation modes and limit operating conditions to prevent instability. further work is also needed to develop practical and optimized implementable control solutions that ensure robust system performance under various contingencies and dynamic events. in summary, this study demonstrates the benefit of a coordinated pss and statcom control strategy for stability enhancement in power systems with high wind penetration. the findings provide an initial platform for further research into implementing robust controls for future renewable-dominant grids. acknowledgments i would like to acknowledge anyone who helped with this paper author contributions: khaled kouider: conceptualization (lead); methodology (lead); validation (equal); writing – original draft (lead); writing – review & editing (equal). abdelkader bekri: supervision (lead); validation(equal); writing – review & editing (equal). funding: the author(s) received no financial support for the research, authorship, and/or publication of this article. conflicts of interest: the authors declare no conflict of interest. references adebayo, i. g., & sun, y. (2017). new performance indices for voltage stability analysis in a power system. energies, 10(12). https://doi.org/10.3390/en10122042 agarala, a., bhat, s. s., mitra, a., zychma, d., & sowa, p. (2022). transient stability analysis of a multi-machine power system integrated with renewables. energies, 15(13). https://doi.org/10.3390/en15134824 alsakati, a. a., vaithilingam, c. a., alnasseir, j., & jagadeeshwaran, a. (2022). simplex search method driven design for transient stability enhancement in wind energy integrated power system using multi-band pss4c. ieee access., 14, 83913–83928. https://doi.org/10.1109/access.2021.3085976 asija, d., choudekar, p., soni, k. m., & sinha, s. k. (2015). power flow study and contingency status of wscc 9 bus test system using matlab. https://dx.doi.org/10.1109/rdcape.2015.7281420 avdakovic, s., nuhanovic, a., kusljugic, m., becirovic, e., & music, m. (2009). identification of low frequency oscillations in power system. 2009 international conference on electrical and electronics engineering eleco 2009. https://doi.org/10.1109/eleco.2009.5355325 bagchi, s., goswami, s., bhaduri, r., ganguly, m., & roy, a. (2016). small signal stability analysis and comparison with dfig incorporated system using facts devices. 2016 ieee 1st international conference on power electronics, intelligent control and energy systems (icpeices) https://dx.doi.org/10.1109/icpeices.2016.7853294 balasubramanian, r., & singh, r. (2011). power system voltage stability analysis using ann and continuation power flow methods. 2011 16th international conference on intelligent system applications to power systems, 1–7. https://dx.doi.org/10.1109/isap.2011.6082192 cai, e., dong, l., & liao, x. (2021). a reduced-order model for representing the reactive power response of a doubly fed wind turbine during zvrt. journal of renewable and sustainable energy, 13(5), 053310. https://doi.org/10.1063/5.0056451 https://doi.org/10.3390/en10122042 https://doi.org/10.3390/en15134824 https://doi.org/10.1109/access.2021.3085976 https://dx.doi.org/10.1109/rdcape.2015.7281420 https://doi.org/10.1109/eleco.2009.5355325 https://dx.doi.org/10.1109/icpeices.2016.7853294 https://dx.doi.org/10.1109/isap.2011.6082192 https://doi.org/10.1063/5.0056451 k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |830 issn: 2252-4940/© 2023. the author(s). published by cbiore carson w. taylor, neal j. balu, dominic maratukulam. (1994). power system voltage stability. mcgraw-hill, new york, chandra, d. r., kumari, m. s., sydulu, m., grimaccia, f., mussetta, m., leva, s., duong, m.q. (2014). impact of scig, dfig wind power plant on ieee 14 bus system with small signal stability assessment. 2014 eighteenth national power systems conference (npsc) https://dx.doi.org/10.1109/npsc.2014.7103883 denholm, paul, trieu mai, rick wallace kenyon, ben kroposki, and mark o’malley. (2020). inertia and the power grid: a guide without the spin (nrel/tp-6120-73856). national renewable energy laboratory (nrel). edrah, m., lo, k. l., & anaya-lara, o. (2015). impacts of high penetration of dfig wind turbines on rotor angle stability of power systems. ieee transactions on sustainable energy, 6(3), 759–766. https://doi.org/10.1109/tste.2015.2412176 edrah, m., lo, k. l., & anaya-lara, o. (2016). reactive power control of dfig wind turbines for power oscillation damping under a wide range of operating conditions. iet generation, transmission & distribution, 10(15), 3777–3785. https://doi.org/10.1049/ietgtd.2016.0132 eriksson, r., modig, n., & elkington, k. (2018). synthetic inertia versus fast frequency response: a definition. iet renewable power generation, 12(5), 507–514. https://doi.org/10.1049/ietrpg.2017.0370 eremia, m., & shahidehpour. (2013). handbook of electrical power system dynamics: modeling, stability, and control. wiley. eshkaftaki, a. a., rabiee, a., kargar, a., & boroujeni, s. t. (2020). an applicable method to improve transient and dynamic performance of power system equipped with dfig-based wind turbines. ieee transactions on power systems, 35(3), 2351–2361. https://doi.org/10.1109/tpwrs.2019.2954497 falehi, a. d., rostami, m., doroudi, a., & ashrafian, a. (2012). optimization and coordination of svc-based supplementary controllers and psss to improve power system stability using a genetic algorithm. turkish journal of electrical engineering and computer sciences, 20(5), 639–654. https://doi.org/10.3906/elk1010-838 fdaili, m., essadki, a., kharchouf, i., & nasser, t. (2021). noncontrolled fault current limiter with reactive power support for transient stability improvement of dfig-based variable speed wind generator during grid faults. iet generation, transmission & distribution, 32(8). https://doi.org/doi.org/10.1002/20507038.12955 gurung, r., n. bhattarai, & kamalasadan, s. (2020). optimal oscillation damping controller design for large-scale wind integrated power grid. ieee transactions on industry applications, 56(4), 4225–4235. https://doi.org/10.1109/tia.2020.2988432 hatziargyriou, n., milanovic, j., rahmann, c., ajjarapu, v., canizares, c., erlich, i., hill, d., hiskens, i., kamwa, i., pal, b., pourbeik, p., sanchez-gasca, j., stankovic, a., van cutsem, t., vittal, v., & vournas, c. (2021). definition and classification of power system stability – revisited & extended. ieee transactions on power systems, 36(4), 3271–3281. https://doi.org/10.1109/tpwrs.2020.3041774 he, p., fang, q., jin, h., ji, y., gong, z., & dong, j. (2022). coordinated design of pss and statcom-pod based on the ga-pso algorithm to improve the stability of wind-pv-thermal-bundled power system. international journal of electrical power & energy systems, 141, 108208. https://doi.org/10.1016/j.ijepes.2022.108208 hemeida, m. g., rezk, h., & hamada, m. m. (2018). a comprehensive comparison of statcom versus svc-based fuzzy controller for stability improvement of wind farm connected to multi-machine power system. electrical engineering, 100(2), 935–951. https://doi.org/10.1007/s00202-017-0559-6 klein, m., rogers, g. j., & prabha, k. (1991). a fundamental study of inter-area oscillations in power systems. ieee transactions on power systems, 6(3), 914–921. https://doi.org/10.1109/59.119229 kothari, i. j., d. p. nagrath. (2019). power system engineering, 3e. mcgraw-hill education pvt. ltd. kumar, s., kumar, a., & sharma, n. k. (2020). a novel method to investigate voltage stability of ieee-14 bus wind integrated system using psat. frontiers in energy, 14(2), 410–418. https://doi.org/10.1007/s11708-016-0440-8 li, s., zhang, h., yan, y., & ren, j. (2022). parameter optimization to power oscillation damper (pod) considering its impact on the dfig. frontiers in energy, 37(2), 1508–1518. https://doi.org/10.1109/tpwrs.2021.3104816 liu, y., liu, q., hu, f., xu, y., chen, x., & chen, s. (2021). transient stability assessment for power system with wind farm considering the stochasticity. international transactions on electrical energy systems, 14(2). https://doi.org/10.1002/2050-7038.12854 ma, y., lv, s., zhou, x., & gao, z. (2017). review analysis of voltage stability in power system. 2017 ieee international conference on mechatronics and automation (icma), 7–12. https://dx.doi.org/10.1109/icma.2017.8015779 milano, f. (2008). power system analysis toolbox version 2.0.0. university college dublin. miller, t. j. e. (2010). theory of the doubly-fed induction machine in the steady state. the xix international conference on electrical machines icem 2010, 1–6. https://doi.org/10.1109/icelmach.2010.5607742 morshed, m. j. (2020). a nonlinear coordinated approach to enhance the transient stability of wind energy-based power systems. ieee/caa journal of automatica sinica, 7(4), 1087–1097. https://doi.org/10.1109/jas.2020.1003255 morshed, m. j., & fekih, a. (2019). a probabilistic robust coordinated approach to stabilize power oscillations in dfig-based power systems. ieee transactions on industrial informatics, 15(10), 5599– 5612. https://doi.org/10.1109/tii.2019.2901935 muñoz, j. c., & cañizares, c. a. (2011). comparative stability analysis of dfig-based wind farms and conventional synchronous generators. 2011 ieee/pes power systems conference and exposition. https://doi.org/10.1109/psce.2011.5772545 nkosi, n. r., bansal, r. c., adefarati, t., naidoo, r. m., & bansal, s. k. (2023). a review of small-signal stability analysis of dfig-based wind power system. international journal of modelling and simulation, 43(3), 153–170. https://doi.org/10.1080/02286203.2022.2056951 pérez-londoño, s., rodríguez-garcía, l., & lópez, y. u. (2012). effects of doubly fed wind generators on voltage stability of power systems. 2012 sixth ieee/pes transmission and distribution: latin america conference and exposition (t&d-la). pillai, a. g., thomas, p. c., sreerenjini, k., baby, s., joseph, t., & srecdharan, s. (2013). transient stability analysis of wind integrate power systems with storage using central area controller. 2013 international conference on microelectronics, communications and renewable energy (icmicr-2013). prabha, k. (1994). power system stability and control. mcgraw-hill education, new york. qiao, w., venayagamoorthy, g. k., & harley, r. g. (2009). real-time implementation of a statcom on a wind farm equipped with doubly fed induction generators. ieee transactions on industry applications, 45(1), 98–107. https://doi.org/10.1109/tia.2008.2009377 ramanujam, r. (2010). power system dynamics: analysis and simulation. phi learning pvt. ltd. reza, m., slootweg, j. g., schavemaker, p. h., kling, w. l., & van der sluis, l. (2003). investigating impacts of distributed generation on transmission system stability. 2003 ieee bologna power tech conference proceedings, 2, 7 pp. vol.2-. https://doi.org/10.1109/ptc.2003.1304341 shabani, h. r., & kalantar, m. (2021). real-time transient stability detection in the power system with high penetration of dfigbased wind farms using transient energy function. international journal of electrical power & energy systems, 131. https://doi.org/10.1016/j.ijepes.2021.107319 shabani, h. r., kalantar, m., & hajizadeh, a. (2021). investigation of the closed-loop control system on the dfig dynamic models in transient stability studies. international journal of electrical power & energy systems, 131. https://doi.org/10.1016/j.ijepes.2021.107084 shahgholian, g., & izadpanahi, n. (2016). improving the performance of wind turbine equipped with dfig using statcom based on input-output feedback linearization controller. environmental engineering science, 4, 65–79. https://dx.doi.org/10.1109/npsc.2014.7103883 https://doi.org/10.1109/tste.2015.2412176 https://doi.org/10.1049/iet-gtd.2016.0132 https://doi.org/10.1049/iet-gtd.2016.0132 https://doi.org/10.1049/iet-rpg.2017.0370 https://doi.org/10.1049/iet-rpg.2017.0370 https://doi.org/10.1109/tpwrs.2019.2954497 https://doi.org/10.3906/elk-1010-838 https://doi.org/10.3906/elk-1010-838 https://doi.org/doi.org/10.1002/2050-7038.12955 https://doi.org/doi.org/10.1002/2050-7038.12955 https://doi.org/10.1109/tia.2020.2988432 https://doi.org/10.1109/tpwrs.2020.3041774 https://doi.org/10.1016/j.ijepes.2022.108208 https://doi.org/10.1007/s00202-017-0559-6 https://doi.org/10.1109/59.119229 https://doi.org/10.1007/s11708-016-0440-8 https://doi.org/10.1109/tpwrs.2021.3104816 https://doi.org/10.1002/2050-7038.12854 https://dx.doi.org/10.1109/icma.2017.8015779 https://doi.org/10.1109/icelmach.2010.5607742 https://doi.org/10.1109/jas.2020.1003255 https://doi.org/10.1109/tii.2019.2901935 https://doi.org/10.1109/psce.2011.5772545 https://doi.org/10.1080/02286203.2022.2056951 https://doi.org/10.1109/tia.2008.2009377 https://doi.org/10.1109/ptc.2003.1304341 https://doi.org/10.1016/j.ijepes.2021.107319 https://doi.org/10.1016/j.ijepes.2021.107084 k. kouider and a. bekri int. j. renew. energy dev 2023, 12(5), 816-831 |831 issn: 2252-4940/© 2023. the author(s). published by cbiore https://www.energyequipsys.com/article_20128_8c989427ad8 dfacc589a02518a7e584f.pdf simani, s., & farsoni, s. (2018). introduction. in farsoni. s. simani. s (ed.), fault diagnosis and sustainable control of wind turbines (pp. 1–12). butterworth-heinemann, elseiver. simon, l., ravishankar, j., & swarup, k. s. (2019). coordinated reactive power and crow bar control for dfig-based wind turbines for power oscillation damping. wind engineering, 43(2), 95–113. https://doi.org/10.1177/0309524x18780385 tang, w., hu, j., chang, y., & liu, f. (2018). modeling of dfig-based wind turbine for power system transient response analysis in rotor speed control timescale. ieee transactions on power systems, 33(6), 6795–6805. https://doi.org/10.1109/tpwrs.2018.2827402 thanpisit, k., & ngamroo, i. (2017). power oscillation damping control by pss and dfig wind turbine under multiple operating conditions. turkish journal of electrical engineering and computer sciences, 25(5), 4354–4368. https://doi.org/10.3906/elk-1702190 xia, s., zhang, q., hussain, s. t., hong, b., & zou, w. (2018). impacts of integration of wind farms on power system transient stability. applied sciences, 8(8). https://doi.org/10.3390/app8081289 xu, y., chi, y., & yuan, h. (2023). power system stability. in stabilityconstrained optimization for modern power system operation and planning (pp. 5–18). https://doi.org/10.1002/9781119848899.ch1 yang, b., jiang, l., wang, l., yao, w., & wu, q. h. (2016). nonlinear maximum power point tracking control and modal analysis of dfig based wind turbine. international journal of electrical power & energy systems, 74, 429–436. https://doi.org/10.1016/j.ijepes.2015.07.036 yang, l., xu, z., østergaard, j., dong, z. y., wong, k. p., & x, ma. (2011). oscillatory stability and eigenvalue sensitivity analysis of a dfig wind turbine system. ieee transactions on energy conversion, 26(1), 328–339. https://doi.org/10.1109/tec.2010.2091130 yu, s. s., chau, t. k., fernando, t., & iu, h. h. (2018). an enhanced adaptive phasor power oscillation damping approach with latency compensation for modern power systems. ieee transactions on power systems, 33(4), 4285–4296. https://doi.org/10.1109/tpwrs.2017.2773632 yunus, a. m. s., saini, m., djalal, m. r., abu‐siada, a., & masoum, m. a. s. (2019). impact of superconducting magnetic energy storage unit on doubly fed induction generator performance during various levels of grid faults. international review of electrical engineering (iree), 14. https://doi.org/10.15866/iree.v14i4.16472 zhang, c., ke, d., sun, y., chung, c. y., xu, j., & shen, f. (2018). coordinated supplementary damping control of dfig and pss to suppress inter-area oscillations with optimally controlled plant dynamics. ieee transactions on sustainable energy, 9(2), 780–791. https://doi.org/10.1109/tste.2017.2761813 zheng, d., jinxin, o., xiaofu, x., & mengyang, l. (2019). rotor angle stability control for dfig-integrated power system considering phase-amplitude characteristics of transient-grid voltage. iet generation, transmission & distribution, 13(16), 3549–3555. https://doi.org/10.1049/iet-gtd.2018.6960 © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/ ) https://www.energyequipsys.com/article_20128_8c989427ad8dfacc589a02518a7e584f.pdf https://www.energyequipsys.com/article_20128_8c989427ad8dfacc589a02518a7e584f.pdf https://doi.org/10.1177/0309524x18780385 https://doi.org/10.1109/tpwrs.2018.2827402 https://doi.org/10.3906/elk-1702-190 https://doi.org/10.3906/elk-1702-190 https://doi.org/10.3390/app8081289 https://doi.org/10.1002/9781119848899.ch1 https://doi.org/10.1016/j.ijepes.2015.07.036 https://doi.org/10.1109/tec.2010.2091130 https://doi.org/10.1109/tpwrs.2017.2773632 https://doi.org/10.15866/iree.v14i4.16472 https://doi.org/10.1109/tste.2017.2761813 https://doi.org/10.1049/iet-gtd.2018.6960 http://creativecommons.org/licenses/by-sa/4.0/ international journal of renewable energy development int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 77 © ijred – issn: 2252-4940, july 15,2015, all rights reserved contents list available at ijred website int. journal of renewable energy development (ijred) journal homepage: http://ejournal.undip.ac.id/index.php/ijred historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency nisith raval and ajay kumar gupta* centre of excellence in nanotechnology (coe-nt), confederation of indian industry (cii), cii house, gulbai tekra road, near panchavati, ahmedabad 380006. gujarat, india. abstract: sun is the continuous source of renewable energy, from where we can get abundant of solar energy. concept of conversion of solar energy into heat was used back in 200 b.c. since then, the solar cells have been developed which can convert solar energy into the electrical energy and these systems have been produced commercially. the technologies to enhance the power conversion efficiency (pce) have been continuously improved. different technologies used for developing solar cells can be categorized either on the basis of material used or techniques of technology development which is further termed as ‘first generation’ (e.g. crystalline silicon), ‘second generation’ (thin films of amorphous silicon, copper indium gallium selenide, cadmium telluride), ‘third generation’ (concentrated, organic and dye sensitize solar cell). these technologies give pce up to 25% depending on the technology and the materials used. nanotechnology enables the use of nanomaterial whose size is below 100 nm with extraordinary properties which has the capability to enhance the pce to greater extent. various nanomaterials like quantum dots, quantum well, carbon nanotubes, nanowire and graphene have been used to make efficient and economical solar cells, which not only provide high conversion efficiency economically but also are easy to produce. today, by using nanotechnology, conversion efficiency up to 44.7 % has been achieved by fraunhofer institute at germany. in this review article, we have reviewed the literature including various patents and publications, summarized the history of solar cell development, development of different technologies and rationale of their development highlighting the advantages and challenges involved in their development for commercial purpose. we have also included the recent developments in solar cell research where different nanomaterials have been designed and used successfully to prove their superiority over conventional systems. keywords: solar cell, nanotechnology, renewable energy, photovoltaic, power conversion efficiency article history: received march 20, 2015; received in revised form may 12, 2015; accepted june 10, 2015; available online how to cite this article: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2), 77-93. http://dx.doi.org/10.14710/ijred.4.2-77-93 * ajay gupta: phone: +917940279996; fax : +917940279999 email: ajay.gupta@cii.in 1. solar cell: historic developments and the current scenario by the year 2050, global energy demand will be more than two times and it may be three times by the end of this century. the scope, to compete these increasing demand is the major challenges for achieving technological advancements, economic growth, and political stability. advancement of existing technologies will not bridge the gaps between today’s need and tomorrow’s requirements. it must require additional source of energy and technologies to harvest the energy efficiency from these source over the next half-century with the capacity to double the today’s energy output. the sun is the natural source of energy which delivers more energy to the earth in an hour than we use in a year from fossil, nuclear, and all other renewable sources together. currently we are using only 0.1 % of our primary energy from sunlight. in the last few decades, there had been growing research and developments in the area of solar cell technologies where the sunlight can be used as a source to generate electricity through photovoltaic effect. however this concept of solar technology is not new and find references as early as in centuries before christ (b.c.). in 3rd century b.c., burning mirrors were http://dx.doi.org/10.14710/ijred.4.2-77-93 citation: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2),77-93, dx.doi.org/10.14710/ijred.4.2.77-93 p a g e | 78 © ijred – issn: 2252-4940, july 15, 2015, all rights reserved used to light torches by greeks and romans for religious purposes (anuradha, lovish & pranjal 2013). later, in 212 b.c., the greek scientist, archimedes used sunlight to set fire to wooden ships by using the reflective properties of bronze shields. during the era of 1st to 17th centuries, there were many reports indicating the use of solar energy in houses and public buildings for getting the warmth in winter seasons. during the next two centuries (i.e. 18th and 19th century), there had been many developments and breakthroughs in the use of solar technologies. these developments are reported in the table 1. all these milestones utilize different processes / technologies to harvest solar energy for use by the society (u.s department of energy 2015). however, still there was no scientific discovery or mechanism available which could be used at a commercial scale by mass public. table 1 history of solar cell (1700 – 1899) sr. no. year historical developments / milestones 1 1767  world’s first solar collector / cooker was developed to cook the foods by swiss scientist horace de saussure. 2 1816  robert stirling and lord kelvin developed a thermo-solar cell to concentrate sun’s thermal energy to produce power. 3 1839  french scientist edmon becquerel discovered the photovoltaic effect for the first time by using an electrolytic cell using electrolysis phenomenon. 4 1860  french august mouchet developed the solar powered steam engine for industrial applications. 5 1873  w. smith discovered the photoconductivity of selenium. 6 1876  william grylls adams and richard e. day discovered the concept of selenium based solar cells and later in year 1883 charles fritts developed the first selenium wafer based solar cells. 7 1880  samuel p langley invented bolometer to measure light from the sunrays. 8 1891  c. kemp patented the first commercial solar water heater. 9 1888 –1897  several scientists granted patents on solar cells. however, at the same time (during the 18th and 19th century) several significant developments took place in the form of the industrial revolution which includes the machine based process, new chemical manufacturing, efficient hydro power production, increase the acceptance of steam and coal as a fuel source. however, as the industrialization progressed across the sectors, demand for the energy kept on rising. earlier, the machines were operated manually which gradually started shifting to automatic operation where human intervention was minimal. these modern machines required energy (e.g. electrical, hydrothermal, steam engines, etc.). due to this, there was growing need of energy for industry as well as society. 1.1 birth of modern technology in order to fulfil the future energy requirements, it was highly desirable to develop the technologies and devices which can capture the sun light and convert it into the form which can be put to use by the society. the beginning of 20th century brought more knowledge and improvement in the technologies used to harvest solar energy. table 2 gives the glimpse of the developments happened during the century (u.s department of energy 2015; affordable solar 2015; wikipedia, timeline of solar cell 2015; experience 2015). table 2 history of solar cell (1900 – 2010) sr. no. year historical developments / milestones 1 1900 –1950  albert einstein discovered the photo-electric effect (year 1905).  carnegie steel company invents copper coil based solar collector (year 1908).  silicon (single crystalline) was developed by jan czochralski, a polish scientist (year 1918).  photovoltaic effect was discovered in cadmium sulfide based solar cell by audobert and stora (year 1932).  passive solar building in us was built (year 1947). 2 1951 – 1960  bell labs invented solar cell (silicone) with about 6% efficiency to run daily equipment’s (year 1954).  western electric licenses commercial solar cell technologies (year 1955).  between years 1957 – 1960, hoffman electronics achieved 8 – 14 % efficient in solar cell.  silicon sensors started producing selenium and silicon solar cells (year 1960). 3 1961 – 1970  the telstar communications satellite is powered by solar cells (year 1962).  sharp corporation produces a module of silicon based solar cells (year 1963).  japan installed a 242 watt solar cell array on a light house (year 1963)  first manned spacecraft (soyuz 1) was powered by solar cells (year 1967).  exon corporation produced cheaper solar cells which could be used in remote locations where grid connections were not feasible (year 1970). 4 1971 – 1980  first amorphous silicon solar cells were created by david carlson and christopher wronski, with 1.1% efficiency (year 1976).  the world production of photovoltaic cells exceeded 500 kw (year 1977).  salyut-1, skylab, florida solar energy centre, the university of delaware’s “solar one” residential building, solar energy research int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 79 © ijred – issn: 2252-4940, july 15,2015, all rights reserved institute at colorado, solar panels in white house, usa, calculators etc. powered by solar energy were developed in this decade.  first thin film solar cell was developed by university of delaware with 10% efficiency using cu2s/cds technology (year 1980). 5 1981 – 1990  arco solar produces > 1 mw of solar modules/year in california (year 1982).  volkswagen, germany tested solar arrays on the roof of dasher station wagons (year 1982).  global photovoltaic production exceeds 21.3 megawatts (year 1983).  centre for photovoltaic engineering created 20% efficient silicon cells at the university of new south wales (year 1985).  world’s largest solar thermal facility located in kramer junction, ca, was commissioned (year 1986).  arco solar released first thin film power module (year 1986).  michael gratzel and brian o'regan created world cheapest dye sensitize solar cell (year 1988).  dr. alvin marks granted patents for lepcon and lumeloid (year 1988). 6 1991 – 2000  national renewable energy laboratory was established by president george h. w. bush (year 1991 – 1994).  university of south florida achieved 15.89 % efficient in thinfilm cell (year 1992).  nrel achieve 30 % efficiency using gainp/gaas two terminal concentrator cell (year 1994).  graetzel, switzerland achieves 11% efficient energy conversion using a photo electrochemical effect in dyesensitized cells that (year 1996).  total worldwide installed photovoltaic power reaches 1,000 megawatts (year 1999).  spectrolabs, inc. and nrel developed solar cells with an efficiency of 32.3% (year 1999).  nrel achieved 18.8% efficient in thin film solar cells (year 1999). 7 2001 – 2010  bp solarex developed highest solar modules of 10.8 % conversion efficiency achieved by 0.5 sq. meter and 10.6 % achieved by 0.9 sq. meter (year 2000).  terra sun llc developed unique method holographic films to concentrate sunlight onto a solar cell (year2001).  power light corporation developed word largest hybrid system (wind and solar energy) to maximize the power generation (year 2001).  the university of delaware achieved a new efficiency of 42.8% (year 2007).  china produced nearly half (1700 mw) of the world production (3800 mw) of solar panels (year 2007).  the installation cost of pv modules decreased from 16000 $/kw in 1990’s to 6000 $/kw (year 2008).  world solar cell production reached 9,340 mw (year 2009). solar cell is an electrical device which converts the light energy into electric energy by photovoltaic effect. in simple words, upon exposure of solar cell with light, it can generate an electric current without being attached to any external voltage source. but do require an external load for power consumption. recent technologies like screen printing and solar fabric can enable to install the solar cell inside the house, even on the roof. today international market have opened up, in which solar panels manufacturer are now playing a key role in the solar power industry. 2. solar cell technology photovoltaic (pv) is a similar term of solar cells, as an electronic devices that convert sunlight directly into electricity. the modern age solar cell was invented at bell telephone laboratories in 1954. a solar cell are grouped together to form a solar module. auxiliary components of solar cell (i.e. balance of system bos) comprise inverter, controls, etc. today, solar cell is one of the fastest growing renewable energy technologies and in the present context, solar cell systems may be considered as one of the most abundant renewable technologies (irena 2012). this is because of the availability of innovative technologies, which makes it feasible to develop efficient modular size with ever increasing power conversion efficiency that they are within the reach of individuals, co-operative societies and small businesses who can install these modules for electricity generation and manage to lock-in prices. over other conventional technologies solar cell technology offers additional benefits like renewable source of energy, readily available across the globe, small and highly modular sizes of solar devices can be installed anywhere at shade-free locations e.g. roofs, gardens etc. solar cell generate electricity by harvesting sunlight and hence require no fossil fuels like nuclear, oil, coal, gas etc., thereby requires relatively low production, operation and maintenance costs. today, wide ranges of technologies are available using different types of materials in solar cell. hence with the huge scope, the future of solar cell is looking bright. solar cell technologies are categories into three generations, viz. first-generation, second generation and third generation based on the material used. 2.1. first generation technologies silicon is a semi-conductive material widely use in solar cell applications. energy band gap of crystalline silicon is about 1.1 ev, which make crystalline silicon more suitable for conversion of sun energy into electric citation: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2),77-93, dx.doi.org/10.14710/ijred.4.2.77-93 p a g e | 80 © ijred – issn: 2252-4940, july 15, 2015, all rights reserved energy. out of all other materials tried till now, silicon have achieved conversion efficiency around 15 to 20 % at commercial scale (kenneth & paul 1982). this technology has a long history of reliable performance with lifetime of more than 25 years. first-generation solar systems use single types of material i.e. crystalline silicon (c-si), either as single crystalline (sc-si) or as multicrystalline (mc-si). single crystalline silicon wafers, which are cut from single crystal silicon ingots to make monocrystalline solar cell. comparatively monocrystalline silicon has capacity to higher efficiency, but it is very difficult to produce from ingot silicon crystal. hence it is very energy intensive and expensive to grow. while multicrystalline wafers are cut directionally solidified blocks or grown in thin sheet so it is easy to build up and they are less expensive. to prepare single crystal silicon from multicrystalline silicon, first and foremost requirement is molten state of silicon and follow by solidification in a manner to arrange the silicon atoms themselves in a perfect lattice as occur naturally. general procedure requires initial contact of single crystal seed of silicon with the molten state and enables to cool down slowly. as the singlecrystal silicon is grown, appropriate substances or dopants are like to introduce to tune the material electric behavior as per requirement. dopants like boron and phosphorus have been used to obtain electron donor and electron acceptor to complete the photoelectric circuit (kenneth & paul 1982). advantages of monocrystalline silicon are that the monocrystalline cells have the highest efficiency, very stable and are simple type of solar cell. drawback of these monocrystalline silicon based solar cells is that they are more expensive and they do not perform efficiently at elevated temperatures (wide bay burnett conservation council inc. wbbcc 2010). crystalline silicon solar cells and modules have dominated photovoltaic technology from the beginning of solar cell history and contribute around 90 % in solar cell market. in 1996, electric output of single crystalline solar cell was 48.7 mw, which share 55 % of global solar cell market (kenneth & paul 1982). multicrystalline solar cell output was 28.4 mw and occupies 32 % global solar cell market. in 2000, single crystalline cell output decreased to 32 % and for multicrystalline it increased to 51%. such changes were observed because of expensive method of fabrication for single crystalline silicon solar cell. in 2004, china enters into the mass production of solar cell technology leads to capture 70 % of total market of solar cell technology. so in 2013, monocrystalline wafer based technology holds 90 % of total market, while multicrystalline about 55 % (fraunhofer institute of solar energy system 2014). 2.2. second generation technologies thin film solar cell as second generation technology, was developed to produce low cost electricity as compared to first generation solar cell. thin film solar cell technology involve the deposition of a thin layer of about 4 µm thickness of photovoltaic material on substrates like steel, glass or plastic. as a result, very less amount of photovoltaic material is required to produce solar cells with good conversion efficiency which makes them feasible for commercial production. thin film solar cells technology utilize non-crystalline material, has lower conversion efficiency ranging from 4 to 12 %. hence its power conversion efficiency is significantly lower in comparison with crystalline solar cells (international energy agency 2013). photovoltaic materials like amorphous silicon (a-si), cadmium telluride (cdte), copper indium selenide (cis) and copper indium gallium diselenide (cigs) are widely used to make thin film solar cell. despite their lower efficiency, thin film solar cells enjoy good stability as well as good shelf-life comparable to c-si module. in addition, they have short energy pay-back times which may be less than 1 year. amorphous silicon (a-si) in thin film solar cell technology, maximum efficiency was achieved with amorphous silicon as a photovoltaic material. defective structure of amorphous silicon leads to high band gap 1.7 ev which in turn increases the absorptivity of sun light to almost 40 times as compared to monocrystalline silicon solar cell (abdin et al. 2013). however, the main problem with amorphous silicon is the degradation of silicon in sun light, so overall efficiency of solar cell diminishes as a function of time. various approaches have been utilized to overcome the above problems. one of the approach is the deposition of amorphous silicon material by plasma vapor deposition technique with as minimum thickness as possible as to strengthen the electric field across the material and its stability (anas i.a.t. 2007). but as the thickness of the film decreases, light absorption capacity decreased and hence cell efficiency also got reduced. so to overcome the lower absorption and conversion efficiency, double, triple or multi junction thin film silicon solar cell approaches were explored. such approaches comprised of the use of thin layer of amorphous silicon and subsequent layer of microcrystalline silicon on to the amorphous silicon. in year 2009, amorphous silicon panels achieved efficiency of 6-9 % (green et al. 2011). afterwards, switzerland based company oerlikon announced test panels that can reach power conversion efficiency of around 11 % using a multifunction technology (greentechmedia 2009). advantage of using amorphous silicon in producing thin film solar cell is that, it is not only very economical but also compatible with other materials. cadmium telluride (cdte) cadmium is a by-product of zinc during mining and tellurium is a by-product of copper processing. cadmium telluride thin film is the most economical thin int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 81 © ijred – issn: 2252-4940, july 15,2015, all rights reserved film technology out of currently available technologies. its band gap is about 1.5 ev, which match with the solar spectrum. in 2011, cdte thin film achieved very high efficiencies up to 16.7% with low production cost than other thin-film technologies (international energy agency 2013). efficiency of cdte solar cells depend on various parameters e.g. deposition technique, substrate and deposition temperature etc. its band gap can be varied by using various approaches like hetero-junction or multi-junction solar cell which lead to increase in efficiency up to 18% (noufi & zweibel 2006). various methods are available to deposit cdte on substrate as mentioned in table 3 (koch et al. 2011). a potential problem of cadmium telluride is the production quantity of telluride and availability, which is depend on extraction, refining and recycling yield of copper industry. toxicity of telluride is another problem for solar cell industry. table 3 thin film deposition technique (koch w et al 2011) sr. no. name of cdte thin-film deposition techniques 1 close space sublimation 2 vapor transport deposition 3 physical vapor deposition 4 sputter deposition 5 electrode deposition 6 metal organic chemical vapor deposition 7 spray deposition 8 screen print deposition copper indium gallium selenide (cigs) as cigs is a new technology so still it is in developing stage and is set to compete with other technologies. solar cells based on cigs offer the highest efficiencies among all thin-film solar cell technologies. power conversion efficiency of solar cell have been achieved up to 20 % at laboratory scale and 7 to 16% was achieve at commercial scale which is close to that of crystalline silicon cells (international energy agency 2013). in this case, band gap can be modified by varying the group iii cations from indium, gallium, and aluminum and the anions between selenium and sulfur. using combinations of different compositions enable to make wider the range of energy band gap. semi conductive materials like cuinse2 (cis), cu(inga)se2 (cigs) and cdte are widely accepted because of their high efficiency, long-term stable performance and potential for low-cost production (noufi & zweibel 2006). thin layer of around 2 mm is sufficient enough to absorb the useful part of the solar spectrum because of its high absorbing power. as compared to other technologies, the manufacturing process is more complex and costly. hence, by replacing indium with low cost materials could help to reduce the cost. currently, researchers are more attracted towards the application of multicrystalline solar cell for space applications due to its excellent photovoltaic effect than silicon. however, thin film manufacturers start adopting modification in existing technology because of low cost crystalline silicon modules are available in the market. four different types of thin film solar cell with power conversion efficiencies are described below in fig 1 (epiaeuropian photovoltaic industry association 2011). fig 1: second generation solar cell efficiency for laboratory scale and commercial scale. 2.3. third generation solar cell third generation solar cell technologies includes concentrating pv (cpv), dye sensitize solar cell and organic solar cells. even with extensive efforts these technologies are still under demonstration and have not yet been widely commercialized. concentrated photovoltaic cell (cpv) concentrated photovoltaic (cpv) cell enable to concentrate a large amount of sunlight onto a small area of solar photovoltaic (pv) cells to generate electricity using optics as a main component which includes lenses or curved mirrors. a concentrator is usually made up of relatively inexpensive materials such as plastic / glass lenses to capture the sunlight falling on a large area and concentrate that energy onto absorbent (irena international renewable energy agency 2013). cpv technology has many advantages like it can operate at elevated temperature, produce more power in minimum time, have slow degradation compared to earlier generation solar technologies and have much higher power conversion efficiency. mirrors or lenses decrease the cost of solar cell significantly than other expensive solar cell (irena international renewable energy agency 2013). cpv systems are classified according to the amount of the concentrated solar energy, measured in ‘suns’ (the square of the magnification): (i) low concentration pv systems with a solar concentration of 2-100 suns, (ii) medium concentration pv systems with a solar concentrations of 100 to 300 suns and (iii) high concentration photovoltaic (hcpv) systems with a solar concentrations of 1000 suns or more (wikipedia, citation: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2),77-93, dx.doi.org/10.14710/ijred.4.2.77-93 p a g e | 82 © ijred – issn: 2252-4940, july 15, 2015, all rights reserved concentrated photovoltaic 2015) . today’s cpv systems have system efficiencies almost 30%, which incorporate highly efficient iii-v silicon solar-cells. the installation costs of cpv systems today are very competitive to flatplate pv systems. at commercial scale silicon based cpv have achieved 20 to 25% power conversion efficiency (robert & vasilis 2012). at laboratory scale, more than 40% of power conversion efficiency was achieved using iii-v semiconductors in multi-junction cpv. their higher efficiency and the smaller pv surface area required may eventually reduce the overall costs. for techno-commercial reasons, the utility market is the most popular market for cpv companies. in 2008, according to a green media survey there were 10, 38, and 46 companies available in the market with their cpv products in residential, commercial and utility segment respectively (sorin, elizabeth & travis . 2008). soitec in europe is another firm with multi junction cells announced for 4-junction devices, at 43.6% conversion efficiency under 319 suns (solar server 2013). in 2012, the fraunhofer institute for solar energy systems fabricated multi-junction photovoltaic cell with 44.7% power conversion efficiency and breaking its own record of 43.6% set just four months prior (maria 2013). dye sensitized solar cell (dssc) dye-sensitized solar cells combining the unique properties of both the organic compound and inorganic compound have attracted the attention of scientists in the recent years. in the conventional solar cell systems, semiconductor material plays the dual role of photon absorber and charge carrier. on the other hand, dssc separate these two functions to two different materials, i.e. the organic molecule usually a dye as a sensitizer absorbs sunlight and the second inorganic molecule then transforms solar energy into electric energy. first, light is illuminated onto a dye which in turn excites electrons in the dye. the excited electrons further excite the electrons in titanium dioxide (tio2) film and ultimately reaching the electrode through diffusion. on the counter electrode, generated electrons reduce i3 － ions to i－ ions and diffuse in the solution, reach the dye, give up electrons, and are oxidized forming i3 － ions (tomohiro & hirohiko 2009). numerous metal complexes and organic dyes have been used as sensitizers in solar cell. photo excitation at a monolayer of organic dye results in the injection of an electron into the conduction band of oxide. then, organic dye restores its original electron configuration by electron donation from the electrolyte, usually an organic system containing redox couples. n3 dye, rul2(ncs)2, and black dye, rul(ncs)3, are the most commonly used as an organic dyes in dscs. main drawback with organic dye based dssc is the toxicity of organic dye as well as the stability of electrolyte at elevated temperature. such problem can be solved by using natural dye because they are biodegradable, cheap and abundant. by using synthetic organic dyes in dssc, highest efficiency of 1112 % have been achieved, while with natural dyes 9.8% was achieved. natural dyes from flower, fruit or vegetable are capable to act as a photo sensitive material in dssc. highest efficiency of 7.6 % have been achieved by using coumarin as a natural dye (souad et al. 2013). however dssc have certain drawbacks e.g. low stability of dye and electrolyte solution at elevated temperature and lesser efficiency compared to other solar cell technologies. need of the hours is to find out the solutions to increase the stability of dyes and electrolytes in these dssc and at the same time increase the efficiency. despite these problems, it is expected that the dssc will become the future of solar cells technologies to provide the cheap energy due with high conversion efficiency. organic solar cell (opv) within the last three decades, tremendous effort have been made to develop organic solar cells common materials used for photovoltaic devices are inorganic. the incorporation of semiconducting polymers into organic solar cells resulted in remarkable improvements within last few years. the sp2hybridization of carbon atom of semiconducting organic materials is responsible for the absorption of visible solar region and to transport the electrons to generate the electricity. basically, four consecutive steps are involved towards working principle of organic solar cell i.e. light absorption, excitation, charge transport and collection (myung 2009). organic solar cells are classified on the bases of number of electron transporting layer available in the device namely single, double and triple layer organic solar cell (daniel & darren 2013). in 1994 a.d. r. n. marks et al. prepared an organic solar cell having poly (p-phenylene vinylene) of 50-320 nm thickness in sandwiched manner between indium titanium oxide (ito) and cathode with 0.1% power conversion efficiency under 0.1mw/cm2 intensity. the reason behind such low efficiency is due to intrinsically low mobility of charges through semiconducting organic. organic solar cells have unique and significant advantage like these devices are not only low costs per module, but also their thinness and extremely small mass could also reduce the costs associated with transportation and installation of modules. so aim of the electronic segment is to set the application of electronics everywhere for the development of next generation. with these, semi conductive organic industry will play a vital role in upcoming future technologies. multidisciplinary approach of organic solar cell with other components like batteries, in fuel cell etc. will enhance their production efficiency. this integration of organic solar cells into many products will be their technological advantage. owing to the lack of scalable high int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 83 © ijred – issn: 2252-4940, july 15,2015, all rights reserved performance donor materials, conversion efficiency of opv is usually lower than other available solar technologies. studies on mass-produced organic photovoltaic (opv) devices lag far behind that on labscale devices. power conversion efficiencies (pce) of 6– 7% have been achieved for opv devices based on different active layers using different organic polymers. recently it has been reported that 8.4% efficient fullerene-free organic solar cells were developed which exploited long-range exciton energy transfer. in one study pce as high as 15% has also been reported and the technology was already patented by the inventors (ning et al. 2013). glimpse of power conversion efficiency (pce) of all three generation technologies for solar cells in terms of laboratory scale and commercial scale as mentioned in table 4 (georg, enrico, & yoann 2012). table 4 the list of technologies with efficiency at laboratory vs. commercial scale with their market share technology material pce at labor atory scale (%) pce at comm ercial scale (%) market share (%) first generation crystalline silicon (multi crystalline and mono crystalline) 28 16-20 85 second generation amorphous silicon 10 8 14 cdte 16 11 cu(inga)se2 20 12 third generation organic solar cell 12 8 1 dye sensitize solar cell (dssc) 10 – 15 11 concentrated solar cell 43.6 31.8 3. current trends towards applications of nanotechnology in solar cell current solar cell technologies have two major drawbacks i.e. low conversion efficiencies and high manufacturing cost for generation of electricity at commercial scale. nanotechnology is an emerging technology that has the potential to provide solution to above problems. as the nanotechnology make use of materials, which are very small in size (less than 100 nm) having very large surface area. such properties can enhance the efficiency manifold and also at the same time reduce the sizes of the devices. currently, most commercial solar cell utilize silicon based semiconductor materials. but in the past few years, scientists have been investigating to increase the efficiency of solar cell using nanotechnology. use of nanomaterial would provide the new ways to capture, store and exchange energy. three basic advantage have been offered by nanotechnology based solar cell. first advantage is the large surface area offer enhanced solar absorption. second, light generated electrons and holes need to travel over a much shorter path and thus recombination losses are greatly reduced. third the tunable energy band gap by varying the size of nanoparticles. this allows for the much needed flexibility in designing of solar cells. currently various nanomaterials have been explored to develop the solar cells. among these materials quantum dots, quantum wells, carbon nanotubes and nanowires are most widely studied (abdin et al. 2013). 3.1 quantum dots (qds) quantum dots are minute crystal composites of size less than 10 nm are having tremendous semi conductive properties. these properties are responsible due to their ability to enhance light absorption and absorb the infrared range. quantum dots has potential to tune the band gap upto optimum value to enhance the maximum efficiency. it can be applied to single junction solar cells with different band gap to form a tandem cells. due to significant properties like significant absorption of infrared radiation, at higher temperatures it increase the photo current and efficient radiation hardness. theoritically it was predicted that the efficiency of qds can be achieve up to ~64% for a well-adjusted intermediate band in a solar cell. a solar cell uses quantum dots as the light absorbing material can be use rather than known bulk materials such as silicon, copper indium gallium selenide (cigs), tio2, zno or cdte (sethi, mukesh, & priti 2011). silicon based qds have been proved to provide better power conversion efficiency as compare to bulk silicon due to the following reasons that (georg, enrico, & yoann 2012): [1] increase in the energy gap with decrease in the size of the qds. due to enhanced energy gap, widening in the distance between the energy states occurs. as a result, the relaxation of the excited carrier is slow down with respect to bulk silicon (hot carrier solar cell). [2] si qds exhibit strong photoluminescence in the visible red region of the solar spectrum. qds can generate multiple excitons in silicon atom with twice or more energy than the band gap, thereby splitting the excited exciton into ‘n’ excitons without losing the excess energy in the form of heat. qds have many advantages like size dependent tunable energy band gap, strong photoluminescence, generation of multiple excitons etc. various approaches have been tried by using qds for solar cell in order to increase power conversion efficiency like (a) qdsscs citation: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2),77-93, dx.doi.org/10.14710/ijred.4.2.77-93 p a g e | 84 © ijred – issn: 2252-4940, july 15, 2015, all rights reserved (quantum dot sensitize solar cell), here qds act as an alternative to the dye molecule in dye sensitized solar cell. (b) crystal qds.to combine the quantum dots inherent tunability with a simple manufacturing process in order to reduce the cost. (c) multi junction solar cell (d) organic solar cell: developments of solar cell using quantum dots (qds) according to us patent us20100012168, us20130042906, us8395042, a quantum dot sensitized solar cell including an anode, a cathode and an electrolyte is provided. the anode includes a semiconductor electrode adsorbed with a plurality of quantum dots. this include a tunable electron conductor that permits greater choices in quantum dots, thereby providing solar cells that can be constructed to utilize a broad light absorption range that covers the ultraviolet, visible and infrared regions leading to an improved conversion efficiency of the solar cell. the size and composition of a quantum dot can determine its band gap and fermi level (mihai, viorel-georgel, cornel, mircea , bogdan-catalin 2010; ming-way 2013; neil, friedrich, timothy, & james, 2013). various approaches have been explored to increase power conversion efficiency like qd’s core-shell structure, qd’s absorb light and emit red shifted light which can absorb by active layer and by depositing qd’s on cnt/ nanowire, 20nm zno mesoporous photo anode to achieve 12 % which was claimed by us patent application us2008/000183, us20120132891 and cn2011/000860 (istvan et al 2006; troy 2008; prashant k. et al 2008; mircea, bogdan-catalin, 2010; leonard & jeeseong, 2012; robert et al 2012; mihai, viorel-georgel, cornel, chunhui 2013). material used for qd’s solar cell are cdse, zns, cdse, ingaas, inga, inas, germanium, cds, pds and inp as per us8658889, us20110146772, us8072039, us20050155641, ep2442326 and us8574685 (hiroaki 2011; kyung-sang, byung-ki. 2011; simon 2011; anna, zhi, linan, marilyn, 2013 & jason, xiaomei 2013; kun-ping 2014). porphyrin-sensitized solar cells with cobalt (ii/iii)–based redox electrolyte and zinc porphyrin dye as sensitizer were succeeded to achieve 12.3% efficiency (aswani et al. 2014). patent us7402832 claimed that core-shell defect free silicon quantum dot exhibits photoluminescence with power conversion efficiency is greater than 10 % (howard & hoon 2008). light scattering silicon and gold nanoparticle shown improved performance of quantumwell solar cells with a 17% power conversion efficiency (derkacs et al. 2008). 94% of the light can absorb by 280 nm thick hydrogenated amorphous silicon (a-si:h) layer in nano dome based solar cell. it is higher than the 65% absorption of flat film devices, due to efficient light management and self-cleaning concept (ekins-daukes et al. 2010). facilitation of significant electric interaction provided by p3ht−cdse nanocomposites having welldefined interface. it also promoting the dispersion of cdse within the p3ht matrix efficiently (xu et al. 2007). power conversion efficiency about 6 % was achieved using co-polymer, poly [n-9-hepta-decanyl-2, 7carbazole-alt-5, 5-(4', 7’-di-2-thienyl-2', 1’, 3’benzothiadiazole) (pcdtbt) in bulk heterojunction composites with the fullerene derivative [6, 6]-phenyl c70-butyric acid methyl ester (pc70bm). conversion of sunlight to electricity to supply for future generation is considered to be a major task. to address a future need of electricity, the cost of solar energy will have to be significantly lower and the efficiency will have to be increase significantly. so with these requirements, research on nanostructure si qds for application in solar cell has enormously increased. however, a deep scientific investigation of using these structures is required to outperform the c-si solar cells to meet future energy needs. 3.2 quantum well (qw) barnham and co-workers at imperial college proposed and demonstrated the first quantum well solar cells (qwscs) a decade ago. it was based on the hypothesis that quantum wells could extend the spectral response and increase the photocurrent of solar cells without degrading their voltage characteristics. hence initial proposal for these devices was motivated by the goal of enhanced energy efficiency. early experiments on algaas-based cells indeed revealed that introduction of gaas quantum wells enhanced both the photocurrent and power conversion efficiency under broadband illumination. these results motivate the researcher to explore quantum well phenomenon to enhance power conversion efficiency (keith et al. 1997). in the past few years researcher are more attracted towards quantum well solar cells (qwsc). quantum well solar cells (qwsc) is a thin layer of material posing a lower band gap than the surrounding barrier or host material. quantum well must be sufficient thin to bring about quantum effect through the confinement of electron and holes, so its dimension is in range of few nanometers. hence its novel structure has potential to achieve high efficiency if the absorption spectra of absorbing material is wide by photo-generated electrons and holes escape from the wells and increase the output current. a qwsc consists of a p–i–n-junction constructed from high band-gap semiconductor with thin layers of lower band-gap semiconductor to form potential wells, as shown in fig 2. by the recombination in the bulk material, the open circuit voltage can be still controlled which is known as barriers. while the sections of lower band-gap material absorb low-energy photons that the barrier material which is known as well. for light with energy greater than the band-gap (e) the quantum-well (qw) cell behave like a conventional solar cell. however, light with energy below and above int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 85 © ijred – issn: 2252-4940, july 15,2015, all rights reserved the energy band-gap (eg), can be absorbed in the quantum wells. in 2009, a single junction gaasp/ingaas quantum well solar cell attained a peak efficiency of 28.3% under solar concentration (ekins-daukes et al. 2013). major benefit of incorporating a quantum well stack into a multi-junction solar cell is to increase the photocurrent up to 40 % with ingap/ mqw/ge. hence additional current flow through the top and middle cells to increase the efficiency. presently, qwsc efficiencies of approximately 20–25% are achievable. while in future, it can be expected that 35% efficiency or beyond is achievable using strain-balanced tandem qwsc in order to extend the spectral range (rault & zahed 2003). for single junction qwsc optical band gap can tune the thickness and composition to get highest efficiency efficiencies. fig 2: quantum well based solar cell. development of solar cell using quantum well (qw) prof keith barnham of imperial college london invented the qw solar cell in 1989, which operated at high current and was made of alternating layers of ingaas and gaasp wile p and n layer are made from gaas. in 2009, prof keith barnham achieved the highest efficiency of a nano structure solar cell at 30.6 % obtained from a tandem junction qw solar cell at a concentration of 54 suns (mazzer et al 2006). us patent us5496415 and us4688068 describe a solar cell formed from a semiconductor having a relative wide band-gap characterized by a multi-quantum well system incorporated in the depletion region of the cell. these quantum wells comprise regions of semiconductor with a smaller band gap. material used for smaller band gap is ingaa, gaas, gap, and gan (gordon 1986; keith 1996; alexandre 2000; liang, ilyas & masud 2013; roger). these kinds of quantum well were able to achieve 12 % power conversion efficiency. it has been also discovered that the power conversion efficiency of ingan/gan based quantum well sola cell are inversely proportional to temperature (ramesh et al 2013). ingan/gan multiple quantum well (mqw) solar cells with enhanced power conversion efficiency using colloidal cds quantum dots (qds) and back-side distributed bragg reflectors (dbrs) up to 20.7% (yu-lin et al. 2013). patent us20120167973, ep2595193 and us8101856 describe that solar cell were designed by using quantum well which is made up of polycrystalline silicon, germanium or gap-si. by incorporating hydrogenated nano crystalline silicon (nc-si:h) and a-si:h (nc-si:h/asi:h) multiple quantum wells (mqw) in the intrinsic region, absorption coefficient increases in three steps starting from 1.0 × 104 cm−1 to 3.0 × 104 cm-1 for photon energy of 1.2 ev to 1.4 ev, respectively. therefore number of incorporation is directly related with the efficiency of solar cell (ankur, manvendra, pratibha 2014). the mitigation of the thermal scattering as well as the control of the light absorption range of quantum energy level is considered as a driving force for next generation solar cell. multiple quantum wells (mqw) utilize absorbing material with multiple band gap to achieve higher conversion efficiency. this mqw thickness of the well layer (5-10 nm) is important to form quantized energy state in the well. mqw based solar cells demonstrate good temperature stability because of multi band gap structures. due to this feature of mqws they are also suitable thermo-photovoltaic applications, where heat generated from combustion process can be converted into electricity. these cells can also be used for spaces applications (seung 2007). 3.3 carbon nanotube (cnt) carbon nano tube (cnt) is made up of graphite sheet, which are rolled like a straw. it may be single rolled sheet single wall carbon nano tube (swcnt) or multiple rolled sheet i.e. multiple wall carbon nano tube (mwcnt). carbon nanotubes is considered as ideal material for solar cell due to wide range of direct band gaps from infrared to ultraviolet and high carrier mobility with reduced carrier transport scattering, which make themselves ideal photovoltaic material. secondary advantage of carbon nanotubes is earth abundant, possess high optical absorption, and superior thermal and photo stability. the extraordinary properties of cnts such as light weight, excellent mechanical strength, three dimensional flexibility and outstanding electro catalytic property, can improve the performance of solar cells. cnt nanostructured surface can also enhance the performance of both charge separation and electron transportation, so they can be applied as either a working electrode or counter electrode, or even as the conductive substrate as shown in fig 3. electron mobility’s on the order of 10,000 cm2/vs have been measured for both metallic and semiconducting tubes. these extraordinary electronic citation: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2),77-93, dx.doi.org/10.14710/ijred.4.2.77-93 p a g e | 86 © ijred – issn: 2252-4940, july 15, 2015, all rights reserved properties are due to low defect density and the one dimensional geometry (yan et al. 2013). utilizing the concept of the photo induced charge transfer between organic conjugated polymers as a donor and nanotubes as acceptor, cnts and different polymers have been used extensively with an aim to increase the efficiency by three times. hence cnt are considering as a promising candidate for solar cell by increasing conversion efficiency, increasing stability and decreasing cost (maciej et al. 2011). fig 3: dye sensitize solar cell containing cnt as charge transfer layer development of solar cell using carbon nanotube (cnt) their low-cost processing is expected to become a method for commercially realistic energy conversion efficiency and the most probable technology to significantly contribute to renewable electricity generation in the coming years. due to their electro catalytic properties carbon nanotube as an electrode increase power energy conversion of solar cell up to 7-9 % (chen et al. 2012). recently, a dye-sensitized solar cell with a tube like structure achieves 4.16 % conversion efficiency, which is different from the sandwich structure of the traditional dye-sensitized solar cell (yongping et al. 2012). so far, cnts have been utilized in solar cell as a substitute of tio2 either in counter electrode or photo anode with simple mixing in solid state electrolyte and with an engineered interfacial layer. recently various approaches utilizing cnts have been explored to increase the conversion efficiency of solar cell. tio2 is considered to be a versatile candidate for the photo electrochemical (pech) application due to its ability to improve the light absorption in visible region (cheng et al. 2013). a titanium dioxide nanoparticle is dispersed on single wall carbon nanotube (swcnt) films to improve photo induced charge separation and transport of carriers to the collecting electrode surface. as far as photoelectron-chemical is concern, charge equilibrium between tio2 and single wall carbon nanotube is a prime important factor to improve the efficiency of solar cell (yongping et al. 2012). however, in case of multi wall carbon nano tube (mwcnt), a controlled layer of multi-wall carbon nanotubes (mwcnt) is usually grown directly on top of fluorine-doped tin oxide (fto) or other glass electrodes as a surface modifier for improving the performance of organic solar cells by 2% (capassoa et al. 2014). charge transport properties enhance the charge collection efficiency of dye sensitized solar cells (dssc) by incorporating oriented tio2 nanotube arrays consisting of closely packed nanotubes having several micrometers in length with typical wall thicknesses, inter tube spacing of 8−10 nm and pore size diameters of about 30 nm (kai et al. 2007). recent patent applications (in2010/000023 and jp2013118127) describe that using tio2 carbon nano tube (mwcnt) nanocomposites comprised dssc prepared by using hydrothermal route leads to the increased efficiency of the dssc (brian et al. 2005; subas, vivek, sarfraj, mujavar, 2010; kiyoshige 2013). recently, several polymers having conducting properties have gained popularity in designing cheap, lightweight, flexible and affordable solar cells. several investigations in this regard have been done by various researchers. for example, single-wall carbon nanotube (swcnt)–polymer solar cells were constructed with region regular poly (3-octylthiophene)-(p3ot) with purity >95% w/w, laser-generated swnts (michael c/o stuttgart technology center dürr, gabriele c/o stuttgart technology center nelles, akio stuttgart technology center yasuda 2005). another invention mentioned in patent ep1507298 describes that solar cell comprising carbon nanotubes (cnt) as an electron conductor and organic substance like polymethacrylate’s for hole’s to get photovoltaic effect (michael c/o stuttgart technology center dürr, gabriele c/o stuttgart technology center nelles, akio stuttgart technology center yasuda 2005). in another such research, by using concentric electro spinning of poly (methyl methacrylate) as a core precursor and polyacrylonitrile as a shell precursor to prepare a hollow active carbon nanofiber have been used. such nanofiber with 190 and 270 nm for core and shell diameter showed excellent mesoporous structure, and 1-d conducting pathway in employing as catalysts of counter electrodes (ces) for dye-sensitized solar cells (dssc). in other developments involving the development of high efficiency solar cells various semiconducting materials have been deployed. patent us20130092236 and us2011/054301 describe such devices where absorbing layer are composed of gallium arsenide (gaas), copper indium selenide (cuinse) etc. to achieve high efficiency with low cost (ranga, christopher, claire, bhaskar, omkaram, srikant, gaurav, sanjayan, kaushal; robert 2012). it was also demonstrated in one of the study that acid infiltration of nanotube networks significantly boosts the cell efficiency to enhance charge separation and transport by reducing the internal resistance to improve the fill factor int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 87 © ijred – issn: 2252-4940, july 15,2015, all rights reserved and by forming photo-electrochemical units. here the efficiency of about 13.8% was achieved by combining carbon nanotubes and silicon (yi et al. 2011). other materials such as silver, palladium, germanium, platinum etc. were also used as conductive adhesives. conductive adhesive enables interconnection of new cells and interconnector concepts, such as heterojunction solar cells with their heat sensitive surface (zemena et al 2013). palladium nanocrystallite decorated tio2 nanotubes (pd/tnts) improve the light absorption in visible region which enhance photo catalytic activity of solar cells (cheng et al. 2013). in another study a composite film composed of platinum nanoparticles and carbon fibers was coated on fto glass as a counter electrode in dye-sensitized solar cell (dssc) shown 7.77% efficiency (cai, chen & peng 2010). patent no 20100089447 describes the use of conductive layers prepared of silver nanoparticles or magnesium nano layer (1-20 nm) thereof mixture (takahito, chihaya, 2007; bulent, burak, richard, 2010). the purpose of carbon nanotube (cnt) based conductive adhesive was to increase the reliability and also to replace these costly heavy metals. defect rich edge planes of bamboo like structure carbon nanotube enable the electron transfer kinetics at the interface of counter electrode and electrolyte and result in improvement of fill factor due to low charge transfer resistance. so in combination with a dyesensitized tio2 photo anode and an organic liquid electrolyte, a multi-wall carbon nanotube (cnt) counter-electrode dssc shows 7.7% energy conversion efficiency. uniform distribution of multiwall carbon nanotubes within the heterojunction organic solar cell are shown to increase in cell fill factor of 50–60% (anthony, ross, ravi 2006). it has been reported that cnt incorporated with tio2 structure could play an important role in transporting the photo generated electrons towards the conductive supports, with the addition of swcnts, the incident power conversion efficiency 7.3 % to 16 %. cnts can be used for the development of flat solar cells and flexible solar cells, where the latter are light weight and have the advantage for signal transmission and pv self-power generation. in the flexible solar cells, rigid substrate made of transparent conducting oxides (tco) has been replaced and it has more power conversion efficiency. the flexible solar cells can be generated through carbon fibers, wire shaped cnt-cnt solar cell and carbon nanoyarn. in addition, the carbon nanostructures due to their efficient power conversion efficiency and 3d bendable structure can be used for various engineering applications such as fiber textile, composites etc. a further research is required, where new designs functional materials with smaller band gaps may enhance power conversion efficiency. wire shaped dsscs with above mentioned characteristics and stability under different environments may be used for various engineering applications. 3.4 nanowire (nw) nanowires are defined as metallic or semiconducting particles having a cross sectional diameters <1 nm, and submicron size length. ideal material for nanowires would have a band gap of around 1.55 ev (qifeng, et al. 2012). the nanowire geometry provides potential advantages like optimum tunable band gap, better light trapping, increase defect tolerance and reduce reflection to increase the efficiency and reduce the cost. 1-d nanowire arrays as photo-electrodes have been widely investigated to provide a rapid collection of carriers and affording large junction areas with less reflectance loss. nanowires are composed of n-type and p-type material, arrange in such a way that one material rolled over another material, so as an when photons irradiate on it then hole and charge move in equal and opposite direction as shown in fig 4. nanowires of si, ge, gan and inas have high electron mobility in the axial direction. therefore nanowires show comparable to or even over the best reported for solar cell efficiency (benjamin & peidong 2009). one of the most successful methods employed for nanowire synthesis is based on chemical vapor deposition (cvd). nanowires grown by cvd employ a so called vapor liquid solid (vls) growth mechanism. vertically aligned oxide nanowire comprising solar cell allows the light to be absorbed in the vertical direction and prevent exciton diffusion to improve the efficiency. in addition, the use of inexpensive, environmentally benign, and stable oxide materials suggests that a device constructed in this manner could have an excellent working lifetime at a fraction of the cost of existing solar cells. figure 4. charge transfer in nanowire based solar cell the x and y axis directions are much smaller than z axis direction of nanowire, which enable to impart nanowires with various applications in electronic and optoelectronic devices. it was found in dye sensitized solar cells (dssc) that, the nanowires provide electron transport 100 times faster than conventional cell by providing direct pathways for electron conduction. semiconducting nanowire (nws) present a fascinating potential opportunity for developing high efficiency, low-cost solar cells by decoupling light absorption and carrier separation. ordered arrays of vertical nanowires citation: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2),77-93, dx.doi.org/10.14710/ijred.4.2.77-93 p a g e | 88 © ijred – issn: 2252-4940, july 15, 2015, all rights reserved with radial junctions has advantages like optical, electrical, strain relaxation, charge separation mechanisms and product cost than thin film solar cell with planner junction. in addition to playing a role in providing direct pathways for electron transport, compared to planar film, the nanowires contribute to enhancing the optical absorption and hence increase the solar cells efficiency (sharma & juhi 2013). development of solar cell using nanowire (nw) according to patent us7893348, solar cell comprising silicon (si) nanowires as active elements have been designed (bastiaan & loucas 2011). patent us201102847230 describe that patterning planar photo-absorbing materials into arrays of nanowires is demonstrated as a method for increasing the total photon absorption in a given thickness of absorbing material (chang & yue, 2010; robert & william 2011). by varying the wire length alters the wire filling ratio required to achieve the same amount of light absorption. optimal designs for antireflective silicon nanowire based solar cells achieve 11% power conversion efficiency (jin et al 2013). radial growth of hydrogenated amorphous silicon over p-doped crystalline silicon nanowires using plasma enhanced chemical vapor deposition technique on glass substrates achieves 5.6 % efficiency (erik & peidong 2010). as with cnts, solar cells have been fabricated using various other materials like silver, gold, titanium, zno, tio2, ingan, gan, gaas, and gainp etc. or in combination with silicon nanowires (fei, siguang, kang, 2007; lars, martin, federico, 2009; juneui, jihyun, sangwoo 2010; yijie, anjia, james, shu, paul 2012; siegfried, 2012; tae, joo, jae, jae, won 2012; zetian 2013). by using tio2 and germanium arsenide (gaas) nanowire either with polymer as composite or as an electrode it can achieve even higher power conversion efficiency of 9.2 % (jiun, shu, ching 2012; hong, shu, ching 2012; sin k. et al 2014). patent applications ib2013/050943 and us20120202314 describe that solar cell comprised of a layer of p/n-doped semiconductor nanowires with embedded polymer layer (samson, hao, felix, 2012; silke, boer, gerhardus, 2013). based on a heterojunction between the vertically aligned gaas/ ag nanowires and polymers like poly (3, 4-ethylenedioxythiophene): poly (styrenesulfonate) and/or incorporating poly (3hexylthiophene) electron blocking layer have been designed to achieved power conversion efficiency of 9.2% (jiun, shu, ching 2012; hong, shu, ching 2012). in another breakthrough, highly flexible and cost-efficient brush painting of flexible organic solar cells (foscs) used same material. in addition, it provides an opportunities to fabricate complex single crystalline semi-conductors devices directly in low cost subtract and electrodes such as al foil, stainless-steel and conductive glass. nano wire based solar cell can be fabricated using a wide variety of materials ranging from si, zn, cdte, cuo2, tio2, in and many other polymer nw combination. power conversion efficiency from these nanowires can be achieved close to 10%. these nanowire provide high efficiency due to their unique geometry for example (a) periodic array of nanowire with radial junctions give the advantage such as reduced reflation, maximum light trapping, radial charge separation and single crystalline synthesis on known epitaxial subtracts (b) axial junction loss the radial charge separation benefit but keep the others. (c) substrate junction lack the radial charge separation benefit and cannot be removed the substrate for testing as single nanowire solar cell. these nanowire have been synthesized by using two most common techniques i.e. chemical vapor deposition and pattern chemical etching. the surface property of the materials, dopant profiling and incorporation of catalyst during growth of nws has been found to affect the electrical properties, charge mobility and diameters. the above factors promised to lower the cost of solar cell by reducing the amount of material needed. it is expected that the cost of multifunction solar cell fabrication using nanowire can be as low as to that of traditional thin film solar cell cost without compromising the power conversion efficiency. however there are many challenges which needs to be addressed before the benefits can be realized commercially, these challenges includes surface characteristics of the material, mechanical and chemical stability, doping control, nanowire array uniformity and scalability. 3.5 graphene graphene is the two dimensional (2d) building block for sp2 carbon allotropes. a flat monolayer of carbon atoms tightly packed into a two dimensional (2d) honeycomb lattice is known as graphene. it is generally wrapped up into zero dimension (0d) fullerenes, rolled into one dimension (1d) nanotubes or stacked into 3d graphite. hence it can be used as a lubricant due to its loosely bonded layered planar structure with the planes of carbon atoms. history of graphene history of graphene is incomplete without describing graphene oxide (go). graphene are called as reduced graphene oxide. which was extensively studied in back 1840s (daniel, rodney, christopher 2010). timeline of selected events occurred in the history of graphene as mentioned below table 5. table 5 historical development of graphene for solar cell (daniel, rodney, christopher 2010) year events 1987 the name “graphene” was first mentioned by s. mouras and co-workers to describe the graphite layers that had various compounds inserted between them forming the so-called graphite intercalation compounds or gic’s. 1999 ruoff and coworkers micromechanically exfoliate int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 89 © ijred – issn: 2252-4940, july 15,2015, all rights reserved graphene into thin lamella comprised of multiple layer of graphene. 2003 in 2003, andre and kostya succeeded in producing the first isolated graphene flakes and this work was published in 2004. 2005 the first major result on the electronic properties was the anomalous quantum hall effect in graphene reported around the same time by andre and kostya and by philip kim’s group at columbia university. 2010 andre and kostya were awarded the 2010 nobel prize in physics for this work and are continuing to unveil new and exciting properties in graphene and other related two dimensional crystal materials. 2014 jonathan coleman and colleagues from trinity college dublin were able to create graphene by mixing the proper amount of graphite powder, water and dishwashing liquid in a high-powered blender. potential for solar cell graphene has very high electrical conductivity due to it contains both holes and electrons as charge carriers. the graphene with each absorb photon it generate an electron hole pair. it was said that graphene’s electrons has similar mobility as photon due to their lack of mass. graphene electrons require to travel sub-micrometer distances without scattering i.e. ballistic transport phenomenon. graphene is used as a light absorbent can able to increase efficiency of solar cell by enhancing absorption capacity. graphene is the strongest material ever discovered with a tensile strength of 300 x 108 pascal’s which is 4 x 108 times stronger than a36 structural steel. the reason behind its strength is 0.142 nm-long carbon bonds. hence graphene is a strong candidate for solar cell (yanwu et al. 2010). development of solar cell using graphene stanford university scientists have developed the first solar cell made entirely of carbon, which is a cheaper alternative to the expensive, silicon based solar cell (standford 2012). bao and her colleagues use graphene sheets as dual material (active layer and electrode). it primarily absorbs near-infrared wavelengths of light so it power conversion efficiency is less than 1% (standford 2012). the university of florida achieves 8.6 % power conversion efficiency by using graphene doped with trifluoromethanesulfonyl-amide (tfsa) coated silicon wafer material. with celebrative research of universitat jaume i in castello and oxford university, they claimed 15.6% power conversion efficiency by combining titanium oxide and graphene as a charge collector and perovskite as a sunlight absorber. its production costs are very low, enable to make it possible for the technology to be used on flexible plastics (gizma 2015). if plasmonic nanostructures of graphene were optimized, it should be possible for falling photon on graphene is converted into current and increase the efficiency of solar cell. that is what industry want to produce said by alexander grigorenko of manchester (physic world 2011). soitec germanyas recently developed, a word class four junction solar cell grown on different iii-v compound materials using cpv technology gives an outstanding 43.6 % pce. this technology was developed in collaboration with the fraunhofer institute of solar energy system and the helmholt zentrum fur materialien und energie in berline. this allows optimal band gap combinations to capture the maximum sunlight’s. recently, multi-junction solar cells with concentrator photovoltaic (cpv) systems have demonstrated solar conversion efficiency over 40. it is cost effective because of broad geographic areas for solar absorption. for metamorphic 3-junction gainp/gainas/ge solar cell, its power conversion efficiency was about 41.6% at 484 suns, the highest efficiency yet demonstrated for this type of cells. professor ravi silva from the university of surrey outlined his definition for 4th generation (4g) solar cells, capable to improve the power conversion efficiency of existing 3rd generation solar cell. 4th generation (4g) solar cells comprise single layer of conducting polymer films (organic) i.e. cheap and flexibility with novel nanostructures (inorganic) i.e. lifetime stability (anthony, ross, ravi 2006). 3.6 nanofabrication: roll to roll process three dimensional (3-d) nanomaterial has potential to boost up the performance of solar cell owing to the power conversion efficiency. however the cost effective production with mass production having unchanged optical and physical properties are still remain challenge for the present technology. though, polymeric solar cell up to some extent are able to overcome the current challenges like high manufacturing cost, lack of ability for mass production or restricted application etc. so in modern scenario, researchers are adopting multidisciplinary approach by combining the nanofabrication technique and polymeric solar cell concept to beneficiate the solar industry. nanofabrication is the technique to manipulate the component or material in controllable manner at nanometer scale in 1, 2 or 3 dimension. such technique gives value addition to the manufacturing process due to change in electronic property of the material at nanometer scale. in last few years, electronic industry has attracted much towards the nanofabrication technique to produce electronic circuits or boards with extraordinary electric performance. now a day’s same concept was applied for solar cell using roll to roll process. roll to roll is the process to deposit, assemble or coat the material on substrate for mass production of industrial product. it has been already used by coating industry to beneficiate of time saving mass production. while in solar industry, roll to roll process is unable to fulfill the prior requirement of power conversion efficiency. by coupling nanofabrication and roll to roll citation: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2),77-93, dx.doi.org/10.14710/ijred.4.2.77-93 p a g e | 90 © ijred – issn: 2252-4940, july 15, 2015, all rights reserved method it is possible to open a wide scope of opportunity for manufacturing paradigm and industrial requirements. nanofabricated roll to roll technique provides a flexible, transparent and light weight substrate for the manufacturing of solar cell. at development stage, materials like silicon, group i-v and polymeric nanoparticles have been investigated as an absorbing component, while plastic or metal strip have been used as a substrate of solar cell. as roll to roll technology for solar cell industry is less mature enough technology than other technologies and hence the power conversion efficiency of solar device achieved very low. the list of different methods of roll to roll process are comprising nanofabrication with their power conversion efficiency in solar cell as shown in table 11. with extensive research and development in last few years power conversion efficiency of flexible solar cell using cigs is 14.1 %, cdte is 11.4 %, a-silicon is 12 % and tio2 is 8 % on plastic substrate at laboratory scale. while at industrial scale for cigs is 17.5 %, cdte is 8 % and asilicon is 8 % (razykov et al 2011). however inherent stability and less production cost indicate a promising potential for these technology. various types of nanofabrication process are available like nanoimprinting, nano-lithography, self-assembling, molecular engineering etc. hence, large scale flexible solar cell will grove into a billion dollars industry over the next few years and will change our view of electronics. the photoelectric market in 2010 with about 2 billion usd of which 20% is flexible. it was projected that by 2020, it rises to almost 60 billion usd, of which 75 % is of flexible (idtechex 2009-2029 market report 2009). table 6 list of roll to roll methods with their pce (%) (markus h. 2013) sr. no. technology material power conversion efficiency (%) 1 slot die coating pedot:pss, p3ht:pcbm 1.74 2 doctor blading pod2tdtbt:pcbm 6.74 3 spray coating p1:pcbm 5.1 4 screen printing ag 20.3 5 flexo printing ag 18.1 6 gravure printing pedot:pss, p3ht:pcbm 2.8 7 inkjet printing ag 4.91 4. future aspects researchers succeed to demonstrate the conversion of solar energy to electric energy after the extensive research of 60 year. today also we are facing crucial limitations like manufacturing cost and poor power conversion efficiency. hence, it is not sufficient to replace the consumption demand of fossil fuel with solar energy. innovative solar cell technologies that utilize nanostructured materials offer great technological potential due to their attractive properties. the late noble physicist richard smalley saw a potential in nanomaterial to bring revolution in energy sector by modifying the fundamentals component of solar cell using value adding property of nanotechnology to make better solar cell. to overcome the problem of reflection, anti-reflecting coating has been used, which increase the cost of solar cell. recently, nonporous material was coated on solar cell surface which is known as black silicon. it increased the 20% power conversion efficiency from conventional solar cell. however, efficiency of solar cell is not only the advantage of nanotechnology. another value addition of nanotechnology is to produce ultra-thin and flexible solar cell, which cross limit of rooftop solar panels. new solar cells are based on films formed by rollto-roll process technology of nanomaterial enabling high surface area, transparency, excellent stability and good electrical conductivity and are ideal for solar applications. it also opens the opportunity to produce cheaper and environmental friendly solar cells. availability of electricity is the major challenge for developing countries, due to lack of fossil fuel and high global price of crude oil. so, nanotechnology has potential to decrease the cost of solar cell and allow to achieve good power conversion efficiency. it also enables to meet the increasing demand of electricity and also accessible for those people. today researchers started exploring nanotechnology in different way to develop the solar cell. therefore, the use of nanotechnology appears to provide a ladder to develop newer solar cell with much high pce at cheaper cost and bring revolution in the solar cell industry in future. 5. acknowledgement the authors are grateful to the confederation of indian industry and govt. of gujarat for their financial support (ref. no.: ic/inc/coe/sanction/667493) in establishing the coe-nt at ahmedabad, gujarat, india. nr is also thankful to the institute of pharmacy, nirma university, ahmedabad for their support by providing library facility. references abdin z, alim a, saidur r, islam r, rashmi w, mekhilef s & wadi a. (2013) solar energy harvesting with the application of nanotechnology, renewable and sustainable energy reviews, 26, 837-852.. affordable solar (2015), http://www.affordable-solar.com/ learning -center/solar-basics/ solar-history; (accessed: 5 february 2015). alexandre f. (2000), multi-quantum well tandem solar cell, (patent application us6147296a). anas i.a.t. (2007), amorphous silicon based solar cells, phd thesis, university of stuttgart. ankur g., manvendra v. & pratibha s. (2014) single junction a-si:h solar cell with a-si:h/nc-si:h/a-si:h quantum wells, thin solid films, 550, 643–648. anna l., zhi z., linan z. & marilyn w. (2013) method for enhancing the conversion efficiency of cdse-quantum dot sensitized solar cells, (patent application ep2442326a3). anthony m., ross h. & ravi s. (2006) interpenetrating multiwall carbon nanotube electrodes for organic solar cells, applied physics letter, 89, 117-133. int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 91 © ijred – issn: 2252-4940, july 15,2015, all rights reserved anthony m., ross h. & ravi s. (2006) interpenetrating multiwall carbon nanotube electrodes for organic solar cells, applied physics letter, 89-95 anuradha t., lovish j. & pranjal b. (2013) solar energy finding new ways, international journal of research in advent technology, 4, 1-6. aswani y, hsuan-wei l, hoi t, chenyi y, aravind c, md.khaja n, eric d, chen-yu y, shaik z & michael g. (2014) porphyrin-sensitized solar cells with cobalt (ii/iii)–based redox electrolyte exceed 12 percent efficiency, science, 334, 605, 629-634.. bastiaan k, loucas t, (2011), nanowires in thin-film silicon solar cells, (patent application us7893348b2). benjamin y. & peidong y. (2009) nanowire-based all-oxide solar cells, journal of american chemical society. 131, 3756–3761. brian l, ryne r, stephanie c & sheila b. (2005) single-wall carbon nanotube–polymer solar cells, progress in photovoltaics: research and applications, 13, 2, 165–172. bulent b., burak m. & richard s. (2010) conductive grids for solar cells, (patent application us20100089447a1). cai j., chen t. & peng s. (2010) all carbon nanotube fiber electrode based dye sensitize photovoiltaic wire, journal of material chemistry, 22, 30, 14856-14860. capasso a, salamandra l, chou a, di-carlo a & motta n. (2014) multiwall carbon nanotube coating of fluorine-doped tin oxide as an electrode surface modifier for polymer solar cells. solar energy materials and solar cells. 122, 297–202. chang c. & yue ma. (2010), increasing solar cell efficiency with silver nanowires, (patent application us20100129949a1). chen t, huisheng p, shutao w, zhibin y, quanyou f, xuemei s, li l & zhong-sheng w. (2012) light weigh, ultrastrong and semiconductive carbon nanotube fibres for highly efficient solar cells, angewandte chemie interational edition; 50, 8, 1815-1819. cheng x, liu h, chen q, li j & wang p. (2013) preparation and characterization of palladium nano-crystallite decorated tio2 nano-tubes photoelectrode and its enhanced photocatalytic efficiency for degradation of diclofenac, journal of hazard mater, 8, 254-255. cheng x, liu h, chen q, li j & wang p. (2013) preparation and characterization of palladium nano-crystallite decorated tio₂ nano-tubes photoelectrode and its enhanced photocatalytic efficiency for degradation of diclofenac, journal of hazard material, 254, 255, 141-148. chunhui l, lei y, junyan x, yih-chyng w.martin s, yanhong l, dongmei l, qingbo m & bo-brummerstedt i. (2013) zno nanoparticle based highly efficient cds/cdse quantum dotsensitized solar cells, chemical physics., 15, 8710-8715. daniel b. & darren l. (2013) green chemistry for organic solar cells, energy environ, science, 6, 2053–2066. daniel d., rodney r. & christopher b. (2010), from conception to realization: an historial account of graphene and some perspectives for its future, angewandte chemie international edition, 49, 9336 – 9345. derkacs d, chen w, matheu p, lim s, yu p & yu e. (2008) nanoparticle-induced light scattering for improved performance of quantum-well solar cells, applied physics letters, 93; 91-107. effie j. cosmos (2013) http://cosmos.ucdavis.edu/archives/2013 /cluster8/jia_effie.pdf; (accessed: 20 october 2015) ekins-daukes n, jia z, ching-mei h, zongfu y, shanhui f & yi c. (2010) nanodome solar cells with efficient light management and self-cleaning, nano letter, 10; 1979– 1984. ekins-daukes n, lee k, hirst l, chan a, fuhrer m, adams j, browne b, barnham k, stavrinou p, connolly j, roberts j, stevens b, airey r & kennedy k. (2013) controlling radiative loss in quantum well solar cells, journal of physics d: applied physics, 46, 26, 264007. epiaeuropian photovoltaic industry association (2011). http://www.epia.org/fileadmin/user_upload/publications/co mpeting_full_report.pdf; (accessed: 5 february 2015). erik g. & peidong y. (2010) light trapping in silicon nanowire solar cells, nano letter, 10, 3, 1082–1087. experience (2015), https://www.experience.com/ alumnus/article? channel_id=energy_utilities&source_page=additional_articles &article_id=article_1130427780670; (accessed: 15 february 2015) fei l., siguang m. & kang w., (2007), carbon nanotube/nanowire thermo-photovoltaic cell, (patent application us20070235076a1). fraunhofer institute for solar energy system ise (2014), photovoltaic report,http://www.ise.fraunhofer.de/de/downloads/pdffiles /aktuelles/ photovoltaics-report-in-englischer-sprache.pdf; (accessed: 5 february 2015) georg p., enrico s. & yoann j. (2012) silicon quantum dots for photovoltaics: a review, quantum dots a variety of new applications, (ed ameenah al-a) pp 10-22 in tech. gizma (2015) http://www.gizmag.com/graphene-solar-cell-recordefficiency/30466/; (accessed: 5th february 2015) green a, keith e, yoshihiro h & wilhelm w. (2011) solar efficiency tables (version 35), progress in photovoltaics: research and applications, 18, 144-150. green tech media (2009) http://www.greentechmedia.com/ articles /read/amorphous-silicon-solar-losing-the-shakeout; (accessed: 5 february 2015). hiroaki t. (2011) method for manufacturing quantum dot-sensitized solar cell electrode, quantum dot-sensitized solar cell electrode and quantum dot-sensitized solar cell, (patent application us20110146772), hong s., shu s. & ching l. (2012) silicon nanowire/organic hybrid solar cell with efficiency of 8.40%, solar energy materials and solar cells, 98, 267–72. howard w & hoon l (2008), quantum dots of group iv semiconductor materials, (patent application us7402832b2). idtechex 2009-2029 market report (2009), http://www. idtechex.com/research/reports/printed_and_thin_film _transistors _and_memory_2009_2029_000221.asp, (accessed: on 5 february 2015). international energy agency (iea), energy technology systems analysis programme (etsap) (2013), http://www.irena.org/documentdownloads/ publications/irenaetsap%20tech%20brief%20e11%20solar%20pv.pdf; (accessed: 5 february 2015) irena international renewable energy agency (2013) http://www.irena.org/documentdownloads/publications/ irenaetsap%20tech%20brief%20e10%20concentrating% 20solar%20power.pdf; (accessed: 5 february 2015). irena international renewable energy agency, renewable energy technologies: cost analysis series (2012), 45. istvan r, vaidyanathan s, masaru k & prashant k. (2006) quantum dot solar cells. harvesting light energy with cdse nanocrystals molecularly linked to mesoscopic tio2 films, journal of american chemical society, 128, 7, 2385– 2393. jason l. & xiaomei j. (2013) electric field tuning of pbs quantum dots for high efficiency solar cell application, (patent application us8574685b1). jin-young j, han-don u, sang-won j, kwang-tae p, jin ho b & jung-ho l. (2013) optimal design for antireflective si nanowire solar cells, solar energy materials and solar cells, 112, 84–90. jiun c., shu s. & ching l. (2012), gaas nanowire/poly(3,4thylene dioxythiophene):poly(styrenesulfonate) hybrid solar cells with incorporating electron blocking poly(3-hexylthiophene) layer, solar energy materials and solar cells,105, 40–45 juneui j., jihyun m. & sangwoo l. (2010) effects of zno nanowire synthesis parameters on the photovoltaic performance of dyesensitized solar cells, thin solid films, 520, 17, 5779–5789 kai z, nathan n, alexander m & arthur f. (2007) enhanced chargecollection efficiencies and light scattering in dye-sensitized solar cells using oriented tio2 nanotubes arrays, nano letter, 7, 1, 69–74. keith b, (1996) concentrator solar cell having a multi-quantum well system in the depletion region of the cell, (patent application us5496415a). citation: raval, n. and gupta, a.k. (2015) historic developments, current technologies and potential of nanotechnology to develop next generation solar cells with improved efficiency, 4(2),77-93, dx.doi.org/10.14710/ijred.4.2.77-93 p a g e | 92 © ijred – issn: 2252-4940, july 15, 2015, all rights reserved keith b, ian b, jenny b, james c, paul g, benjamin k, jenny n, ernest t, alexander z. (1997) quantum well solar cells, applied surface science, 113, 14, 722-733.. kenneth z & paul h. (1982) basic photovoltaic principles and methods, solar energy research institute, technology and engineering, 249. kiyoshige (2013) kojima, photoelectrode using carbon nano-tube, (patent application jp2013118127). koch w (2011) handbook of photovoltaic science and engineering, (eds antonio l & steven h) pp 241-65, wiley, england. kun-ping h. (2014) quantum dot thin film solar cell, (patent application us8658889). kyung-sang c. & byung-ki k. (2011) energy conversion film and quantum dot film comprising quantum dot compound, energy conversion layer including the quantum dot film, and solar cell including the energy conversion layer, (patent number us8072039). lars s., martin m. &federico c., (2009), nanowire-based solar cell structure, (patent application wo2008156421a3). leonard p & jeeseong h, (2012) precision quantum dot clusters, (patent number us20120132891) liang w., ilyas m. & masud b. (2013) quantum efficiency of multiple quantum wells, (patent application wo2014008412a3). sibinski, m, jakubowska m, znajdek, k, słoma m & guzowski b. (2011) carbon nanotube transparent conductive layers for solar cells applications, optica applicata, 12, 2, 375-381.. maria j. (2013) record breaking solar cell approaches 45% efficiency; http://forceofthesun.com/record-breaking-solarcell-approaches-45-efficiency. (accessed: 5 february 2015). markus h. (2013) large-scale roll-to-roll fabrication of organic solar cells for energy production, phd thesis, technical university of denmark. mazzera m, barnham k, ballard i, bessiere a, ioannides a, johnson d, lynch m, tibbits t, roberts j, hill g & calderc c. (2006) progress in quantum well solar cells, thin solid films, 76– 83. michael c/o stuttgart technology center dürr, gabriele c/o stuttgart technology center nelles, akio stuttgart technology center yasuda (2005) carbon nanotubes based solar cells, (patent application ep1507298a1). mihai m, viorel-georgel d, cornel c, mircea b & bogdan-catalin s (2010) quantum dot solar cell, (patent number us20100012168). mihai m, viorel-georgel d, cornel c, mircea b & bogdan-catalin s, (2010), quantum dot solar cell, (patent number us20100012168). ming-way l (2013) quantum-dot sensitized solar cell (patent number us20130042906). myung k. (2009) understanding organic photovoltaic cells: electrode, nanostructure, reliability, and performance, phd thesis, university of michigan. neil d, friedrich p, timothy h, & james m, (2013) quantum dot solar cell with quantum dot bandgap gradients (patent number us8395042) ning l, derya b, karen f, florian m, tayebeh a, mathieu t, miguel c, martin d, antonio f, frederik k & christoph b. (2013) towards 15% energy conversion efficiency: a systematic study of the solution-processed organic tandem solar cells based on commercially available materials, energy environmental science, 6, 3407-3413. noufi r. & zweibel k., (2006) high-efficiency cdte and cigs thin-film solar cells: highlights and challenges, ieee 4th world conference on photovoltaic energy conversion (wcpec-4) waikoloa, hawaii physic world (2011) http://physicsworld.com/cws/article/news /2011/sep/05/graphene-could-make-perfect-solar-cells; (accessed: 5th february 2015) prashant k. (2008) quantum dot solar cells. semiconductor nanocrystals as light harvesters, journal of physical chemistry c, 112, 48, 18737–18753. qifeng z, supan y, junting x, daniel m & guozhong c. (2012) oxide nanowires for solar cell applications, nanoscale, 4; 1436–1445 ramesh l, baolai l, zachary b, tugba n, seth h, andrew n& diana h. (2013) gasb/ingaas quantum dot–well hybrid structure active regions in solar cells, solar energy materials and solar cells, 1, 14, 165–171. ranga a., christopher b., claire c., bhaskar k., omkaram n., srikant r., gaurav s., sanjayan s, kaushal s. & robert v., (2012) high efficiency solar cell device with gallium arsenide absorber layer, (patent application wo2012044978a3). rault, f & zahed, a, monash university’s (2003). http://solar.org.au/papers/03papers/rault.pdf; (accessed: 21 october 2015). razykov t, ferekides c, morel d, stefanakos e,ullal h & upadhyaya h. (2011) solar photovoltaic electricity: current status and future prospects, solar energy, 85, 1580–1608. robert c. & vasilis f. (2012) third generation photovoltaics, (ed vasilis f) pp 167-182, in tech. robert p, serge p , jessica k & michael g. (2012) quantum dot sensitization of organic− inorganic hybrid solar cells, journal of physical chemistry b, 106, 31, 7578–7580. robert s. & william w., (2011), disordered nanowire solar cell, (patent application us20110083728a1). roger c. & gordon o. (1986), quantum well multijunction photovoltaic cell, (patent application us4688068a). samson j., hao x. & felix k., (2012), solar cells based on polymer nanowires, (patent application us20120202314a1). sethi v., mukesh p. & priti s. (2011) use of nanotechnology in solar pv cell, international journal of chemical engineering and applications, 2, 77-80. seung m. (2007), recent progress in inorganic solar cells using quantum structures, recent patents on nanotechnology, 1, 1, 67-73. sharma r. & juhi n. (2013), absorption of light in silicon nanowire solar cells: designing of solar cells, international journal of computational engineering research, 3; 7; 25-28. siegfried k, (2012), nanowire multijunction solar cell, (patent application us8242353b2). silke d., boer k. & gerhardus d, (2013), flexible nanowire based solar cell, (patent application wo2013118048a1). simon f. (2011) solar cell with epitaxially grown quantum dot material, (patent application us7863516b2). sin-bi k, yong-jin n, seok-in n & han-ki k. (2014) brush-painted flexible organic solar cells using highly transparent and flexible ag nanowire network electrodes, solar energy materials and solar cells,122, 152– 157.solar server (2013); http://www.solarserver.com/solar-magazine/ solarnews/archive2013/2013/kw21/soitec-produces-436efficient-4-junction-solar-pv-cell.html; (accessed: 20 february 2015). sorin g., elizabeth w. & travis b. (2008) concentrating solar power technology, cost, and markets, prometheus institute for sustainable development and greentech media. souad b, iraj a & iis s. (2013) natural photosensitizers for dye sensitized solar cells, international journal of renewable energy research, 3(1), 139-143.. standford (2012) http://news.stanford.edu/news/2012/october/ carbon-solar-cell-103112.html; (accessed: 5th february 2015) subas m., vivek d., sarfraj h. & mujavar o., (2010). high efficient dyesensitized solar cells using tio2multiwalled carbon nano tube (mwcnt) nanocomposite, (patent application wo2010079516a1). tae k., joo y., jae j., jae y. & won p., (2012), solar cell using p-i-n nanowire, (patent application us20120097232a1) takahito o, chihaya a, (2007) organic solar cell, (patent application us20090199903a1). tomohiro n. & hirohiko m. (2009) development of dye-sensitized solar cells, ulvac technical journal, 70, 1-5. troy h. (2008) quantum dot photovoltaic device, (patent number us2008000183). u.s. department of energyenergy efficient and renewable energy, the history of solar, https://www1.eere.energy.gov/solar/pdfs/ solar_timeline.pdf; (accessed: 5 february 2015). int. journal of renewable energy development 4 (2) 2015:77-93 p a g e | 93 © ijred – issn: 2252-4940, july 15,2015, all rights reserved wide bay burnett conservation council inc (wbbcc) (2010) https://wbbcc.files.wordpress.com/2010/08/solartechnology-explained.pdf; (accessed: 5 february 2015) wikipedia (2015), http://en.wikipedia.org/ wiki/timeline_of_solar _cells; (accessed: 5 february 2015). jun x , jun w , mike m , prasun m, malika j,jacob p & zhiqun l. (2007) organic-inorganic nanocomposites via directly grafting conjugated polymers onto quantum dots, journal of american chemical society, 129, 42, 12828-12833. yan j, uddina m, dickensa t, okolia o. (2013) carbon nanotubes (cnts) enrich the solar cells, solar energy, 96, 239–252. yanwu z, shanthi m, weiwei c, xuesong l, ji-won s, jeffrey p, & rodney r. (2010), graphene and graphene oxide: synthesis, properties, and applications, advance material, 20, 1–19. yi j, anyuan c, xi b, zhen l, luhui z, ning g, jinquan w, kunlin w, hongwei z, dehai w, & ajayan p. (2011), achieving high efficiency silicon-carbon nanotube heterojunction solar cells by acid doping, nano letter, 11, 5, 1901–1905.. yijie h., anjia g., james h., shu h. & paul m., (2012), nano-wire solar cell or detector, (patent application us20120006390a1). yongping f, zhibing l, hongwei w, saocong h, xin c, dang w & dechun z. (2012) dye-sensitized solar cell tube, solar energy materials and solar cells, 112; 212–219. yu-lin t, chien-chung l, hau-vei h, chun-kai c, hsin-chu c, kuo-ju c, wei-chi l, jin-kong s, fang-i l, peichen y & hao-chung k. (2013) improving efficiency of ingan/gan multiple quantum well solar cells using cds quantum dots and distributed bragg reflectors, solar energy materials and solar cells, 217, 531– 536.. zemen y, schulz s, trommler h, buschhorn s, bauhofer w & karl s. (2013) comparison of new conductive adhesives based on silver and carbon nanotubes for solar cells interconnection, solar energy materials and solar cells, 109; 155–159.. zetian m. & md k. (2013). high efficiency broadband semiconductor nanowire devices, (patent application us20130240348a1). international journal of renewable energy development int. j. renew. energy dev. 2023, 12(4), 720-740 | 720 https://doi.org/10.14710/ijred.2023.54612 issn: 2252-4940/© 2023.the author(s). published by cbiore contents list available at ijred website international journal of renewable energy development journal homepage: https://ijred.undip.ac.id a comprehensive review on the use of biodiesel for diesel engines van giao nguyen1, minh tuan pham2,*, nguyen viet linh le3, huu cuong le4, thanh hai truong5, dao nam cao5,* 1institute of engineering, hutech university, ho chi minh city, viet nam. 2school of mechanical engineering, hanoi university of science and technology, hanoi, viet nam. 3faculty of automotive engineering, school of technology, van lang university, ho chi minh city, vietnam. 4institute of maritime, ho chi minh city university of transport, ho chi minh city, viet nam. 5patet research group, ho chi minh city university of transport, ho chi minh city, viet nam. abstract. fossil fuels are the main source of energy for transportation operations around the world. however, fossil fuels cause extremely negative impacts on the environment, as well as uneven distribution across countries, increasing energy insecurity. biodiesel is one of the potential and feasible options in recent years to solve energy problems. biodiesel is a renewable, low-carbon fuel source that is increasingly being used as a replacement for traditional fossil fuels, particularly in diesel engines. biodiesel has several potential benefits such as reducing greenhouse gas emissions, improving air quality, and energy independence. however, there are also several challenges associated with the use of biodiesel including the compatibility of biodiesel with existing engine technologies and infrastructure as well as the cost of production, which can vary depending on factors such as location, climate, and competing uses for the feedstocks. meanwhile, studies aimed at comprehensively assessing the impact of biodiesel on engine power, performance, and emissions are lacking. this becomes a major barrier to the dissemination of this potential energy source. therefore, this study will provide a comprehensive view of the physicochemical properties of biodiesel that affect the performance and emission properties of the engine, as well as discuss the difficulties and opportunities of this potential fuel source. keywords: biodiesel; physicochemical properties; engine performance; emissions characteristics; blended fuel; nano-additives. @ the author(s). published by cbiore. this is an open access article under the cc by-sa license (http://creativecommons.org/licenses/by-sa/4.0/). received: 25th april 2023; revised:15th june 2023; accepted: 20th june 2023; available online: 28th june 2023 1. introduction fossil fuel reserves include areas where the existence of fossil fuels is "proven, probable, or possible" to approach and extract (speight, 2011). there are obvious differences between reserve and resource. while all reserves are resources, the reverse does not happen. resources become reserves when two conditions are met: (i) they must be discovered and recorded, (ii) the economic feasibility of being able to access and extract mineral resources. therefore, though there are many calculations about how much time humanity has left before fossil fuel reserves are exhausted, most of the calculations are inaccurate when it is common to consider only "proven, probable, or possible" reserves while the number of resources that exist is still unexplored (perera and nadeau, 2022; plantinga and scholtens, 2021). with technology in the field of resource extraction increasingly developed, more and more resources are discovered as well as the ability to exploit reserves that were previously unexploitable is also improving. that is good news when in the short term, the problem of running out of fuel is not a threat (shafiee and topal, 2009). however, the nature of these resources is still non-renewable, not to mention the number of actual resources being able to become reserves is unknown. on the other hand, despite the increasingly developed technologies that help machines operate * corresponding author email: nam.cao@ut.edu.vn (d. n. cao); tuan.phamminh@hust.edu.vn (m.t.pham) more efficiently and smoothly, directly helping to improve energy efficiency, the world's resource consumption is increasing every year and there is no sign of a decline (martins et al., 2018; peters et al., 2017). therefore, their depletion is inevitable. research to find and shift to renewable resources is essential and helps humanity best prepare before any serious energy crisis can occur. energy has always been a burning issue throughout the development of mankind. along with the population explosion, the demand for energy of each individual also increases, making energy is never enough even though newly invented technology has helped people increasingly exploit and create more energy from different sources (pham et al., 2023). transport is an important industry in every economy (hoang et al., 2022a). they not only serve the travel needs of people but also play a lifeline role in the supply chain, especially in the current period of globalization (nguyen and bui, 2021; rudzki et al., 2022). most of the energy supplied to the transportation industry comes from fossil fuel sources such as gasoline or diesel (fernández et al., 2020; serbin et al., 2021). even so, the misuse of fossil fuels can create negative impacts on society (stelmasiak et al., 2017; yang et al., 2019). energy consumption is an important economic driver fueling growth and prosperity review article https://doi.org/10.14710/ijred.2023.54612 https://doi.org/10.14710/ijred.2023.54612 mailto:nam.cao@ut.edu.vn mailto:tuan.phamminh@hust.edu.vn https://orcid.org/0000-0002-4641-8563 http://crossmark.crossref.org/dialog/?doi=10.14710/ijred.2023.54612&domain=pdf v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 721 issn: 2252-4940/© 2023. the author(s). published by cbiore (venugopal et al., 2023). recent decades have observed exceptional growth in global energy demand with forecasts predicting a steady increase in the coming years as countries around the world continue on their paths of economic development. since it was first invented, internal combustion engines have played an important role in propelling societies forward in both the literal and metaphorical senses of the world (balasubramanian et al., 2022; sharma et al., 2023). today, the internal combustion engine model is one of the most common heat engines installed among vehicles, production machinery, and manufacturing equipment. due to the better energy conversion efficiency and cheaper fuel costs, the diesel engine is often favored over the gasoline counterpart as the main power source in electric generators, machinery, and equipment that are used in various sectors including construction, agriculture, and heavy industry, as well as among road vehicles and maritime transport fleets (hoang and pham, 2019; lamas et al., 2015). one of the biggest drawbacks of diesel engines is the significant amounts of air pollutants emitted during the combustion process. with the common application of diesel engines worldwide, this negative impact on the environment is only further exacerbated by the annual increase in the number of passenger vehicles. high-density urban areas are often subjected to hazardous air quality conditions due to the heavy city traffic which has become a fairly common occurrence (hoang et al., 2021a). as a proportion of the urban population in countries around the world continues to grow, environmental and health impacts caused by poor air quality present a major challenge to today’s government leaders (bakır et al., 2022. on the other hand, the urgent need for alternative sources of energy that could potentially replace traditional fossil fuels has become increasingly apparent. according to a study by lamb et al. (lamb et al., 2021), the total ghg emissions from transport operations worldwide were about 8.5 gtco2eq in 2018, accounting for about 14% of total emissions. among them, emissions from road transport activities account for a staggering 73% of the industry's emissions. figure 1 shows the contribution of ghg emissions in the transport industry in 2018 (lamb et al., 2021). with the characteristics of being able to be used flexibly for short distances and only having to bear a small load when compared to aircraft and ships, along with a densely located and easily accessible distribution and repair facility, the application of alternative fuel sources in road traffic is not only easier but also much safer than changing fuel sources for aviation or ships. in addition, since it accounts for the majority of the industry's emissions, being able to successfully use environmentally friendly fuel sources in road transport will rapidly reduce emissions in the transportation industry. therefore, research on alternative fuels for cars and motorcycles is of the utmost interest and development. another serious problem is the uneven distribution of fossil fuel deposits around the world, this leads to energy insecurity in these resource-deficient countries. according to the latest statistics in 2022, only the ten countries possessing the largest oil reserves in the world account for more than 85% of the total oil reserves of the whole world (“oil reserves by country 2022,” n.d.). for the above reasons, researchers have been making great efforts to find a fuel that can be widely used to replace fossil fuels. among them, biodiesel is considered to be one of the most suitable and potential alternatives (nguyen and vu, 2019; prabhu et al., 2023). four generations of biodiesel have been researched and developed depending on the raw materials used for production (singh et al., 2019). while first-generation biodiesel uses edible resources like rapeseed oil, palm oil, and soybean oil for production, second and third-generation biodiesel uses nonedible resources (goh et al., 2022). jatropha curcas, rubber seed, or neem oil are commonly used to produce second-generation biodiesel (singh et al., 2020), while animal fat and waste cooking oil are the main sources of third-generation biodiesel (n et al., 2023; hadiyanto et al., 2018). in addition, algae are usually used for synthesizing fourth-generation biodiesel (jeyakumar et al., 2022; maroušek et al., 2023b). the difference between the second, third, and fourth-generation biodiesel is that the third and fourth-generation use more economically optimal raw materials as well as do not depend on the seasonal characteristics of the crop, and do not affect the food chain and use the land for cultivation (sakthivel et al., 2018). usually, food crops such as rapeseed, soybean, sunflower, safflower, palm, coconut, and animal fats, etc. are processed and it is converted into biodiesel. these biodiesels are obtained and are named 1st generation biofuels as this was the earliest alternative idea in the production of biodiesel. still, various types of research have been carried out and many found that biodiesel production can also be done by the processing of non-food crops and novel starch like jatropha, pongamia, mahua, pine, nerium, and calophyllum inophyllum, etc. these biodiesels obtained are named as 2nd generation biofuels as this was taken as the next initiative in the production of biofuels. many improvements were found in 2nd generation biofuels when comparing them with the 1st generation biofuels in terms of performance, combustion, and emission characteristics of the diesel engine. but the availability of these 1st and 2nd generation biofuels is limited. for large-scale production of biodiesel from algae in countries like india, vietnam, etc. is available in plenty among various water reserves. hence, the yielding of biofuel from the micro-algae feedstock seems to be better than the other feedstocks in the asean countries. therefore, biodiesel is renewable energy with extremely diverse production materials and can be found in every country (silviana et al., 2022; zullaikah et al., 2021; hadiyanto et al., 2016). figure 2 shows the main sources of biodiesel production (“global biodiesel production is increasing renewable carbon news,” n.d.). it can be seen that biodiesel production sources are extremely diverse, and it should be noted that these are statistics on commercialized biodiesel sources when the percentage of edible sources still accounts for a fairly high proportion. meanwhile, in recent years, advances in the field of fuels have helped researchers to propose more efficient and less socially harmful sources of biodiesel production (kolakoti et al., 2022; fig. 1. contribution of ghg emissions in the transport industry in 2018 (lamb et al., 2021) v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 722 issn: 2252-4940/© 2023. the author(s). published by cbiore widayat et al., 2023). however, it is necessary to have a comprehensive review of the use of biodiesel for diesel engines. this study will focus on analyzing the properties of biodiesel and their influence on engine performance and emissions. in addition, the study also discusses challenges and provides the latest methods to solve outstanding problems. 2. biodiesel properties with the main purpose to replace mineral diesel in the internal combustion engine itself, biodiesel's physical and chemical properties have many similarities with diesel. these similar properties make the blending of biodiesel easy, as well as the use of biodiesel does not require the engine to be seriously modified. however, differences are still present and need to be analyzed to understand their effects before being widely used. table 1 shows the important properties of some common biodiesel compared to diesel (ayhan et al., 2020; dinesha et al., 2019; hoang et al., 2021b; jafari et al., 2019; nagaraja et al., 2012; nayak et al., 2021). this section will focus on analyzing each fuel's characteristics and comparing it to diesel fuel. 2.1. kinematic viscosity viscosity is one of the most important fuel parameters that every fuel research must consider first. viscosity represents the ability of the fuel to flow and this parameter will directly affect how the fuel injection system works. from the data in table 1, it can be seen that the viscosity of biodiesel is higher than diesel oil. especially, kinematic viscosity will greatly affect performance and even cause engine damage if operating at low temperatures. this is the main reason for the obstacles when using biodiesel in engines that operate in low-temperature environments and can only be used as a secondary fuel to blend with diesel if there are no suitable engine modifications (hoang, 2021a). the reason for the high kinematic viscosity of biodiesel is because of their high chemical structure and molecular weight. the suitable viscosity range according to astm d445 is 1.9–6.0 mm2/s and according to en iso 3104 is 3.5–5.0 mm2/s (balat, 2011; balat and balat, 2010). the viscosity of biodiesel is still in the area of satisfying the above standards, but when applied to each specific engine, it should be considered very carefully. 2.2. density density is the weight per unit volume, density of the fuel is also a highly significant factor since it has been associated with other characteristics of fuel like the cetane number and heating value (tesfa et al., 2010). besides, based on density, engineers can measure and design the fuel tank and the amount of fuel in the system (alptekin and canakci, 2008). there is not much of a density difference between diesel and biodiesel although the density of biodiesel tends to be slightly higher than that of diesel. fig. 2. main sources of biodiesel production (sarin, 2012) table 1 physicochemical properties of diesel fuel and some popular biodiesel fuel properties diesel fuel palm oil methyl ester (pome) corn oil methyl ester (come) coconut biodiesel waste cooking oil biodiesel honge oil methyl ester (home) rice bran oil-based biodiesel (rbo) fish oil biodiesel (fbd) kinematic viscosity (cst) 3.18 4.5 4.3 4.82 4.36 4.7 4.68 4.91 density (kg/m3) 839.0 870.0 870-880 860.0 890.0 890.0 892 877 lower heating value (mj/kg) 44.8 37 39.6 37.2 38.8 38.9 42.2 41 cetane number 40-55 56.5 >55 58.6 53.4 54 63.8 59 flashpoint (oc) 68 178 >100 150 to 170 175.4 210 183 156 pour point (oc) -7 -5 to -10 -10 to -15 5 to 12 0 to -5 -3 to -12 -11 5 cloud point (oc) -10 to 6 5 to 10 -5 to -10 9 to 14 -3 0 to -9 -10 9 oxygen content (%) 0 11.26 10.96 11.54 11 11.3 10-12 9.3 v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 723 issn: 2252-4940/© 2023. the author(s). published by cbiore besides density, relative density, which is the density of the component compared to the density of water, is also an important parameter of the fuel to compute flow and viscosity characteristics, convert mass to volume, and assess the homogeneity of biodiesel tanks. 2.3. calorific value or lower heating value calorific value (cv) or lower heating value (lhv) is a measure of the amount of heat generated from the full combustion of a hydrocarbon not accounting for the heat contained in combustion products if not returned to precombustion temperature. cv is the real energy content and is also an important parameter for estimating the design parameters of the combustion process (giakoumis and sarakatsanis, 2018; kumar et al., 2013). the cv of diesel is approximately 12% higher on a weight basis when compared to biodiesel, which also means diesel fuel has higher energy content than biodiesel. however, biodiesel has a slightly higher density than diesel, so the cv of diesel is only about 8% higher than biodiesel on a volume basis (ozcanli et al., 2013). 2.4. cetane number the cetane number (cn) shows the ability of the fuel to autoignite quickly after being injected. the higher the cetane number, the shorter the time between the start of fuel injection into the combustion chamber and the ignition process. the higher the cetane number, the better the ignition quality of the fuel (karmakar et al., 2010; lapuerta et al., 2008). this is one of the extremely important indicators in choosing the right fuel for the engine. in the table, it can be seen that the cetane number of diesel is slightly lower than that of biodiesel. the advantages of having a higher cetane number in biodiesel are manifold. these include shorter ignition delay, lower nox emissions, and a decreased incidence of knocking during the combustion process (godiganur et al., 2010; kumar and kumar, 2010; reyes and sepúlveda, 2006). 2.5. flash point, cloud point, and pour point the temperature at which a fuel will catch fire when exposed to a flame or a spark is known as its flash point. flashpoint varies inversely with the fuel’s volatility. the flash point is a significant property that pertains to the combustibility characteristics of liquids (mejía et al., 2013). the flash point values of methyl esters derived from vegetable oil are considerably lower compared to the flash point values of the original vegetable oils. moreover, as the quantity of residual alcohol increases, the flash point of these methyl esters decreases (černoch et al., 2010). it can be seen in the flash point of biodiesel that is much higher than that of diesel, which will make it safer to handle, store, and transport fuel. the cloud point is the temperature at which wax crystals first become visible when the fuel is cooled, whereas the pour point of a liquid is the temperature below which the liquid loses its flow characteristics. it is defined as the minimum temperature at which the oil can pour down from a beaker (lopes et al., 2008). these indicators are very important especially when the engine has to work at low temperatures. unsuitable cloud point and pour point fuels can cause the fuel to solidify and clog the vehicle's fuel system and filters, directly affect engine performance, and cause long-term engine damage. 2.6. oxygen content one of the biggest differences between biodiesel and diesel fuel is the oxygen content (coşofreţ et al., 2016). while the oxygen content of diesel is very low or even absent, biodiesel is an oxygen-rich fuel. oxygen content in biodiesel is about 10 to 12% weight depending on the type of biodiesel. the oxygen content in biodiesel will help fuel burn cleaner and significantly reduce the number of unburned hydrocarbons (uhc). 2.7. stability of oxidation the stability of oxidation in fuels refers to their resistance to oxidative reactions, which can lead to the formation of harmful by products, degradation of the fuel quality, and potential engine performance issues. the stability of oxidation is particularly important for hydrocarbon-based fuels like gasoline and diesel. the oxidative stability of biodiesel fuel is influenced by the number of bis-allylic sites present in unsaturated biodiesel compounds. factors such as the biodiesel's age, the composition of fatty acid methyl esters, and storage conditions contribute to biodiesel's oxidation stability (rajamohan et al., 2022). due to its molecular structure, biodiesel fuels are more susceptible to oxidative degradation compared to fossil diesel fuels. astm d6751 and en-14214 are two standards for evaluating the oxidation stability of fuels. while the minimum requirement for oxidation stability at 110°c with astm d6751 standard is 3 hours, the en-14214 standard is stricter when it comes to the requirement that the fuel maintains 6 hours under similar conditions. nevertheless, biodiesel in its pure form, derived from various feedstocks, typically fails to meet this requirement (sakthivel et al., 2018). therefore, some additives are often added to biodiesel to enhance its stability of oxidation. 3. effects of biodiesel on combustion, performance, and emissions characteristics of the engine with the above properties, the direct use of biodiesel on the engine is feasible, but its effects on the engine need to be comprehensively evaluated. therefore, a lot of research has been done recently to study different aspects of engines using biodiesel as well as improve both its performance and emission characteristics. this section will present a comprehensive perspective on the above problem. limits and opportunities of potential biodiesel fuel sources will also be mentioned and discussed. 3.1. effects of biodiesel on combustion characteristics to better understand the effect of biodiesel on the engine, it is necessary to analyze the combustion process. the most important parameters in combustion analysis are ignition delay (id), heat release rate (hrr), and pressure rise rate (prr) which are commonly calculated and measured. table 2 shows the newest finding on engine combustion characteristics as well as performance. all the data is collected when the engine is operating at 100% load if the engine load is not specified. the trends of increasing or decreasing the above parameters when compared with diesel-only engines can be predicted in theory based on the physicochemical properties of biodiesel (sharma et al., 2022). the parameters of the engine are most closely related to each other, and changing one parameter will affect all the other v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 724 issn: 2252-4940/© 2023. the author(s). published by cbiore parameters (tuan hoang et al., 2021). although it is possible to predict the trend of the parameters, it is still necessary to perform specific simulations or experiments to accurately determine the effect of the fuel on the engine because any small t a b le 2 e ff e c ts o f b io d ie se l o n e n g in e p e rf o rm a n c e a n d c o m b u st io n b io d ie s e l r a ti o e n g in e p e rf o rm a n c e e n g in e c h a ra c te ri s ti c r e f b s f c b t e e g t ig n it io n d e la y ( °c a ) h e a t re le a s e r a te (j / °c a ) p re s s u re r is e r a te (m p a / °c a ) j a tr o p h a o il m e th y l e st e r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 1 0 % , 1 2 % , 1 8 % , 2 4 % a n d 3 3 % fo r b 2 0 , b 4 0 , b 6 0 , b 8 0 a n d b 1 0 0 a t 7 5 % l o a d ↓ 4 % , 6 % , 8 % , 1 0 % a n d 1 1 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 a n d b 1 0 0 a t 7 5 % lo a d ↓ 3 % , 4 % , 6 % , 7 % a n d 8 .5 % fo r b 2 0 , b 4 0 , b 6 0 , b 8 0 a n d b 1 0 0 a t 7 5 % l o a d (g a d e t a l. , 2 0 2 1 ) j a tr o p h a o il b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↑ 4 % , 7 % , 9 % , 1 0 % , 1 4 % fo r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 5 % , 7 % , 8 % , 9 % , 1 1 % fo r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↑ 1 % , 3 % , 4 % , 5 .5 % a n d 6 % fo r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 2 .5 % , 3 % , 3 % , 7 % a n d 8 % fo r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 4 % f o r b 1 0 0 ↓ 6 % , 1 0 % , 1 0 % , 1 3 % a n d 1 4 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 (r a o e t a l. , 2 0 0 7 ) j a tr o p h a o il b 2 0 ↓ 5 % ↑ 5 % ↓ 9 % ↓ 1 3 % (d e e p a n ra j e t a l. , 2 0 1 7 ) k a ra n ja o il b 2 0 ↓ 5 .2 % ↑ 1 1 % ↓ 1 2 .5 % ↓ 1 5 % (d e e p a n ra j e t a l. , 2 0 1 7 ) w a st e c o o k in g o il b 1 0 0 ↑ 1 3 % ↑ 5 % ↓ 7 .8 % ↓ 6 .5 % (a n e t a l. , 2 0 1 3 ) w a st e c o o k in g o il b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↑ 1 6 % , 1 9 % , 3 2 % , 3 5 % a n d 4 5 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 1 6 % , 2 2 % , 3 2 % , 3 2 % a n d 4 7 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 1 0 % , 1 9 % , 2 8 % , 3 7 % a n d 4 6 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 3 % , 5 % , 7 % , 9 % a n d 1 1 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 3 % , 5 % , 7 % , 8 % a n d 9 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 (m o h a m e d e t a l. , 2 0 2 0 ) c a n o la o il a n d s a ff lo w e r o il b 1 0 0 ↑ 1 0 % ↑ 1 .5 % ↑ i n si g n if ic a n t ↑ i n si g n if ic a n t (a lp te k in , 2 0 1 7 ) s o y b e a n b 1 0 , b 2 0 , b 5 0 , b 1 0 0 ↑ 2 % , 4 % , 7 % a n d 9 % fo r b 1 0 , b 2 0 , b 5 0 , b 1 0 0 ↓ w it h th e in c re a se o f b io d ie se l in t h e m ix tu re ↓ i n si g n if ic a n t ↑ i n si g n if ic a n t (ö ze n e r e t a l. , 2 0 1 4 ) s o y b e a n b 3 0 , b 5 0 , b 8 0 , b 1 0 0 ↑ 7 % , 8 % , 1 0 % a n d 1 2 % fo r b 3 0 , b 5 0 , b 8 0 , b 1 0 0 ↓ i n si g n if ic a n t ↓ i n si g n if ic a n t a t 9 0 % lo a d ↓ 2 .5 % b 1 0 0 a t 9 0 % lo a d (q i e t a l. , 2 0 1 0 ) r a p e se e d o il b 5 , b 2 0 , b 7 0 , b 1 0 0 ↑ 2 .5 % , 3 % , 5 .5 % a n d 7 .5 % f o r b 5 , b 2 0 , b 7 0 , b 1 0 0 ↑ i n si g n if ic a n t ↑ 3 % f o r b 1 0 0 ↓ 9 .7 % , 1 7 % , 3 0 % a n d 4 8 % fo r b 5 , b 2 0 , b 7 0 , b 1 0 0 ↓ 5 .4 % , 8 ,4 % , 1 2 % a n d 1 6 % fo r b 5 , b 2 0 , b 7 0 , b 1 0 0 ↓ 3 % , 3 .8 % , 4 ,2 % a n d 5 % f o r b 5 , b 2 0 , b 7 0 , b 1 0 0 (b u y u k k a y a , 2 0 1 0 ) f is h o il b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↑ 4 % f o r b 2 0 ↓ 1 .5 % , 9 .8 % , 1 1 % a n d 1 3 .5 % fo r b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↑ 2 % , 3 % , 4 % a n d 5 % fo r b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ w it h th e in c re a se o f b io d ie se l in t h e m ix tu re ↓ 2 .5 % , 1 2 % , 1 4 % , 1 8 % a n d 2 2 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 4 .5 % , 1 2 % , 1 2 % , 2 3 % a n d 3 0 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 (g n a n a se k a ra n e t a l. , 2 0 1 6 ) f is h o il b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↑ 3 .7 % f o r b 2 0 , ↓ 1 .8 % , 6 .4 % , 1 1 .3 % a n d 1 2 .4 % fo r b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↑ 0 .4 % , 2 .4 % , 2 .1 % , 1 .1 % a n d 3 .3 % fo r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 2 .2 % , 2 .5 % , 4 % , 1 5 % , 2 4 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 2 .8 % , 1 2 .2 % , 2 0 % , 2 2 .5 % a n d 2 4 ,5 % fo r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 ↓ 8 .6 % , 1 9 % , 2 4 .8 % , 4 7 % a n d 7 1 % f o r b 2 0 , b 4 0 , b 6 0 , b 8 0 , b 1 0 0 (s a k th iv e l e t a l. , 2 0 1 4 ) r ic e b ra n o il b 2 0 ↓ 3 .4 5 % ↑ 7 .8 1 2 5 % ↓ 6 .7 % ↑ 1 7 .6 % ↑ 1 3 .7 % (d h a m o d a ra n e t a l. , 2 0 1 7 ) n e e m o il b 2 0 ↓ 1 0 .3 4 % ↑ 5 .6 5 % ↓ 1 4 .2 % ↑ 1 2 .5 % ↑ 8 .5 % (d h a m o d a ra n e t a l. , 2 0 1 7 ) v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 725 issn: 2252-4940/© 2023. the author(s). published by cbiore change can have a bigger impact in the long run, not to mention that most internal combustion engines degrade in efficiency and fail very quickly at the end of their life. ignition delay (id) is the period between the start of fuel injection and the start of combustion (soc) and is usually shown by the crank angle (oca) (saravanan et al., 2014). during this time, the fuel is atomized and mixed with air, and then the heat generated by compression raises the temperature and pressure of the fuel-air mixture until it reaches its ignition point (aldhaidhawi et al., 2017; pham and cao, 2023). once this temperature is reached, the fuel starts to burn rapidly, releasing energy to drive the engine. biodiesel helps to reduce id and this trend becomes more obvious as the proportion of biodiesel in fuel increases (allen et al., 2013). this is explained by two main reasons: the oxygen content and the cetane number of the fuel. biodiesel becomes more combustible due to the increased presence of oxygen in the fuel mixture, which also assists in breaking down larger fatty acids in biodiesel into smaller molecules, resulting in the production of a greater number of volatile substances (singh et al., 2021). in addition, the cetane number is often inversely related to the ignition delay time and the cetane number of biodiesel is usually higher than that of diesel (bittle et al., 2010). therefore, it is not surprising that many studies in table 2 have shown that increasing the ratio of biodiesel in the fuel will decrease the ignition delay. this is an advantage of biodiesel since with a shorter id, the fuel will have more time to burn, leading to more complete combustion within the engine cylinder (agarwal et al., 2013; bednarski et al., 2019). this makes it possible for the engine to take advantage of more of the potential energy generated from the fuel and thus also reduce the power loss represented by the exhaust gases, such as smoke, particulate matter, or unburnt hydrocarbon. however, a very short ignition delay may cause a knock or excessive pressure rise, leading to engine damage. therefore, understanding the properties of different types of diesel engines will help manufacturers recommend suitable biodiesel or its ratio in the mixture of fuel. heat release rate (hrr) has an extremely close relationship with ignition delay. the hrr of a diesel engine is a measure of the amount of heat energy released by the combustion process in the engine cylinder over time (kaya and kökkülünk, 2020). the ignition delay affects the heat release rate because it determines the timing and duration of the combustion process. given the lower calorific value (cv) of biodiesel but a shorter ignition delay (id), it is not difficult to see that the heat release rate of an engine using either biodiesel fuel or a biodiesel-diesel fuel mixture will be significantly reduced compared to engines using only diesel. besides, the fact that biodiesel has a higher kinematic viscosity and density than diesel is also another important factor that reduces the hrr of the engine (shahabuddin et al., 2013). it is easy to see that a higher hrr will help the engine produce better fuel energy, and biodiesel makes the hrr of the engine lower, which significantly affects engine performance. pressure rise rate (prr) is another important factor to control the performance and durability of the engine (wei et al., 2018). although the pressure rise rate of the engine at a low load increases more rapidly when using biodiesel, a similar trend does not occur at a high load. temperature and pressure are two interrelated quantities that exhibit a direct proportionality. additionally, diesel fuel fig. 3. the different between cylinder gas pressure and heat release rate with the crank angle of diesel fuel and different biodiesels (ramalingam and mahalakshmi, 2020; seraç et al., 2020; silitonga et al., 2017; zarrinkolah and hosseini, 2022)(a) soybean-based biodiesel (b) moringa oleifera biodiesel (c) jatropha curcas biodiesel (d) sunflower methyl ester biodiesel v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 726 issn: 2252-4940/© 2023. the author(s). published by cbiore generally possesses greater energy density than biodiesel (bergthorson and thomson, 2015; elkelawy et al., 2019). therefore, under heavier loads, engines running on biodiesel tend to experience a slower increase in pressure rate and consequently, the peak pressure is also lower compared to when using diesel fuel (tamilselvan et al., 2017). figure 3 shows the difference between cylinder gas pressure and heat release rate with the crank angle of diesel fuel and different biodiesels (ramalingam and mahalakshmi, 2020; seraç et al., 2020; silitonga et al., 2017; zarrinkolah and hosseini, 2022). the difference is not too large to require serious modifications to the engine, however, to avoid affecting the power experience that people are used to with diesel engines, many measures are still proposed by researchers which will be clarified in the next section. 3.2. effects of biodiesel on engine performance with the effects on the engine characteristics recorded above, biodiesel surely has effects on engine performance. common parameters used to evaluate engine performance include brake-specific fuel consumption (bsfc), brake thermal efficiency (bte), and exhaust gas temperature (egt). while brake-specific fuel consumption (bsfc) and brake thermal efficiency (bte) are intended to help comprehensively evaluate the engine's ability to generate power as well as fuel consumption, exhaust gas temperature (egt) provide valuable information about the combustion efficiency, the state of the engine components, and the overall health of the engine (sivaramakrishnan and ravikumar, 2014). monitoring and controlling these parameters is crucial to prevent damage to the engine and its components. it also helps optimize the performance, efficiency, and durability of the engine while also ensuring compliance with emission regulations. bsfc is a measure of the fuel efficiency of an engine. it is defined as the amount of fuel consumed per unit of power produced by the engine and it also expresses the proportion of fuel mass used by the engine relative to the amount of braking power it generates. bsfc of engines using biodiesel will increase significantly. this is explained by bsfc having a close relationship with the viscosity, density, and especially the calorific value of the fuel (a.v.s.l et al., 2021). usually, the lower the calorific value, the higher the bsfc will be. meanwhile, higher viscosity and density of fuel can also increase the bsfc because they can increase the friction between the fuel and the engine components, resulting in more energy losses due to friction. this, in turn, can cause the engine to work harder and consume more fuel to produce the same amount of power output, leading to a higher bsfc value. additionally, higher viscosity and density can also cause the fuel to atomize less effectively, leading to incomplete combustion and further increasing fuel consumption and bsfc (kathirvelu et al., 2017; temizer et al., 2020). however, if biodiesel is used as a fuel blend with diesel fuel in small proportions, typically less than 20% of the fuel density, the bsfc is recorded to be insignificant (canakci and van gerpen, 2003; pullagura et al., 2023) in theory, changing the compress ratio appropriately will improve the bsfc for all fuels, however, the efficiency when changing the compress ratio on biodiesel engines is noted to be much better than diesel engines (suresh et al., 2018). with high compression ratios, biodiesel is reported to have lower volatility and higher kinematic viscosity, which directly improves engine performance. figure 4a compares the fuel consumption of some biodiesel with diesel. some exceptions like argemone biodiesel or home oil-based biodiesel have significantly improved bsfc. these fuel sources are said to be extremely potent, and more research is needed to understand this phenomenon. in general, although the bsfc of the engine depends a lot on many different parameters such as engine type, engine operating conditions, fuel injection pressure (fip), the fuel injection method, and so on, biodiesel will always tend to increase bsfc under the same operating conditions if compared with diesel. meanwhile, brake thermal efficiency (bte) is a measure of the efficiency of an engine in converting the energy contained fig. 4. comparison of engine performance using biodiesel and diesel; (a) difference of bsfc (b) difference of bte v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 727 issn: 2252-4940/© 2023. the author(s). published by cbiore in the fuel into useful work. it is defined as the ratio of the engine's brake power output to the energy content of the fuel consumption by the engine. the poor vaporization characteristics of biodiesel make the engine tend to use a lot of energy to produce useful work, which causes the bte of the biodiesel engine to be significantly reduced at low engine speed and load ranges (karthikeyan et al., 2020). however, at higher engine loads and speeds, where the vaporization of the fuel is also smoother, the oxygen content in the component helps the fuel to burn more cleanly, reducing the loss of useful power (jindal et al., 2010). both the bsfc and bte of the engine are significantly affected by the difficulty of vaporization of biodiesel, especially in the low load ranges of the engine (agarwal et al., 2017). researchers have proposed an extremely effective method to solve the above problem is adding an amount of alcohol to the fuel. the amount of alcohol with the characteristic of having low viscosity and density will reduce the overall viscosity and density of the fuel (veza et al., 2022), making it easier for vaporization to occur, thereby improving the efficiency of the engine (duraisamy et al., 2021; padhee and raheman, 2015; truong et al., 2021). blends of biodiesel with ethanol or methanol have been reported to significantly improve bte over diesel fuel regardless of operating conditions, in which biodiesel alone shows weakness (el-seesy et al., 2021; venu and madhavan, 2017). however, the ratio between biodiesel and alcohol content needs to be studied and calculated carefully because an excessive alcohol ratio will cause a cooling effect due to the high latent heat of the vaporization of alcohol (erdiwansyah et al., 2019; yilmaz et al., 2016). this will in turn reduce the bte. it can be seen that the relationship between the physicochemical properties of the fuel and the bte of the engine is not quite linear and it is necessary to find the optimal points to help the engine operate more smoothly. figure 4b shows the comparison between the brake thermal efficiency of some biodiesel with diesel fuel (riyadi et al., 2023). diesel engines have a continuous development history to exploit and make the most of the energy produced from diesel fuel. when changing diesel with biodiesel, regardless of the change in their composition and chemical properties as mentioned above immediately affect the performance of the engine, both negative and positive effects (nagarajan et al., 2022). however, the negative effects are more obvious. data in table 2 shows recent research on the effect of biodiesel on engine performance and emission characteristics. most biodiesel reduces engine performance, as shown by increasing bsfc and decreasing bte (dubey et al., 2022; more et al., 2020; perumal and ilangkumaran, 2018). the use of biodiesel as the main fuel without any improvement in fuel or engine characteristics has been proven both theoretically and experimentally to reduce the performance of the engine significantly. however, when using biodiesel fuel as a second fuel source to mix with diesel fuel, the engine has minimized its power loss. many studies have shown that the 20% biodiesel blending ratio (b20) is the optimal fuel ratio for the engine when the changes in bte and bsfc are insignificant (canakci and van gerpen, 2003; jindal et al., 2010; lue et al., 2001). exhaust gas temperature (egt) is another parameter that also needs to be paid attention to when analyzing the combustion of an engine. during low engine loads, the exhaust gas temperature (egt) tends to be lower due to reduced fuel consumption and subsequent lower heat production. conversely, at high engine loads, the egt typically rises as a result of increased fuel combustion and the generation of greater heat (mehta et al., 2010; uyumaz et al., 2014). heat is also a type of energy produced by combustion, so a higher exhaust gas temperature also means that combustion produces more energy. however, this is wasted energy that is not useful for engine operating processes. the cause of higher exhaust gas temperature can also come from incomplete combustion leading to a significant amount of unburned particles in the exhaust gas. besides, too high exhaust gas temperature will put pressure on machine parts such as the pistons, valves, and exhaust system, reducing their durability and causing damage if continued for a long time. usually, the egt of a diesel engine can range from 300°c to 700°c or even higher, depending on the operating conditions and characteristics of the fuel. however, the recognition of the trend of the egt trend of biodiesel engines is not unanimous. some studies recorded a decrease in egt (ghazali et al., 2015; kegl, 2011; shrivastava et al., 2019) while others noted an increase in egt (abed et al., 2018; kerihuel et al., 2005). however, the trend of increasing egt is more recognized by many studies. in the case of a decrease in egt, the phenomenon is explained that biodiesel has a higher oxygen content and a lower carbon-to-hydrogen ratio than diesel fuel as well as its lower heating value (lhv), which can lead to more complete combustion and less unburned fuel in the exhaust gases (al-lwayzy and yusaf, 2017; haşimoğlu et al., 2008; yilmaz et al., 2014). conversely, the higher viscosity of methyl esters, leading to inadequate fuel atomization and vaporization, can account for the delayed combustion of injected fuel. this delayed burning process elucidates the rise in exhaust gas temperature (egt) in the engine (yilmaz et al., 2014; yilmaz and atmanli, 2017). 3.3. effects of biodiesel on emissions characteristics in the opposite direction, unlike creating negative effects on engine performance, biodiesel has always been known as a fuel source to help reduce emissions. biodiesel is an oxygenated fuel with a more complete combustion process, leading to significantly improved emission parameters (elkelawy et al., 2021). table 3 shows recent research about the effects of biodiesel on emission characteristics and compounds. all the data is collected when the engine is operating at 100% load if the engine load is not specified. most studies show that the use of biodiesel significantly reduces uhc and co (ahmad and saini, 2022; joy et al., 2018; vellaiyan and partheeban, 2018). the unburnt hydrocarbons (uhc) as the name suggests are the result of incomplete combustion of the fuel in the engine. the term "uhc" refers to all varieties of hydrocarbon compounds produced by an engine, but which cannot be assessed separately. as a result, depending on their makeup, they are categorized and referred to as uhc emissions comparable to c1, c3, or c6. when the air-fuel ratio is either too rich or too low for auto-ignition, uhc emissions take place and the combination cannot sustain a flame or ignite automatically (mofijur et al., 2016). the fact that biodiesel has a significantly higher oxygen content than diesel fuel, and the cetane number of biodiesel is also slightly higher than diesel fuel; thus biodiesel could help the combustion process be more complete. this directly reduces the amount of uhc formed (abed et al., 2019; e et al., 2017). interestingly, mixing a small amount of alcohol as mentioned above to improve engine performance also reduces the amount of uhc. this makes the solution of mixing alcohols of particular attention to researchers. a similar problem occurs with co emissions because co is the result of fuel combustion under bad conditions such as a lack of oxygen or improper air-to-fuel ratio (kim et al., 2018). therefore, the presence of oxygen in the composition of biodiesel also helps to reduce co formation significantly by creating conditions for co emissions to be able to convert into co2, v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 728 issn: 2252-4940/© 2023. the author(s). published by cbiore leading to a significant reduction in co emissions (abed et al., 2018). however, one major difference between uhc and co is that mixing a small amount of alcohol can significantly increase the amount of co. specifically, the amount of co was reported t a b le 3 e ff e c ts o f b io d ie se l o n e m is si o n c h a ra c te ri st ic s b le n d e d f u e l r a ti o e m is s io n p e rf o rm a n c e r e f u h c c o n o x s m o k e r ic e b ra n m e th y l e st e r b 2 0 ↓ 1 0 .5 % ↑ 1 6 % ↓ 1 0 % (j a y a p ra b a k a r a n d k a rt h ik e y a n , 2 0 1 6 ) b 1 0 d 9 0 , b 2 0 d 8 0 , b 5 0 d 5 0 , b 1 0 0 ↔ f o r b 2 0 . ↑ 3 0 % , 1 8 % , 3 3 % fo r b 1 0 , b 5 0 , b 1 0 0 ↑ 2 0 % f o r b 1 0 ↔ f o r b 5 0 ↓ 4 0 % a n d 2 0 % fo r b 2 0 a n d b 1 0 0 ↔ f o r b 1 0 , b 2 0 ↑ 6 .2 5 % a n d 9 .2 % f o r b 5 0 , b 1 0 0 ↓ 2 7 % a n d 3 3 % f o r b 1 0 , b 2 0 ↑ 2 % f o r b 5 0 , b 1 0 0 (s h a h , 2 0 1 5 ) j a tr o p h a m e th y l e st e r b 1 0 , b 2 0 ↓ 2 .5 % a n d 1 0 .3 % f o r b 1 0 , b 2 0 ↓ 2 0 % a n d 2 9 % f o r b 1 0 , b 2 0 ↑ 3 .1 % a n d 6 .3 % f o r b 1 0 , b 2 0 (m o fi ju r e t a l. , 2 0 1 3 ) b 2 0 ↓ 2 8 .6 % ↓ 4 5 % ↑ 9 .3 % (k u m a r a n d s h a rm a , 2 0 1 6 ) b 2 0 , b 1 0 0 ↓ ↓ 3 4 % a n d 5 0 % f o r b 2 0 , b 1 0 0 ↑ 1 2 .8 % a n d 5 7 % f o r b 2 0 , b 1 0 0 (k a th ir v e lu e t a l. , 2 0 1 7 ) f is h o il m e th y l e st e r b 2 0 , b 1 0 0 ↓ ↓ 2 5 % a n d 4 7 % f o r b 2 0 , b 1 0 0 ↑ 1 3 .2 a n d 2 9 % f o r b 2 0 , b 1 0 0 (k a th ir v e lu e t a l. , 2 0 1 7 ) m o ri n g a m e th y l e st e r b 2 0 ↓ 2 8 .6 % ↓ 3 0 % ↑ 9 .3 % (k u m a r a n d s h a rm a , 2 0 1 6 ) r a p e se e d m e th y l e st e r b 2 5 , b 5 0 , b 7 5 , b 1 0 0 ↓ 1 5 .4 % , 2 6 .7 % , 3 2 .2 % a n d 4 2 .1 % f o r b 2 5 , b 5 0 , b 7 5 a n d b 1 0 0 ↓ 7 .6 % , 2 2 .7 % , 3 0 .4 % , 3 5 .4 % f o r b 2 5 , b 5 0 , b 7 5 a n d b 1 0 0 ↑ 1 4 .4 % , 2 1 .6 % , 2 8 .5 % a n d 3 2 .9 % fo r b 2 5 , b 5 0 , b 7 5 a n d b 1 0 0 ↑ 5 .6 % fo r b 2 5 a n d 1 0 .3 % fo r b 1 0 0 (r a m a n e t a l. , 2 0 1 9 ) p a lm m e th y l e st e r b 2 0 , b 4 0 , b 6 0 , b 1 0 0 ↑ 1 4 % , 4 3 % f o r b 6 0 , b 1 0 0 ↑ 1 0 0 % f o r b 4 0 a n d b 2 0 ↔ f o r b 1 0 0 ↑ 1 0 % f o r b 6 0 , b 2 0 ↑ 2 2 % f o r b 4 0 ↓ i n si g n if ic a n tl y f o r b 2 0 , b 4 0 ↓ 7 % a n d 1 6 % f o r b 6 0 , b 1 0 0 (i q b a l e t a l. , 2 0 1 3 ) b 2 0 ↓ 3 8 % ↓ 1 4 % ↑ 3 7 % ↓ 1 1 .7 % (r o sh a e t a l. , 2 0 1 9 ) b 2 0 d 8 0 ↓ 4 2 .8 % ↓ 5 0 % ↑ 1 .6 % (k u m a r a n d s h a rm a , 2 0 1 6 ) b 3 0 , b 1 0 0 ↓ 1 8 .2 6 % f o r b 3 0 ↑ 3 0 % f o r b 1 0 0 ↑ 2 .3 % f o r b 3 0 (v e rm a a n d s h a rm a , 2 0 1 5 ) s o y b e a n m e th y l e st e r b 3 0 , b 5 0 , b 7 0 ↓ 1 .8 3 % , 2 .9 4 % , 4 .1 8 % fo r b 3 0 , b 5 0 a n d b 7 0 ↓ 1 5 .0 2 % , 3 3 .8 1 % , a n d 3 0 .7 3 % fo r b 3 0 , b 5 0 a n d b 7 0 ↑ 4 .2 8 % , 5 .5 2 % , 1 1 .9 % f o r b 3 0 , b 5 0 a n d b 7 0 ↓ 5 .4 % , 1 8 .0 2 % , 3 4 .0 9 % f o r b 3 0 , b 5 0 a n d b 7 0 (e lk e la w y e t a l. , 2 0 1 9 ) c o c o n u t m e th y l e st e r b 3 0 ↑ 3 1 .2 1 % ↑ 3 .8 % (v e rm a a n d s h a rm a , 2 0 1 5 ) c o tt o n se e d m e th y l e st e r b 5 , b 1 0 , b 2 0 ↓ 1 6 .2 % , 4 1 % a n d 4 7 % f o r b 5 , b 1 0 , b 2 0 ↓ 4 0 % , 6 8 % a n d 7 6 % f o r b 5 , b 1 0 , b 2 0 (s in h a a n d m u ru g a v e lh , 2 0 1 6 ) w a st e c o o k in g m e th y l e st e r b 2 0 , b 5 0 ↓ 2 5 % , 2 7 % f o r b 2 0 , b 5 0 ↓ 1 0 .7 % , 3 2 .1 % f o r b 2 0 , b 5 0 ↑ 1 8 % , 4 1 % f o r b 2 0 , b 5 0 (e la d a w y e t a l. , 2 0 1 3 ) j u li fl o ra m e th y l e st e r b 2 0 , b 3 0 , b 4 0 , b 1 0 0 ↓ 6 % , 9 % , 1 3 % a n d 1 7 % f o r b 2 0 , b 3 0 , b 4 0 , b 1 0 0 ↔ f o r b 2 0 , b 4 0 ↑ 7 % f o r b 3 0 ↓ 3 .8 % f o r b 1 0 0 ↓ 1 .6 % , 1 .4 % , 1 ,1 % f o r b 2 0 , b 3 0 , b 4 0 ↑ 0 .6 % f o r b 1 0 0 ↓ 1 1 .7 % , 4 .4 % f o r b 2 0 , b 3 0 ↑ 1 0 .3 % , 1 4 % fo r b 4 0 a n d b 1 0 0 (a so k a n e t a l. , 2 0 1 9 ) v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 729 issn: 2252-4940/© 2023. the author(s). published by cbiore to have increased by 39.95%, 38.83%, and 12.6% for propanol, butanol, and pentanol, respectively (uyumaz, 2018; zhang et al., 2022). figures 5a and figure 5b show a comparison of the uhc and co emissions of engines using biodiesel with diesel fuel and the similarity in their trends is shown clearly when most biodiesel will reduce uhc and co emissions except in some special cases. on the contrary, the availability of oxygen in the biodiesel composition will increase nox emissions, which is a typical trade-off relationship in most combustion fuel studies (duraisamy et al., 2021; manigandan et al., 2020). nitrogen oxides (nox) are the most hazardous pollutants generated by engines and are dependent on factors such as the combustion temperature and the length of time it is exposed to a hightemperature environment (above 1400oc), the chemical composition of the fuel, and the availability of oxygen in these high-temperature areas (appavu et al., 2021). the cause of nox emissions when using biodiesel increases can be explained by the earlier combustion of the fuel in the combustion chamber along with the improved combustion process in both quality and speed, causing the 87 in the combustion chamber to rise, thereby enhancing the formation of nox (chen et al., 2018; mirhashemi and sadrnia, 2020). in the combustion chamber equipped with electronic injectors, unsaturated biodiesel with higher iodine value is also the main cause of the increase in nox emissions (mofijur et al., 2019). the same phenomenon does not occur with the engine with a common rail direct injection system. all the above signs indicate that the increase in nox emissions of the engine is bound by many causes, however, their relationship has not been interesting and clarified by many studies (rathinam et al., 2018; varatharajan and cheralathan, 2012; zare et al., 2021). not many studies have documented a reduction in nox emissions as shown in table 3 and in those cases, it is explained that the lower heating value of biodiesel compared to diesel fuel and lower id, which forces hot gases to stay in the combustion chamber at high temperature to generate less nox. figure 5c compares the nox emission of different biodiesel to diesel fuel. in addition, the alcohols once again show a suitable fit and have a very important role in turning biodiesel into a more user-friendly fuel as the addition of longer chain alcohols will reduce nox emissions by approximately 27.44%, 19.27%, and 15.05% for pentanol, butanol and propanol compared with a 50% biodiesel blend (uyumaz, 2018). it is undeniable that biodiesel will increase nox emissions, the overall emissions are still significantly reduced, and biodiesel is still widely considered by researchers as a much cleaner energy source than diesel fuel. therefore, instead of completely using biodiesel in an internal combustion engine, using biodiesel as a secondary fuel source to blend to help overcome the inherent weaknesses of diesel while still being able to partly avoid depending on them is a much more potential approach. although it is difficult to find a type of biodiesel that, after blending with diesel oil, can improve all aspects of engine performance as well as reduce all types of emissions, there are three important issues to be aware of when using this method: (i) engine performance should not necessarily be improved, but should focus on reducing emissions, (ii) biodiesel improvement studies are still being carried out and biodiesel applications will be more and more perfect in the future and (iii) fuel blending will help reduce consumption as well as ensure energy security for many countries. of course, the higher the efficiency of the engine, the better, however, the internal combustion engine has been in common use for a long time and the efficiency of the internal combustion engine has thus been widely accepted. as long as the fuels used do not reduce engine performance or do not significantly reduce it, it is acceptable. in other words, the priority when researching fuel blends should be to help reduce emissions rather than improve engine performance. with this approach, the disadvantages of diesel oil cannot be completely overcome but will be improved and also easier to use. furthermore, finding a fuel source that can be mixed with diesel will greatly reduce the need for diesel. this has not been able to completely solve the burning problems of diesel fuel, but it will help countries lacking oil reserves reduce pressure on energy and also give humanity more time to find a solution to the problem. besides, this method is considered one of the extremely simple, economical, and proactive methods to solve fuel and emissions problems in the short term. the physicochemical properties of diesel and biodiesel can vary significantly, and these differences can impact their fig. 2.comparison of emission from the engine using biodiesel and diesel; (a) difference of hc emission (b) difference of co emission (c) difference of nox emission v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 730 issn: 2252-4940/© 2023. the author(s). published by cbiore performance and compatibility when used as blended fuel. t a b le 4 . n a n o p a rt ic le a p p ro a c h e s in i m p ro v in g e n g in e p e rf o rm a n c e a n d e m is si o n c h a ra c te ri st ic s b io d ie s e l n a n o p a rt ic le s e n g in e p e rf o rm a n c e e m is s io n p e rf o rm a n c e r e f b s f c ( o r s f c ) b t e h c c o n o x s m o k e j a tr o p h a c e o 2 p a rt ic le s (4 0 p p m a n d 8 0 p p m ) ↑ in si g n if ic a n t ↓ 1 6 .7 % , 3 8 .9 % fo r 4 0 p p m a n d 8 0 p p m c e o 2 ↓ 2 1 % , 4 0 % fo r 4 0 p p m a n d 8 0 p p m c e o 2 ↓ 7 % , 2 1 % f o r 4 0 p p m a n d 8 0 p p m c e o 2 (s a ji th e t a l. , 2 0 1 0 ) c o 3 o 4 p a rt ic le s ↓ 2 % ↑ 0 .6 % a t 5 0 % l o a d ↓ 8 3 % a t 7 5 % lo a d ↓ 5 0 % a t 7 5 % l o a d ↓ 4 7 % a t 7 5 % l o a d (g a n e sh a n d g o w ri sh a n k a r, 2 0 1 1 ) a lm g p a rt ic le s ↓ 1 % ↑ 0 .2 % a t 5 0 % l o a d ↓ 7 0 % a t 5 0 % lo a d ↓ 6 6 % a t 5 0 % l o a d ↓ 2 0 % a t 5 0 % l o a d (g a n e sh a n d g o w ri sh a n k a r, 2 0 1 1 ) a l 2 o 3 p a rt ic le s (5 0 p p m ) ↓ 1 3 .5 % ↑ 1 2 % ↓ 1 3 .3 % ↓ 4 0 % ↓ 2 0 .8 % ↓ 1 3 .4 % (b a sh a a n d a n a n d , 2 0 1 3 ) t io 2 ↓ 5 % ↑ 6 % ↓ 2 0 % ↔ ↑ 2 1 % (v e n u a n d m a d h a v a n , 2 0 1 6 ) z ro 2 ↔ ↔ ↓ 2 0 % ↔ ↓ 2 1 % (v e n u a n d m a d h a v a n , 2 0 1 6 ) m u st a rd t io 2 n a n o fl u id (1 0 0 a n d 2 0 0 p p m ) ↓ 3 % , 4 .2 % f o r 1 0 0 p p m a n d 2 0 0 p p m ↓ 8 % , 1 3 % fo r 1 0 0 p p m a n d 2 0 0 p p m ↓ i n si g n if ic a n t ↓ 6 ,7 % , 1 6 .7 % f o r 1 0 0 p p m a n d 2 0 0 p p m (y u v a ra ja n e t a l. , 2 0 1 8 ) p o u lt ry l it te r a l 2 o 3 p a rt ic le s ↓ 6 .7 % ↓ 2 5 .4 % ↓ 2 8 .5 % ↓ 4 .4 % ↑ 1 % (r a m e sh e t a l. , 2 0 1 8 ) n e e m c a rb o n n a n o tu b e s ↓ ↑ 3 .4 3 % ↓ 8 .1 5 % ↓ 1 1 .7 7 % ↓ 4 .6 7 % ↓ 5 .7 4 % (g n a n a si k a m a n i e t a l. , 2 0 1 5 ) c a lo p h y ll u m in o p h y ll u m t io 2 p a rt ic le s ↓ 1 3 .8 % ↑ 3 .1 % ↓ 1 2 % ↓ 2 3 % ↑ i n si g n if ic a n t ↓ 7 .7 % (p ra v e e n e t a l. , 2 0 1 8 ) c a u le rp a ra c e m o sa ir o n i i a n d i ro n i ii n a n o fl u id ↓ 1 1 .2 % ↓ 8 .9 % ↓ 1 0 % ↓ 1 0 .3 % (k a rt h ik e y a n a n d p ra th im a , 2 0 1 6 ) m a h u a c u o ↑ in si g n if ic a n t ↔ ↓ 3 3 % ↑ i n si g n if ic a n t ↓ 1 2 .5 % (c h a n d ra se k a ra n e t a l. , 2 0 1 6 ) n e e m c a rb o n n a n o tu b e s (5 0 p p m a n d 1 0 0 p p m ) ↓ 5 .1 % a n d 6 .7 % fo r 5 0 p p m a n d 1 0 0 p p m ↓ 2 % a n d 5 .8 % f o r 5 0 p p m a n d 1 0 0 p p m ↓ 6 .3 % a n d 9 .2 % f o r 5 0 p p m a n d 1 0 0 p p m ↓ 6 .3 % a n d 7 .8 % fo r 5 0 p p m a n d 1 0 0 p p m (r a m a k ri sh n a n e t a l. , 2 0 1 9 ) v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 731 issn: 2252-4940/© 2023. the author(s). published by cbiore however, by carefully adjusting the fuel mix ratio, it is possible to resolve these differences and achieve a desirable balance between the properties of the two fuels and help the engine operate smoothly with high efficiency. in the efforts to help overcome the remaining weaknesses of biodiesel, researchers have applied many different technologies and have achieved remarkable achievements in recent years. one of the recent approaches of researchers is to use nanoparticles additives (hoang, 2021b; kandasamy and sundararaj, 2018; pradeep and senthilkumar, 2021; rameshbabu and senthilkumar, 2021; sathish et al., 2023). applications of nanoparticles are diverse such as nano metalbased particles such as cerium oxide (kumar et al., 2019; shaisundaram et al., 2021), titan oxide (nanthagopal et al., 2017; sunil et al., 2021), zinc oxide (javed et al., 2016; vali et al., 2022), copper oxide (kalaimurugan et al., 2019; rozina et al., 2022), carbon-based nanoparticles (murugesan et al., 2022), nanofluids (kannan et al., 2011; khalife et al., 2017; shaafi et al., 2015). nanoparticles can be used as additives in diesel and biodiesel to increase surface area to volume ratio as well as increase catalytic activity in nano-size metal oxides and metals (hoang et al., 2022b). nano additives directly improve engine combustion by improving heat transfer, catalytic activity, and air fuel mixing rate (karthikeyan et al., 2017; tomar and kumar, 2020). table 4 shows the comparison between with and without nano-additives on engine performance and emission characteristics. the data shows that using nano additives or nanofluid significantly reduces emissions such as hc, co, and smoke, especially in some research, the results show that nanotechnology can even reduce nox emissions, which solves the trade-off problem of emissions in the combustion process of the engine. regarding the engine performance, many studies also show an improvement in bsfc and bte of engines powered by nanoparticles-included fuels although it does not completely solve the problem. besides, because nanoparticles are used as a catalyst in the combustion process, their shelf life and performance will be maintained for a long time if there are no problems such as poisoning or thermal degradation, leading to deactivation. this will make the cost of applying nano additives not too high, but the effect is extremely stable. however, studies aimed at comprehensively assessing the potential of this method are very limited. therefore, it is necessary to have a comprehensive assessment of the use of nanoparticles for blending with biodiesel. 4. challenges and opportunities with the ever-increasing energy demand and political instability directly affecting the supply stability of fossil fuels, the development and utilization of biodiesel becomes more relevant than ever (coşofreţ et al., 2016). the diversity in production inputs, coupled with the fact that it has been shown to significantly reduce greenhouse gas emissions, which has been a sore point in recent years, creates extremely favorable conditions for biodiesel to compete with fossil fuel sources. unlike traditional fossil fuels, biodiesel can be produced domestically, reducing the need for foreign oil imports. this can also help to stimulate local economies by creating jobs in the production and distribution of biodiesel. another opportunity for biodiesel is its potential to reduce greenhouse gas emissions. biodiesel is considered a low-carbon fuel, meaning that it produces fewer greenhouse gas emissions than traditional fossil fuels (semwal et al., 2022). this can help countries and companies meet their emissions reduction targets and contribute to global efforts to combat climate change (babatunde et al., 2022). additionally, biodiesel can be used in a wide range of applications, including transportation, heating, and electricity generation. this versatility makes biodiesel a flexible and adaptable fuel source that can meet a variety of energy needs. however, like any new technology or industry, biodiesel faces both challenges and opportunities. one of the primary challenges of biodiesel is its cost, while the production of biodiesel has become more efficient and costeffective in recent years, it is still more expensive than traditional fossil fuels (maroušek et al., 2023a; meira et al., 2015). this is partly due to the higher cost of feedstocks, such as soybean oil, which is used to produce biodiesel. additionally, the cost of production equipment, infrastructure, and transportation can also be higher than traditional fossil fuel production (kumar et al., 2021). however, according to many researchers, the cost of raw materials, especially biomass feedstock, accounts for most of the financial structure (apostolakou et al., 2009). therefore, finding cheap fuel will greatly improve profits. to solve the above problem, the use of biomass sources to convert into biodiesel must be considered carefully. one potential approach not only to reducing the cost of biodiesel production but also improving other environmental issues is by utilizing biomass sources derived from by-products and waste products from production and living activities (baldia et al., 2023; sung et al., 2021; vani et al., 2022). however, they are still in the early stages of development and are not yet widely available or cost-effective. secondly, biodiesel can hurt engine performance in cold weather conditions (rochelle and najafi, 2019). this makes the use of biodiesel at certain times of the year or in some countries with cold climates extremely unsuitable. biodiesel has a higher cloud point and pours point than petroleum diesel, which means that it solidifies at a higher temperature, making it challenging fig. 3. comparison between diesel and biodiesel in different start temperatures (zare et al., 2017) v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 732 issn: 2252-4940/© 2023. the author(s). published by cbiore to use in low-temperature conditions (sani et al., 2018; su et al., 2021). this can result in engine starting problems and fuel filter clogging, which can affect engine performance and reliability (nursyairah et al., 2022). many studies have been done to evaluate the performance of biodiesel engines at low temperatures and most of the results show difficulty in cold-start the engine (hadi et al., 2023; yubaidah, 2023). besides, biodiesel that has solidified or thickened due to cold weather may not properly atomize and mix with air in the combustion chamber, resulting in incomplete combustion and reduced engine performance (chaichan et al., 2020; clenci et al., 2016; jiaqiang et al., 2019). figure 6 compares the advantages and disadvantages of biodiesel and diesel in cold-start and hot-start (zare et al., 2017). it can be seen that the use of biodiesel at low temperatures completely loses the natural advantages of this fuel. this is a research direction that needs to receive a lot of attention to make it possible to use biodiesel in different weather and temperature conditions. besides improving engine performance and reducing emissions, another problem for the long-term and widespread use of biodiesel that needs to be analyzed and evaluated is the possibility of engine damage (dharma et al., 2023). engine performance and emission index are parameters that only show the immediate suitability of biodiesel. in case the use of biodiesel causes the engine to degrade quickly, biodiesel is very unlikely to be considered a sustainable alternative fuel source. one of the criteria to evaluate the suitability of the fuel, in the long run, is the degree of deposit formation of the fuel when used in the engine (zhang et al., 2020). some studies have shown that biodiesel has poor atomization and low evaporation, which leads to larger fuel droplets as well as the heterogeneity of the fuel mixture, which directly increases the possibility of deposit formation (liaquat et al., 2014). however, the three most important causes of scale formation are thought to be temperature, nozzle geometry, and fuel composition (leedham et al., 2004). birgel et al. (birgel et al., 2011) experimented with the deposit formation on the injector using different fuels, and the results were shown in figure 7. it can be seen clearly with the naked eye that biodiesel significantly increases the amount of deposit formation on the injector (hoang and le, 2019). this is still a challenge with efforts to bring biodiesel into widespread use. many researchers propose several solutions to try to limit the formation of deposits. mulyono et al. (mulyono et al., 2018) used the hydrotreating method and got some positive results. the results show that the formation and growth of scale are slower in hydrotreated vegetable oil than in biodiesel. besides, the above phenomenon can be partly solved by improving the parameters of biodiesel. biodiesel has high viscosity, while fuel with high viscosity requires a longer ignition delay because the fuel droplets take longer to vaporize, which makes scale formation more likely to occur (emiroğlu, 2019). although there are many theoretical studies explaining the cause for the formation of biodiesel scale, the solutions to solve this phenomenon are still very limited. this also could be a potential direction for biodiesel fuel researchers another aspect that needs to be considered in the long term is the engine corrosion of biodiesel. the corrosive potential of biodiesel is rated as much higher than that of diesel because of its high oxygen content (fazal et al., 2012; hoang et al., 2020). in addition, the biodiesel production process can generate impurities such as free fatty acids, glycerol, and metal catalyst residues. if not handled properly, these impurities can participate in reactions that corrode metals. fazal et al. (fazal et al., 2010) observed corrosion rate of palm oil biodiesel with copper, and aluminium is 0.586 and 0.202 mils per year (mpy) while for diesel, the corrosion rate is only 0.3 mpy and 0.15 mpy respectively. in another study, saravana kannan thangavelu et. al also reported a higher corrosion rate when blending biodiesel into diesel. specifically, b20d75e5 (20% biodiesel, 75% diesel, and 5% ethanol) and b20d70e10 (20% biodiesel, 70% diesel, and 10% ethanol) have a corrosion rate of 0.1572 and 0.1817 mpy respectively while diesel has a corrosion rate of only 0.1572 and 0.1817 mpy, respectively. 0.0523 mpy (thangavelu et al., 2016). this is a significant increase when just mixing 20% biodiesel into the fuel, but it increases the corrosion rate by more than 3 times. this is considered a serious problem because it not only reduces engine performance but also causes engine damage, directly increases warranty and repair costs, or even raises a big question mark to safety concerns. these are all major obstacles to the widespread dissemination of biodiesel and require research to come up with optimal solutions. with its enormous potential but still underappreciated today by businesses and citizens, the role of government in promoting and supporting this fuel is more widespread is indispensable. policies for biodiesel around the world vary widely, depending on factors such as government priorities, energy security goals, and environmental concerns (austin et al., 2022). some countries have implemented ambitious targets for biodiesel production and use, while others have been slower to adopt this renewable fuel source. in europe, renewable energy directive has set a target of 14% renewable energy in transportation by 2030, which includes the use of biodiesel (long et al., 2021). many european countries have implemented mandatory biodiesel blending policies, with blending ratios ranging from 7% to 20% (chong et al., 2021). for example, in germany, diesel fuel must contain a minimum of 7% biodiesel, while in france the mandatory blend is 8.5%. in the united states, the federal government has implemented a renewable fig. 4. the optical investigation for deposit formation level evaluation on injectors (a) new nozzle, (b) diesel fuel, (c) bd30, (d) bd100 (birgel et al., 2011) v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 733 issn: 2252-4940/© 2023. the author(s). published by cbiore fuel standard program, which requires a certain volume of renewable fuel to be blended into gasoline and diesel fuel. biodiesel is included as a renewable fuel under the rfs, and the program has helped to promote the growth of the domestic biodiesel industry. in addition, some states, such as california, have implemented low-carbon fuel standards that incentivize the use of biodiesel and other low-carbon fuels. in south america, brazil is a leading producer and user of biodiesel, with a mandatory blending policy that requires all diesel fuel to contain at least 13% biodiesel (de souza et al., 2022). argentina and colombia have also implemented mandatory blending policies, with blending ratios of 10% and 8%, respectively (canabarro et al., 2023). in asia, several countries have implemented biodiesel policies to promote renewable energy and reduce dependence on imported fossil fuels. for example, indonesia has set a target of 30% renewable energy in transportation by 2025, with biodiesel playing a key role in achieving this goal (kharina et al., 2016). malaysia has also implemented a biodiesel blending policy, with a mandatory blend of 10% (zulqarnain et al., 2020). overall, biodiesel policies around the world are evolving as governments seek to promote sustainable energy sources and reduce their reliance on fossil fuels. while the specifics of these policies vary widely, they all aim to promote the growth of the biodiesel industry and reduce greenhouse gas emissions from transportation. 5. conclusions the study presents the important properties of biodiesel and updates the latest research to improve engine performance and reduce emissions when using biodiesel. it can be seen that, with its physicochemical properties, biodiesel improves the engine's emission indicators significantly, however, operational issues such as performance or durability of the engine have become a problem that scientists are trying to resolve. completely independent use of biodiesel without engine modifications is theoretically possible but is a huge minus point for both engine performance and operating costs. the use of biodiesel as a fuel mixed with diesel fuel will be more reasonable at present. besides, to fully realize the opportunities of biodiesel, several steps need to be taken. one of the most important is to continue to invest in research and development to improve the efficiency and cost-effectiveness of biodiesel production. this includes developing new feedstocks and production methods, as well as improving the efficiency of existing production processes. in addition, social policies such as tax incentives, subsidies, and mandates that require a certain percentage of transportation fuel to be made from renewable sources are also necessary to support the commercialization of biodiesel. this also will help to educate the public and raise awareness about the benefits of biodiesel. success in utilizing the fullest potential of biodiesel will relieve pressure on energy issues in many countries, setting the stage for sustainable development. conflicts of interest: the authors declare no conflict of interest. references a.v.s.l, s.b., subramaniapillai, n., khadhar mohamed, m.s.b., narayanan, a., 2021. effect of rubber seed oil biodiesel on engine performance and emission analysis. fuel 296, 120708. https://doi.org/10.1016/j.fuel.2021.120708 abed, k.a., el morsi, a.k., sayed, m.m., shaib, a.a. el, gad, m.s., 2018. effect of waste cooking-oil biodiesel on performance and exhaust emissions of a diesel engine. egypt. j. pet. 27, 985–989. https://doi.org/10.1016/j.ejpe.2018.02.008 abed, k.a., gad, m.s., el morsi, a.k., sayed, m.m., elyazeed, s.a., 2019. effect of biodiesel fuels on diesel engine emissions. egypt. j. pet. 28, 183–188. https://doi.org/10.1016/j.ejpe.2019.03.001 agarwal, a.k., gupta, j.g., dhar, a., 2017. potential and challenges for large-scale application of biodiesel in automotive sector. prog. energy combust. sci. 61, 113–149. https://doi.org/10.1016/j.pecs.2017.03.002 agarwal, a.k., srivastava, d.k., dhar, a., maurya, r.k., shukla, p.c., singh, a.p., 2013. effect of fuel injection timing and pressure on combustion, emissions and performance characteristics of a single cylinder diesel engine. fuel 111, 374–383. ahmad, k., saini, p., 2022. effect of butanol additive with mango seed biodiesel and diesel ternary blends on performance and emission characteristics of diesel engine. energy sources, part a recover. util. environ. eff. 44, 9988–10005. https://doi.org/10.1080/15567036.2022.2143954 al-lwayzy, s.h., yusaf, t., 2017. diesel engine performance and exhaust gas emissions using microalgae chlorella protothecoides biodiesel. renew. energy 101, 690–701. https://doi.org/10.1016/j.renene.2016.09.035 aldhaidhawi, m., chiriac, r., badescu, v., 2017. ignition delay, combustion and emission characteristics of diesel engine fueled with rapeseed biodiesel – a literature review. renew. sustain. energy rev. 73, 178–186. https://doi.org/10.1016/j.rser.2017.01.129 allen, c., toulson, e., tepe, d., schock, h., miller, d., lee, t., 2013. characterization of the effect of fatty ester composition on the ignition behavior of biodiesel fuel sprays. fuel 111, 659–669. https://doi.org/10.1016/j.fuel.2013.03.057 alptekin, e., 2017. emission, injection and combustion characteristics of biodiesel and oxygenated fuel blends in a common rail diesel engine. energy 119, 44–52. https://doi.org/10.1016/j.energy.2016.12.069 alptekin, e., canakci, m., 2008. determination of the density and the viscosities of biodiesel-diesel fuel blends.renewable energy https://doi.org/10.1016/j.renene.2008.02.020 an, h., yang, w.m., maghbouli, a., li, j., chou, s.k., chua, k.j., 2013. performance, combustion and emission characteristics of biodiesel derived from waste cooking oils. appl. energy 112, 493– 499. https://doi.org/10.1016/j.apenergy.2012.12.044 apostolakou, a.a., kookos, i.k., marazioti, c., angelopoulos, k.c., 2009. techno-economic analysis of a biodiesel production process from vegetable oils. fuel process. technol. 90, 1023–1031. https://doi.org/10.1016/j.fuproc.2009.04.017 appavu, p., ramanan m, v., jayaraman, j., venu, h., 2021. no x emission reduction techniques in biodiesel-fuelled ci engine: a review. aust. j. mech. eng. 19, 210–220. https://doi.org/10.1080/14484846.2019.1596527 asokan, m.a., senthur prabu, s., bade, p.k.k., nekkanti, v.m., gutta, s.s.g., 2019. performance, combustion and emission characteristics of juliflora biodiesel fuelled di diesel engine. energy 173, 883–892. https://doi.org/10.1016/j.energy.2019.02.075 austin, k.g., jones, j.p.h., clark, c.m., 2022. a review of domestic land use change attributable to u.s. biofuel policy. renew. sustain. energy rev. 159, 112181. https://doi.org/10.1016/j.rser.2022.112181 ayhan, v., çangal, ç., cesur, i̇., çoban, a., ergen, g., çay, y., kolip, a., özsert, i̇., 2020. optimization of the factors affecting performance and emissions in a diesel engine using biodiesel and egr with taguchi method. fuel 261, 116371. babatunde, k.a., salam, k.k., aworanti, o.a., olu-arotiowa, o.a., alagbe, s.o., oluwole, t.d., 2022. transesterification of castor oil: neuro-fuzzy modelling, uncertainty quantification and optimization study. syst. microbiol. biomanufacturing. https://doi.org/10.1007/s43393-022-00120-9 bakır, h., ağbulut, ü., gürel, a.e., yıldız, g., güvenç, u., soudagar, m.e.m., hoang, a.t., deepanraj, b., saini, g., afzal, a., 2022. forecasting of future greenhouse gas emission trajectory for india using energy and economic indexes with various metaheuristic algorithms. j. clean. prod. 360, 131946. https://doi.org/10.1016/j.jclepro.2022.131946 balasubramanian, d., wongwuttanasatian, t., venugopal, i.p., rajarajan, a., 2022. exploration of combustion behavior in a https://doi.org/10.1016/j.fuel.2021.120708 https://doi.org/10.1016/j.ejpe.2018.02.008 https://doi.org/10.1016/j.ejpe.2019.03.001 https://doi.org/10.1016/j.pecs.2017.03.002 https://doi.org/10.1080/15567036.2022.2143954 https://doi.org/10.1016/j.renene.2016.09.035 https://doi.org/10.1016/j.rser.2017.01.129 https://doi.org/10.1016/j.fuel.2013.03.057 https://doi.org/10.1016/j.energy.2016.12.069 https://doi.org/10.1016/j.renene.2008.02.020 https://doi.org/10.1016/j.apenergy.2012.12.044 https://doi.org/10.1016/j.fuproc.2009.04.017 https://doi.org/10.1080/14484846.2019.1596527 https://doi.org/10.1016/j.energy.2019.02.075 https://doi.org/10.1016/j.rser.2022.112181 https://doi.org/10.1007/s43393-022-00120-9 https://doi.org/10.1016/j.jclepro.2022.131946 v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 734 issn: 2252-4940/© 2023. the author(s). published by cbiore compression ignition engine fuelled with low-viscous pimpinella anisum and waste cooking oil biodiesel blends. j. clean. prod. 331, 129999. https://doi.org/10.1016/j.jclepro.2021.129999 balat, m., 2011. potential alternatives to edible oils for biodiesel production – a review of current work. energy convers. manag. 52, 1479–1492. https://doi.org/10.1016/j.enconman.2010.10.011 balat, m., balat, h., 2010. progress in biodiesel processing. appl. energy 87, 1815–1835. https://doi.org/10.1016/j.apenergy.2010.01.012 baldia, a., rajput, d., kumar, a., pandey, a., dubey, k.k., 2023. engineering microalgae as the next-generation food. syst. microbiol. biomanufacturing 3, 166–178. https://doi.org/10.1007/s43393-022-00144-1 basha, j.s., anand, r.b., 2013. the influence of nano additive blended biodiesel fuels on the working characteristics of a diesel engine. j. brazilian soc. mech. sci. eng. https://doi.org/10.1007/s40430013-0023-0 bednarski, m., orliński, p., wojs, m.k., sikora, m., 2019. evaluation of methods for determining the combustion ignition delay in a diesel engine powered by liquid biofuel. j. energy inst. 92, 1107– 1114. https://doi.org/10.1016/j.joei.2018.06.007 bergthorson, j.m., thomson, m.j., 2015. a review of the combustion and emissions properties of advanced transportation biofuels and their impact on existing and future engines. renew. sustain. energy rev. 42, 1393–1417. https://doi.org/10.1016/j.rser.2014.10.034 birgel, a., ladommatos, n., aleiferis, p., milovanovic, n., lacey, p., richards, p., 2011. investigations on deposit formation in the holes of diesel injector nozzles, in: sae technical papers. pp. 123–131. bittle, j.a., knight, b.m., jacobs, t.j., 2010. interesting behavior of biodiesel ignition delay and combustion duration. energy & fuels 24, 4166–4177. https://doi.org/10.1021/ef1004539 buyukkaya, e., 2010. effects of biodiesel on a di diesel engine performance, emission and combustion characteristics. fuel 89, 3099–3105. https://doi.org/10.1016/j.fuel.2010.05.034 canabarro, n.i., silva-ortiz, p., nogueira, l.a.h., cantarella, h., macielfilho, r., souza, g.m., 2023. sustainability assessment of ethanol and biodiesel production in argentina, brazil, colombia, and guatemala. renew. sustain. energy rev. 171, 113019. https://doi.org/10.1016/j.rser.2022.113019 canakci, m., van gerpen, j.h., 2003. comparison of engine performance and emissions for petroleum diesel fuel, yellow grease biodiesel, and soybean oil biodiesel. trans. asae 46, 937-. https://doi.org/10.13031/2013.13948 černoch, m., hájek, m., skopal, f., 2010. relationships among flash point, carbon residue, viscosity and some impurities in biodiesel after ethanolysis of rapeseed oil. bioresour. technol. 101, 7397– 7401. https://doi.org/10.1016/j.biortech.2010.05.003 chaichan, m., gaaz, t.s., al-amiery, a., mohd, kadhum, a.a., 2020. biodiesel blends startability and emissions during cold, warm and hot conditions. j. nanofluids 9, 75–89. https://doi.org/10.1166/jon.2020.1732 chandrasekaran, v., arthanarisamy, m., nachiappan, p., dhanakotti, s., moorthy, b., 2016. the role of nano additives for biodiesel and diesel blended transportation fuels. transp. res. part d transp. environ. 46, 145–156. https://doi.org/10.1016/j.trd.2016.03.015 chen, h., xie, b., ma, j., chen, y., 2018. nox emission of biodiesel compared to diesel: higher or lower? appl. therm. eng. 137, 584– 593. https://doi.org/10.1016/j.applthermaleng.2018.04.022 chong, c.t., loe, t.y., wong, k.y., ashokkumar, v., lam, s.s., chong, w.t., borrion, a., tian, b., ng, j.-h., 2021. biodiesel sustainability: the global impact of potential biodiesel production on the energy–water–food (ewf) nexus. environ. technol. innov. 22, 101408. https://doi.org/10.1016/j.eti.2021.101408 clenci, a., niculescu, r., danlos, a., iorga-simăn, v., trică, a., 2016. impact of biodiesel blends and di-ethyl-ether on the cold starting performance of a compression ignition engine. energies 9,284. https://doi.org/10.3390/en9040284 coşofreţ, d., popa, c., ristea, m., 2016. study on the greenhouse gases generated by the direct injection diesel engines running on biodiesel. int. conf. knowledge-based organ. 22, 616–621. https://doi.org/10.1515/kbo-2016-0106 de souza, t.a.z., pinto, g.m., julio, a.a.v., coronado, c.j.r., perezherrera, r., siqueira, b.o.p.s., da costa, r.b.r., roberts, j.j., palacio, j.c.e., 2022. biodiesel in south american countries: a review on policies, stages of development and imminent competition with hydrotreated vegetable oil. renew. sustain. energy rev. 153, 111755. https://doi.org/10.1016/j.rser.2021.111755 deepanraj, b., srinivas, m., arun, n., sankaranarayanan, g., abdul salam, p., 2017. comparison of jatropha and karanja biofuels on their combustion characteristics. int. j. green energy 14, 1231– 1237. dhamodaran, g., krishnan, r., pochareddy, y.k., pyarelal, h.m., sivasubramanian, h., ganeshram, a.k., 2017. a comparative study of combustion, emission, and performance characteristics of rice-bran-, neem-, and cottonseed-oil biodiesels with varying degree of unsaturation. fuel 187, 296–305. https://doi.org/10.1016/j.fuel.2016.09.062 dharma, s., silitonga, a.s., shamsuddin, a.h., sebayang, a.h., milano, j., sebayang, r., sarjianto, ibrahim, h., bahri, n., ginting, b., damanik, n., 2023. properties and corrosion behaviors of mild steel in biodiesel-diesel blends. energy sources, part a recover. util. environ. eff. 45, 3887–3899. https://doi.org/10.1080/15567036.2019.1668883 dinesha, p., kumar, s., rosen, m.a., 2019. combined effects of water emulsion and diethyl ether additive on combustion performance and emissions of a compression ignition engine using biodiesel blends. energy 179, 928–937. https://doi.org/10.1016/j.energy.2019.05.071 dubey, a., prasad, r.s., kumar singh, j., nayyar, a., 2022. optimization of diesel engine performance and emissions with biodiesel-diesel blends and egr using response surface methodology (rsm). clean. eng. technol. 8, 100509. https://doi.org/10.1016/j.clet.2022.100509 duraisamy, b., velmurugan, k., venkatachalapathy, v.s.k., thiyagarajan, s., varuvel, e.g., 2021. effect of amyl alcohol addition in a ci engine with prosopis juliflora oil–an experimental study. energy sources, part a recover. util. environ. eff. https://doi.org/10.1080/15567036.2021.1996489 e, j., pham, m., zhao, d., deng, y., le, d., zuo, w., zhu, h., liu, t., peng, q., zhang, z., 2017. effect of different technologies on combustion and emissions of the diesel engine fueled with biodiesel: a review. renew. sustain. energy rev. 80, 620–647. https://doi.org/10.1016/j.rser.2017.05.250 el-adawy, m., ibrahim, a., el-kassaby, m.m., 2013. an experimental evaluation of using waste cooking oil biodiesel in a diesel engine. energy technol. 1, 726–734. https://doi.org/10.1002/ente.201300100 el-seesy, a.i., waly, m.s., he, z., el-batsh, h.m., nasser, a., el-zoheiry, r.m., 2021. influence of quaternary combinations of biodiesel/methanol/n-octanol/diethyl ether from waste cooking oil on combustion, emission, and stability aspects of a diesel engine. energy convers. manag. 240, 114268. https://doi.org/10.1016/j.enconman.2021.114268 elkelawy, m., alm-eldin bastawissi, h., el shenawy, e.a., taha, m., panchal, h., sadasivuni, k.k., 2021. study of performance, combustion, and emissions parameters of di-diesel engine fueled with algae biodiesel/diesel/n-pentane blends. energy convers. manag. x 10, 100058. https://doi.org/10.1016/j.ecmx.2020.100058 elkelawy, m., alm-eldin bastawissi, h., esmaeil, k.k., radwan, a.m., panchal, h., sadasivuni, k.k., ponnamma, d., walvekar, r., 2019. experimental studies on the biodiesel production parameters optimization of sunflower and soybean oil mixture and di engine combustion, performance, and emission analysis fueled with diesel/biodiesel blends. fuel 255, 115791. https://doi.org/10.1016/j.fuel.2019.115791 emiroğlu, a.o., 2019. effect of fuel injection pressure on the characteristics of single cylinder diesel engine powered by butanol-diesel blend. fuel 256, 115928. https://doi.org/10.1016/j.fuel.2019.115928 erdiwansyah, mamat, r., sani, m.s.m., sudhakar, k., kadarohman, a., sardjono, r.., 2019. an overview of higher alcohol and biodiesel as alternative fuels in engines. energy reports 5, 467–479. https://doi.org/10.1016/j.egyr.2019.04.009 https://doi.org/10.1016/j.jclepro.2021.129999 https://doi.org/10.1016/j.enconman.2010.10.011 https://doi.org/10.1016/j.apenergy.2010.01.012 https://doi.org/10.1007/s43393-022-00144-1 https://doi.org/10.1007/s40430-013-0023-0 https://doi.org/10.1007/s40430-013-0023-0 https://doi.org/10.1016/j.joei.2018.06.007 https://doi.org/10.1016/j.rser.2014.10.034 https://doi.org/10.1021/ef1004539 https://doi.org/10.1016/j.fuel.2010.05.034 https://doi.org/10.1016/j.rser.2022.113019 https://doi.org/10.13031/2013.13948 https://doi.org/10.1016/j.biortech.2010.05.003 https://doi.org/10.1166/jon.2020.1732 https://doi.org/10.1016/j.trd.2016.03.015 https://doi.org/10.1016/j.applthermaleng.2018.04.022 https://doi.org/10.1016/j.eti.2021.101408 https://doi.org/10.3390/en9040284 https://doi.org/10.1515/kbo-2016-0106 https://doi.org/10.1016/j.rser.2021.111755 https://doi.org/10.1016/j.fuel.2016.09.062 https://doi.org/10.1080/15567036.2019.1668883 https://doi.org/10.1016/j.energy.2019.05.071 https://doi.org/10.1016/j.clet.2022.100509 https://doi.org/10.1080/15567036.2021.1996489 https://doi.org/10.1016/j.rser.2017.05.250 https://doi.org/10.1002/ente.201300100 https://doi.org/10.1016/j.enconman.2021.114268 https://doi.org/10.1016/j.ecmx.2020.100058 https://doi.org/10.1016/j.fuel.2019.115791 https://doi.org/10.1016/j.fuel.2019.115928 https://doi.org/10.1016/j.egyr.2019.04.009 v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 735 issn: 2252-4940/© 2023. the author(s). published by cbiore fazal, m.a., haseeb, a., masjuki, h.h., 2010. comparative corrosive characteristics of petroleum diesel and palm biodiesel for automotive materials. fuel process. technol. 91, 1308–1315. fazal, m.a., haseeb, a.s.m.a., masjuki, h.h., 2012. degradation of automotive materials in palm biodiesel. energy 40, 76–83. https://doi.org/10.1016/j.energy.2012.02.026 fernández, i.a., gómez, m.r., gómez, j.r., lópez-gonzález, l.m., 2020. generation of h2 on board lng vessels for consumption in the propulsion system. polish marit. res. 27, 83–95. https://doi.org/10.2478/pomr-2020-0009 gad, m.s., el-shafay, a.s., abu hashish, h.m., 2021. assessment of diesel engine performance, emissions and combustion characteristics burning biodiesel blends from jatropha seeds. process saf. environ. prot. 147, 518–526. https://doi.org/10.1016/j.psep.2020.11.034 ganesh, d., gowrishankar, g., 2011. effect of nano-fuel additive on emission reduction in a biodiesel fuelled ci engine, in: 2011 international conference on electrical and control engineering, icece 2011 proceedings. https://doi.org/10.1109/iceceng.2011.6058240 ghazali, w.n.m.w., mamat, r., masjuki, h.h., najafi, g., 2015. effects of biodiesel from different feedstocks on engine performance and emissions: a review. renew. sustain. energy rev. 51, 585–602. giakoumis, e.g., sarakatsanis, c.k., 2018. estimation of biodiesel cetane number, density, kinematic viscosity and heating values from its fatty acid weight composition. fuel 222, 574–585. https://doi.org/10.1016/j.fuel.2018.02.187 global biodiesel production is increasing renewable carbon news [www document], n.d. gnanasekaran, s., saravanan, n., ilangkumaran, m., 2016. influence of injection timing on performance, emission and combustion characteristics of a di diesel engine running on fish oil biodiesel. energy 116, 1218–1229. https://doi.org/10.1016/j.energy.2016.10.039 gnanasikamani, b., balaji, g., cheralathan, m., 2015. effect of cnt as additive with biodiesel on the performance and emission characteristics of a di diesel engine effect of additives on biodiesel view project emission reduction view project effect of cnt as additive with biodiesel on the performance and em. artic. int. j. chemtech res. 7, 1230–1236. godiganur, s., suryanarayana murthy, c., reddy, r.p., 2010. performance and emission characteristics of a kirloskar ha394 diesel engine operated on fish oil methyl esters. renew. energy 35, 355–359. https://doi.org/10.1016/j.renene.2009.07.007 goh, b.h.h., chong, c.t., ong, h.c., milano, j., shamsuddin, a.h., lee, x.j., ng, j.-h., 2022. strategies for fuel property enhancement for second-generation multi-feedstock biodiesel. fuel 315, 123178. https://doi.org/10.1016/j.fuel.2022.123178 hadi, s., ghodrat, m., baghban, m., 2023. case studies in thermal engineering effects of blending energetic iron nanoparticles in b20 fuel on lower co and uhc emissions of the diesel engine in cold start condition. case stud. therm. eng. 41, 102658. https://doi.org/10.1016/j.csite.2022.102658 hadiyanto, h., yuliandaru, i. and hapsari, r. 2018. production of biodiesel from mixed waste cooking and castor oil. matec web conf., 156 (2018) 03056, https://doi.org/10.1051/matecconf/201815603056 hadiyanto, h., lestari, s.p., abdullah, a., widayat, w, sutanto, h. 2016. the development of fly ash-supported cao derived from mollusk shell of anadara granosa and paphia undulata as heterogeneous cao catalyst in biodiesel synthesis. int j energy environ eng 7, 297–305. https://doi.org/10.1007/s40095-0160212-6 haşimoğlu, c., ciniviz, m., özsert, i̇., i̇çingür, y., parlak, a., sahir salman, m., 2008. performance characteristics of a low heat rejection diesel engine operating with biodiesel. renew. energy 33, 1709–1715. https://doi.org/10.1016/j.renene.2007.08.002 hoang, a.t., 2021a. prediction of the density and viscosity of biodiesel and the influence of biodiesel properties on a diesel engine fuel supply system. j. mar. eng. technol. 20, 299–311. https://doi.org/10.1080/20464177.2018.1532734 hoang, a.t., 2021b. combustion behavior, performance and emission characteristics of diesel engine fuelled with biodiesel containing cerium oxide nanoparticles: a review. fuel process. technol. 218, 106840. https://doi.org/10.1016/j.fuproc.2021.106840 hoang, a.t., le, a.t., 2019. a review on deposit formation in the injector of diesel engines running on biodiesel. energy sources, part a recover. util. environ. eff. 41, 584–599. https://doi.org/10.1080/15567036.2018.1520342 hoang, a.t., pham, v.v., 2019. a study of emission characteristic, deposits, and lubrication oil degradation of a diesel engine running on preheated vegetable oil and diesel oil. energy sources, part a recover. util. environ. eff. 41, 611–625. https://doi.org/10.1080/15567036.2018.1520344 hoang, a.t., pham, v.v., nguyen, x.p., 2021a. integrating renewable sources into energy system for smart city as a sagacious strategy towards clean and sustainable process. j. clean. prod. 305, 127161. https://doi.org/10.1016/j.jclepro.2021.127161 hoang, a.t., sirohi, r., pandey, a., nižetić, s., lam, s.s., chen, w.-h., luque, r., thomas, s., arıcı, m., pham, v.v., 2022a. biofuel production from microalgae: challenges and chances. phytochem. rev. https://doi.org/10.1007/s11101-022-09819-y hoang, a.t., tabatabaei, m., aghbashlo, m., 2020. a review of the effect of biodiesel on the corrosion behavior of metals/alloys in diesel engines. energy sources, part a recover. util. environ. eff. 42, 2923–2943. https://doi.org/10.1080/15567036.2019.1623346 hoang, a.t., tabatabaei, m., aghbashlo, m., carlucci, a.p., ölçer, a.i., le, a.t., ghassemi, a., 2021b. rice bran oil-based biodiesel as a promising renewable fuel alternative to petrodiesel: a review. renew. sustain. energy rev. 135, 110204. https://doi.org/10.1016/j.rser.2020.110204 hoang, a.t., xuan le, m., nižetić, s., huang, z., ağbulut, ü., veza, i., said, z., tuan le, a., dung tran, v., phuong nguyen, x., 2022b. understanding behaviors of compression ignition engine running on metal nanoparticle additives-included fuels: a control comparison between biodiesel and diesel fuel. fuel 326, 124981. https://doi.org/10.1016/j.fuel.2022.124981 inbanaathan, p.v., balasubramanian, d., nguyen, v.n., le, v.v., waehayee, m., r, r., veza, i., yukesh, n., kalam, m.a., sonthalia, a., varuvel, e.g., 2023. comprehensive study on using hydrogengasoline-ethanol blends as flexible fuels in an existing variable speed si engine. int. j. hydrogen energy. https://doi.org/10.1016/j.ijhydene.2023.03.107 iqbal, a.m., zainal, z.a., mazlan, m., mustafa al-bakri, a.m., salim, m.s., 2013. performance and emission characteristics of diesel engine running on blended palm oil. adv. mater. res. 795, 164– 169. https://doi.org/10.4028/www.scientific.net/amr.795.1 64 jafari, m., verma, p., bodisco, t.a., zare, a., surawski, n.c., borghesani, p., stevanovic, s., guo, y., alroe, j., osuagwu, c., milic, a., miljevic, b., ristovski, z.d., brown, r.j., 2019. multivariate analysis of performance and emission parameters in a diesel engine using biodiesel and oxygenated additive. energy convers. manag. 201, 112183. https://doi.org/10.1016/j.enconman.2019.112183 javed, s., satyanarayana murthy, y.v.v., satyanarayana, m.r.s., rajeswara reddy, r., rajagopal, k., 2016. effect of a zinc oxide nanoparticle fuel additive on the emission reduction of a hydrogen dual-fuelled engine with jatropha methyl ester biodiesel blends. j. clean. prod. 137, 490–506. https://doi.org/10.1016/j.jclepro.2016.07.125 jayaprabakar, j., karthikeyan, a., 2016. performance and emission characteristics of rice bran and alga biodiesel blends in a ci engine. mater. today proc. 3, 2468–2474. https://doi.org/10.1016/j.matpr.2016.04.164 jeyakumar, n., hoang, a.t., nižetić, s., balasubramanian, d., kamaraj, s., lakshmana pandian, p., sirohi, r., nguyen, p.q.p., nguyen, x.p., 2022. experimental investigation on simultaneous production of bioethanol and biodiesel from macro-algae. fuel 329, 125362. https://doi.org/10.1016/j.fuel.2022.125362 jiaqiang, e., liu, g., zhang, z., han, d., chen, j., wei, k., gong, j., yin, z., 2019. effect analysis on cold starting performance enhancement of a diesel engine fueled with biodiesel fuel based on an improved thermodynamic model. appl. energy 243, 321– 335. https://doi.org/10.1016/j.apenergy.2019.03.204 jindal, s., nandwana, b.p., rathore, n.s., vashistha, v., 2010. experimental investigation of the effect of compression ratio and injection pressure in a direct injection diesel engine running on jatropha methyl ester. appl. therm. eng. https://doi.org/10.1016/j.energy.2012.02.026 https://doi.org/10.2478/pomr-2020-0009 https://doi.org/10.1016/j.psep.2020.11.034 https://doi.org/10.1109/iceceng.2011.6058240 https://doi.org/10.1016/j.fuel.2018.02.187 https://doi.org/10.1016/j.energy.2016.10.039 https://doi.org/10.1016/j.renene.2009.07.007 https://doi.org/10.1016/j.fuel.2022.123178 https://doi.org/10.1016/j.csite.2022.102658 https://doi.org/10.1051/matecconf/201815603056 https://doi.org/10.1007/s40095-016-0212-6 https://doi.org/10.1007/s40095-016-0212-6 https://doi.org/10.1016/j.renene.2007.08.002 https://doi.org/10.1080/20464177.2018.1532734 https://doi.org/10.1016/j.fuproc.2021.106840 https://doi.org/10.1080/15567036.2018.1520342 https://doi.org/10.1080/15567036.2018.1520344 https://doi.org/10.1016/j.jclepro.2021.127161 https://doi.org/10.1007/s11101-022-09819-y https://doi.org/10.1080/15567036.2019.1623346 https://doi.org/10.1016/j.rser.2020.110204 https://doi.org/10.1016/j.fuel.2022.124981 https://doi.org/10.1016/j.ijhydene.2023.03.107 https://doi.org/10.4028/www.scientific.net/amr.795.164 https://doi.org/10.4028/www.scientific.net/amr.795.164 https://doi.org/10.1016/j.enconman.2019.112183 https://doi.org/10.1016/j.jclepro.2016.07.125 https://doi.org/10.1016/j.matpr.2016.04.164 https://doi.org/10.1016/j.fuel.2022.125362 https://doi.org/10.1016/j.apenergy.2019.03.204 v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 736 issn: 2252-4940/© 2023. the author(s). published by cbiore https://doi.org/10.1016/j.applthermaleng.2009.10.004 joy, n., devarajan, y., nagappan, b., anderson, a., 2018. exhaust emission study on neat biodiesel and alcohol blends fueled diesel engine. energy sources, part a recover. util. environ. eff. 40, 115– 119. kalaimurugan, k., karthikeyan, s., periyasamy, m., mahendran, g., dharmaprabhakaran, t., 2019. experimental studies on the influence of copper oxide nanoparticle on biodiesel-diesel fuel blend in ci engine. energy sources, part a recover. util. environ. eff. 1–16. https://doi.org/10.1080/15567036.2019.1679290 kandasamy, s., sundararaj, s., 2018. improvement of emission reduction in nano additive simarouba glauca biodiesel blends. energy sources, part a recover. util. environ. eff. 40, 1929–1934. https://doi.org/10.1080/15567036.2018.1488900 kannan, g.r., karvembu, r., anand, r., 2011. effect of metal based additive on performance emission and combustion characteristics of diesel engine fuelled with biodiesel. appl. energy 88, 3694–3703. https://doi.org/10.1016/j.apenergy.2011.04.043 karmakar, a., karmakar, s., mukherjee, s., 2010. properties of various plants and animals feedstocks for biodiesel production. bioresour. technol. 101, 7201–7210. https://doi.org/10.1016/j.biortech.2010.04.079 karthikeyan, s., kalaimurugan, k., prathima, a., 2017. investigation on the emission quality characteristics of a diesel engine fueled with algae biofuel with nano additives. energy sources, part a recover. util. environ. eff. 39, 2046–2052. https://doi.org/10.1080/15567036.2017.1349216 karthikeyan, s., periyasamy, m., prathima, a., 2020. combustion analysis of a ci engine with caulerpa racemosa algae biofuel with nano additives. mater. today proc. 33, 3324–3329. https://doi.org/10.1016/j.matpr.2020.04.780 karthikeyan, s., prathima, a., 2016. environmental effect on the impact of ferrofluid on caulerpa racemosa oil methyl ester from marine macroalgae. energy sources, part a recover. util. environ. eff. 38, 3242–3248. kathirvelu, b., subramanian, s., govindan, n., santhanam, s., 2017. emission characteristics of biodiesel obtained from jatropha seeds and fish wastes in a diesel engine. sustain. environ. res. 27, 283–290. https://doi.org/10.1016/j.serj.2017.06.004 kaya, c., kökkülünk, g., 2020. biodiesel as alternative additive fuel for diesel engines: an experimental and theoretical investigation on emissions and performance characteristics. energy sources, part a recover. util. environ. eff. 1–23. https://doi.org/10.1080/15567036.2020.1774685 kegl, b., 2011. influence of biodiesel on engine combustion and emission characteristics. appl. energy 88, 1803–1812. https://doi.org/10.1016/j.apenergy.2010.12.007 kerihuel, kumar, m.s., bellettre, j., tazerout, m., 2005. investigations on a ci engine using animal fat and its emulsions with water and methanol as fuel. sae tech. pap. https://doi.org/10.4271/2005-01-1729 khalife, e., tabatabaei, m., demirbas, a., aghbashlo, m., 2017. impacts of additives on performance and emission characteristics of diesel engines during steady state operation. prog. energy combust. sci. https://doi.org/10.1016/j.pecs.2016.10.001 kharina, a., malins, c., beijing, s.s., 2016. biofuels policy in indonesia: overview and status report. kim, d.-s., hanifzadeh, m., kumar, a., 2018. trend of biodiesel feedstock and its impact on biodiesel emission characteristics. environ. prog. sustain. energy 37, 7–19. https://doi.org/10.1002/ep.12800 kolakoti, a., setiyo, m., rochman, m.l., 2022. a green heterogeneous catalyst production and characterization for biodiesel production using rsm and ann approach. int. j. renew. energy dev. 11, 703–712. kumar, k., sharma, m.p., 2016. performance and emission characteristics of a diesel engine fuelled with biodiesel blends. int. j. renew. energy res. 6, 658–662. https://doi.org/10.20508/ijrer.v6i2.3827.g6831 kumar, l.r., yellapu, s.k., tyagi, r.d., drogui, p., 2021. biodiesel production from microbial lipid obtained by intermittent feeding of municipal sludge and treated crude glycerol. syst. microbiol. biomanufacturing 1, 344–355. https://doi.org/10.1007/s43393021-00030-2 kumar, n., kumar, v. and a., 2010. biodiesel as an alternative fuel for ci engines: environmental effect. indian j. sci. technol. 3, 602– 606. https://doi.org/10.17485/ijst/2010/v3i5.23 kumar, n., varun, chauhan, s.r., 2013. performance and emission characteristics of biodiesel from different origins: a review. renew. sustain. energy rev. 21, 633–658. https://doi.org/10.1016/j.rser.2013.01.006 kumar, s., dinesha, p., rosen, m.a., 2019. effect of injection pressure on the combustion, performance and emission characteristics of a biodiesel engine with cerium oxide nanoparticle additive. energy 185, 1163–1173. lamas, m.i., c.g., r., j., t., j.d., r., 2015. numerical analysis of emissions from marine engines using alternative fuels. polish marit. res. 22, 48–52. https://doi.org/10.1515/pomr-2015-0070 lamb, w.f., wiedmann, t., pongratz, j., andrew, r., crippa, m., olivier, j.g.j., wiedenhofer, d., mattioli, g., khourdajie, a. al, house, j., pachauri, s., figueroa, m., saheb, y., slade, r., hubacek, k., sun, l., ribeiro, s.k., khennas, s., de la rue du can, s., chapungu, l., davis, s.j., bashmakov, i., dai, h., dhakal, s., tan, x., geng, y., gu, b., minx, j., 2021. a review of trends and drivers of greenhouse gas emissions by sector from 1990 to 2018. environ. res. lett. 16, 073005. https://doi.org/10.1088/17489326/abee4e lapuerta, m., armas, o., rodríguez-fernández, j., 2008. effect of biodiesel fuels on diesel engine emissions. prog. energy combust. sci. https://doi.org/10.1016/j.pecs.2007.07.001 leedham, a., caprotti, r., graupner, o., klaua, t., 2004. impact of fuel additives on diesel injector deposits. sae 2004-01-2935. liaquat, a.m., masjuki, h.h., kalam, m.a., rizwanul fattah, i.m., 2014. impact of biodiesel blend on injector deposit formation. energy 72, 813–823. https://doi.org/10.1016/j.energy.2014.06.006 long, a., bose, a., o’shea, r., monaghan, r., murphy, j.d., 2021. implications of european union recast renewable energy directive sustainability criteria for renewable heat and transport: case study of willow biomethane in ireland. renew. sustain. energy rev. 150, 111461. https://doi.org/10.1016/j.rser.2021.111461 lopes, j.c.a., boros, l., kráhenbúhl, m.a., meirelles, a.j.a., daridon, j.l., pauly, j., marrucho, i.m., coutinho, j.a.p., 2008. prediction of cloud points of biodiesel. energy and fuels 22, 747–752. https://doi.org/10.1021/ef700436d/asset/images/larg e/ef-2007-00436d_0011.jpeg lue, y.f., yeh, y.y., wu, c.h., 2001. the emission characteristics of a small d.i. diesel engine using biodiesel blended fuels. j. environ. sci. heal. part a toxic/hazardous subst. environ. eng. 36, 845– 859. https://doi.org/10.1081/ese-100103765 manigandan, s., gunasekar, p., nithya, s., devipriya, j., 2020. effects of nanoadditives on emission characteristics of engine fuelled with biodiesel. energy sources, part a recover. util. environ. eff. 42, 1– 9. https://doi.org/10.1080/15567036.2019.1587048 maroušek, j., gavurová, b., strunecký, o., maroušková, a., sekar, m., marek, v., 2023a. techno-economic identification of production factors threatening the competitiveness of algae biodiesel. fuel 344. https://doi.org/10.1016/j.fuel.2023.128056 maroušek, j., maroušková, a., gavurová, b., tuček, d., strunecký, o., 2023b. competitive algae biodiesel depends on advances in mass algae cultivation. bioresour. technol. 374, 128802. https://doi.org/10.1016/j.biortech.2023.128802 martins, f., felgueiras, c., smitková, m., 2018. fossil fuel energy consumption in european countries. energy procedia 153, 107– 111. https://doi.org/10.1016/j.egypro.2018.10.050 mehta, r.n., chakraborty, m., mahanta, p., parikh, p.a., 2010. evaluation of fuel properties of butanol-biodiesel-diesel blends and their impact on engine performance and emissions. ind. eng. chem. res. 49, 7660–7665. https://doi.org/10.1021/ie1006257/asset/images/large /ie-2010-006257_0005.jpeg meira, m., quintella, c.m., ribeiro, e.m.o., silva, h.r.g., guimarães, a.k., 2015. overview of the challenges in the production of biodiesel. biomass convers. biorefinery 5, 321–329. https://doi.org/10.1007/s13399-014-0146-2 mejía, j.d., salgado, n., orrego, c.e., 2013. effect of blends of diesel and palm-castor biodiesels on viscosity, cloud point and flash point. ind. crops prod. 43, 791–797. https://doi.org/10.1016/j.indcrop.2012.08.026 https://doi.org/10.1016/j.applthermaleng.2009.10.004 https://doi.org/10.1080/15567036.2019.1679290 https://doi.org/10.1080/15567036.2018.1488900 https://doi.org/10.1016/j.apenergy.2011.04.043 https://doi.org/10.1016/j.biortech.2010.04.079 https://doi.org/10.1080/15567036.2017.1349216 https://doi.org/10.1016/j.matpr.2020.04.780 https://doi.org/10.1016/j.serj.2017.06.004 https://doi.org/10.1080/15567036.2020.1774685 https://doi.org/10.1016/j.apenergy.2010.12.007 https://doi.org/10.4271/2005-01-1729 https://doi.org/10.1016/j.pecs.2016.10.001 https://doi.org/10.1002/ep.12800 https://doi.org/10.20508/ijrer.v6i2.3827.g6831 https://doi.org/10.1007/s43393-021-00030-2 https://doi.org/10.1007/s43393-021-00030-2 https://doi.org/10.17485/ijst/2010/v3i5.23 https://doi.org/10.1016/j.rser.2013.01.006 https://doi.org/10.1515/pomr-2015-0070 https://doi.org/10.1088/1748-9326/abee4e https://doi.org/10.1088/1748-9326/abee4e https://doi.org/10.1016/j.pecs.2007.07.001 https://doi.org/10.1016/j.energy.2014.06.006 https://doi.org/10.1016/j.rser.2021.111461 https://doi.org/10.1021/ef700436d/asset/images/large/ef-2007-00436d_0011.jpeg https://doi.org/10.1021/ef700436d/asset/images/large/ef-2007-00436d_0011.jpeg https://doi.org/10.1081/ese-100103765 https://doi.org/10.1080/15567036.2019.1587048 https://doi.org/10.1016/j.fuel.2023.128056 https://doi.org/10.1016/j.biortech.2023.128802 https://doi.org/10.1016/j.egypro.2018.10.050 https://doi.org/10.1021/ie1006257/asset/images/large/ie-2010-006257_0005.jpeg https://doi.org/10.1021/ie1006257/asset/images/large/ie-2010-006257_0005.jpeg https://doi.org/10.1007/s13399-014-0146-2 https://doi.org/10.1016/j.indcrop.2012.08.026 v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 737 issn: 2252-4940/© 2023. the author(s). published by cbiore mirhashemi, f.s., sadrnia, h., 2020. nox emissions of compression ignition engines fueled with various biodiesel blends: a review. j. energy inst. 93, 129–151. https://doi.org/10.1016/j.joei.2019.04.003 mofijur, m., masjuki, h.h., kalam, m.a., atabani, a.e., 2013. evaluation of biodiesel blending, engine performance and emissions characteristics of jatropha curcas methyl ester: malaysian perspective. energy 55, 879–887. https://doi.org/10.1016/j.energy.2013.02.059 mofijur, m., rasul, m., hassan, n.m.s., uddin, m.n., 2019. investigation of exhaust emissions from a stationary diesel engine fuelled with biodiesel. energy procedia 160, 791–797. https://doi.org/10.1016/j.egypro.2019.02.159 mofijur, m., rasul, m.g., hyde, j., azad, a.k., mamat, r., bhuiya, m.m.k., 2016. role of biofuel and their binary (diesel–biodiesel) and ternary (ethanol–biodiesel–diesel) blends on internal combustion engines emission reduction. renew. sustain. energy rev. 53, 265–278. https://doi.org/10.1016/j.rser.2015.08.046 mohamed, m., tan, c.-k., fouda, a., gad, m.s., abu-elyazeed, o., hashem, a.-f., 2020. diesel engine performance, emissions and combustion characteristics of biodiesel and its blends derived from catalytic pyrolysis of waste cooking oil. energies 13, 5708. https://doi.org/10.3390/en13215708 more, g.v., koli, s.r., rao, y.v.h., prasad, p.i., rao, b.n., 2020. effect of compression ratio on compression ignition engine with ruco biodiesel/ diethyl ether/ diesel fuel blends. energy sources, part a recover. util. environ. eff. 1–20. https://doi.org/10.1080/15567036.2020.1785593 mulyono, a.b., sugiarto, b., suryantoro, m.t., setiapraja, h., yubaidah, s., attharik, m.i., ariestiawan, m.r., cohen, a., 2018. effect of hydrotreating in biodiesel on the growth of deposits in the combustion chamber as a solution for the deposits reduction in the usage of biodiesel. e3s web conf. 67, 02014. https://doi.org/10.1051/e3sconf/20186702014 murugesan, p., hoang, a.t., perumal venkatesan, e., santosh kumar, d., balasubramanian, d., le, a.t., pham, v.v., 2022. role of hydrogen in improving performance and emission characteristics of homogeneous charge compression ignition engine fueled with graphite oxide nanoparticle-added microalgae biodiesel/diesel blends. int. j. hydrogen energy 47, 37617–37634. https://doi.org/10.1016/j.ijhydene.2021.08.107 n, s., afzal, a., v, s.h., ağbulut, ü., alahmadi, a.a., gowda, a.c., alwetaishi, m., shaik, s., hoang, a.t., 2023. poultry fat biodiesel as a fuel substitute in diesel-ethanol blends for di-ci engine: experimental, modeling and optimization. energy 270, 126826. https://doi.org/10.1016/j.energy.2023.126826 nagaraja, s., sakthivel, m., sudhakaran, r., 2012. comparative study of the combustion, performance, and emission characteristics of a variable compression ratio engine fuelled with diesel, corn oil methyl ester, and palm oil methyl ester. j. renew. sustain. energy 4, 063122. https://doi.org/10.1063/1.4768543 nagarajan, j., balasubramanian, d., khalife, e., usman, k.m., 2022. optimization of compression ignition engine fuelled with cotton seed biodiesel using diglyme and injection pressure. j. technol. innov. 2, 52–61. https://doi.org/10.26480/jtin.02.2022.52.61 nanthagopal, k., ashok, b., tamilarasu, a., johny, a., mohan, a., 2017. influence on the effect of zinc oxide and titanium dioxide nanoparticles as an additive with calophyllum inophyllum methyl ester in a ci engine. energy convers. manag. 146, 8–19. https://doi.org/10.1016/j.enconman.2017.05.021 nayak, s.k., hoang, a.t., nayak, b., mishra, p.c., 2021. influence of fish oil and waste cooking oil as post mixed binary biodiesel blends on performance improvement and emission reduction in diesel engine. fuel 289, 119948. https://doi.org/10.1016/j.fuel.2020.119948 nguyen, h.p., bui, v.d., 2021. sustainable development of vietnam’s transportation from analysis of car freight management. int. j. knowledge-based dev. 12, 77–96. https://doi.org/10.1504/ijkbd.2021.121707 nguyen, x.p., vu, h.n., 2019. corrosion of the metal parts of diesel engines in biodiesel-based fuels. int. j. renew. energy dev. 8, 119– 132. https://doi.org/10.14710/ijred.8.2.119-132 nursyairah, j., lau, h.l.n., jalal, r.i.a., loh, s.k., 2022. effect of palm biodiesel blends on cold start performance and emissions of common rail turbocharged engine at moderately cold ambient temperatures. environ. prog. sustain. energy 1–9. https://doi.org/10.1002/ep.14037 oil reserves by country 2022 [www document], n.d. ozcanli, m., gungor, c., aydin, k., 2013. biodiesel fuel specifications: a review. energy sources, part a recover. util. environ. eff. 35, 635– 647. https://doi.org/10.1080/15567036.2010.503229 özener, o., yüksek, l., ergenç, a.t., özkan, m., 2014. effects of soybean biodiesel on a di diesel engine performance, emission and combustion characteristics. fuel 115, 875–883. https://doi.org/10.1016/j.fuel.2012.10.081 padhee, d., raheman, h., 2015. performance, emissions and combustion characteristics of a single cylinder diesel engine fuelled with blends of jatropha methyl ester and diesel. int. j. renew. energy dev. 3, 125–131. https://doi.org/10.14710/ijred.3.2.125-131 perera, f., nadeau, k., 2022. climate change, fossil-fuel pollution, and children’s health. n. engl. j. med. 386, 2303–2314. https://doi.org/10.1056/nejmra2117706 perumal, v., ilangkumaran, m., 2018. water emulsified hybrid pongamia biodiesel as a modified fuel for the experimental analysis of performance, combustion and emission characteristics of a direct injection diesel engine. renew. energy 121, 623–631. https://doi.org/10.1016/j.renene.2018.01.060 peters, g.p., andrew, r.m., canadell, j.g., fuss, s., jackson, r.b., korsbakken, j.i., le quéré, c., nakicenovic, n., 2017. key indicators to track current progress and future ambition of the paris agreement. nat. clim. chang. 7, 118–122. https://doi.org/10.1038/nclimate3202 pham, m.t., cao, d.n., 2023. research on numerical simulation of pcci engine: a review. j. technol. innov. 3, 38–45. https://doi.org/10.26480/jtin.02.2023.38.45 pham, m.t., pham, v.t., cao, d.n., 2023. design and fabrication of heating device for vegetable oil used for diesel engines. j. technol. innov. 3, 29–37. https://doi.org/10.26480/jtin.01.2023.29.37 plantinga, a., scholtens, b., 2021. the financial impact of fossil fuel divestment. clim. policy 21, 107–119. https://doi.org/10.1080/14693062.2020.1806020 prabhu, c., navaneetha krishnan, b., prakash, t., rajasekar, v., balasubramanian, d., le, v.v., linh le, n.v., phong nguyen, p.q., nguyen, v.n., 2023. biodiesel unsaturation and the synergic effects of hydrogen sharing rate on the characteristics of a compression ignition engine in dual-fuel mode. fuel 334, 126699. https://doi.org/10.1016/j.fuel.2022.126699 pradeep, p., senthilkumar, m., 2021. simultaneous reduction of emissions as well as fuel consumption in ci engine using water and nanoparticles in diesel-biodiesel blend. energy sources, part a recover. util. environ. eff. 43, 1500–1510. praveen, a., lakshmi narayana rao, g., balakrishna, b., 2018. performance and emission characteristics of a diesel engine using calophyllum inophyllum biodiesel blends with tio2 nanoadditives and egr. egypt. j. pet. 27, 731–738. https://doi.org/10.1016/j.ejpe.2017.10.008 pullagura, g., vadapalli, s., v. v. s., p., rao chebattina, k.r., 2023. effect of dispersant added graphene nanoplatelets with diesel– sterculia foetida seed oil biodiesel blends on diesel engine: engine combustion, performance and exhaust emissions. biofuels 14, 461–472. https://doi.org/10.1080/17597269.2022.2148876 qi, d.h., chen, h., geng, l.m., bian, y.z., 2010. experimental studies on the combustion characteristics and performance of a direct injection engine fueled with biodiesel/diesel blends. energy convers. manag. 51, 2985–2992. https://doi.org/10.1016/j.enconman.2010.06.042 rajamohan, s., hari gopal, a., muralidharan, k.r., huang, z., paramasivam, b., ayyasamy, t., nguyen, x.p., le, a.t., hoang, a.t., 2022. evaluation of oxidation stability and engine behaviors operated by prosopis juliflora biodiesel/diesel fuel blends with presence of synthetic antioxidant. sustain. energy technol. assessments 52, 102086. https://doi.org/10.1016/j.seta.2022.102086 ramakrishnan, g., krishnan, p., rathinam, s., thiyagu, r., devarajan, y., 2019. role of nano-additive blended biodiesel on emission characteristics of the research diesel engine. int. j. green energy 16, 435–441. https://doi.org/10.1080/15435075.2019.1577742 https://doi.org/10.1016/j.joei.2019.04.003 https://doi.org/10.1016/j.energy.2013.02.059 https://doi.org/10.1016/j.egypro.2019.02.159 https://doi.org/10.1016/j.rser.2015.08.046 https://doi.org/10.3390/en13215708 https://doi.org/10.1080/15567036.2020.1785593 https://doi.org/10.1051/e3sconf/20186702014 https://doi.org/10.1016/j.ijhydene.2021.08.107 https://doi.org/10.1016/j.energy.2023.126826 https://doi.org/10.1063/1.4768543 https://doi.org/10.26480/jtin.02.2022.52.61 https://doi.org/10.1016/j.enconman.2017.05.021 https://doi.org/10.1016/j.fuel.2020.119948 https://doi.org/10.1504/ijkbd.2021.121707 https://doi.org/10.14710/ijred.8.2.119-132 https://doi.org/10.1002/ep.14037 https://doi.org/10.1080/15567036.2010.503229 https://doi.org/10.1016/j.fuel.2012.10.081 https://doi.org/10.14710/ijred.3.2.125-131 https://doi.org/10.1056/nejmra2117706 https://doi.org/10.1016/j.renene.2018.01.060 https://doi.org/10.1038/nclimate3202 https://doi.org/10.26480/jtin.02.2023.38.45 https://doi.org/10.26480/jtin.01.2023.29.37 https://doi.org/10.1080/14693062.2020.1806020 https://doi.org/10.1016/j.fuel.2022.126699 https://doi.org/10.1016/j.ejpe.2017.10.008 https://doi.org/10.1080/17597269.2022.2148876 https://doi.org/10.1016/j.enconman.2010.06.042 https://doi.org/10.1016/j.seta.2022.102086 https://doi.org/10.1080/15435075.2019.1577742 v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 738 issn: 2252-4940/© 2023. the author(s). published by cbiore ramalingam, s., mahalakshmi, n. v, 2020. influence of moringa oleifera biodiesel–diesel–hexanol and biodiesel–diesel–ethanol blends on compression ignition engine performance, combustion and emission characteristics. rsc adv. 10, 4274–4285. raman, l.a., deepanraj, b., rajakumar, s., sivasubramanian, v., 2019. experimental investigation on performance, combustion and emission analysis of a direct injection diesel engine fuelled with rapeseed oil biodiesel. fuel 246, 69–74. https://doi.org/10.1016/j.fuel.2019.02.106 ramesh, d.k., dhananjaya kumar, j.l., hemanth kumar, s., namith, v., basappa jambagi, p., sharath, s., 2018. study on effects of alumina nanoparticles as additive with poultry litter biodiesel on performance, combustion and emission characteristic of diesel engine. mater. today proc. 5, 1114–1120. https://doi.org/10.1016/j.matpr.2017.11.190 rameshbabu, a., senthilkumar, g., 2021. emission and performance investigation on the effect of nano-additive on neat biodiesel. energy sources, part a recover. util. environ. eff. 43, 1315–1328. rao, g.l.n., prasad, b.d., sampath, s., rajagopal, k., 2007. combustion analysis of diesel engine fueled with jatropha oil methy lesterdiesel blends. int. j. green energy 4, 645–658. https://doi.org/10.1080/15435070701665446 rathinam, s., justin abraham baby, s., devarajan, y., t, a., 2018. influence of water on exhaust emissions on unmodified diesel engine propelled with biodiesel. energy sources, part a recover. util. environ. eff. 40, 2511–2517. https://doi.org/10.1080/15567036.2018.1503756 reyes, j.f., sepúlveda, m.a., 2006. pm-10 emissions and power of a diesel engine fueled with crude and refined biodiesel from salmon oil. fuel 85, 1714–1719. https://doi.org/10.1016/j.fuel.2006.02.001 riyadi, t.w.b., spraggon, m., herawan, s.g., idris, m., paristiawan, p.a., putra, n.r., r, m.f., silambarasan, r., veza, i., 2023. biodiesel for hcci engine: prospects and challenges of sustainability biodiesel for energy transition. results eng. 17, 100916. https://doi.org/10.1016/j.rineng.2023.100916 rochelle, d., najafi, h., 2019. a review of the effect of biodiesel on gas turbine emissions and performance. renew. sustain. energy rev. 105, 129–137. https://doi.org/10.1016/j.rser.2019.01.056 rosha, p., mohapatra, s.k., mahla, s.k., cho, h., chauhan, b.s., dhir, a., 2019. effect of compression ratio on combustion, performance, and emission characteristics of compression ignition engine fueled with palm (b20) biodiesel blend. energy 178, 676–684. https://doi.org/10.1016/j.energy.2019.04.185 rozina, chia, s.r., ahmad, m., sultana, s., zafar, m., asif, s., bokhari, a., nomanbhay, s., mubashir, m., khoo, k.s., show, p.l., 2022. green synthesis of biodiesel from citrus medica seed oil using green nanoparticles of copper oxide. fuel 323, 124285. https://doi.org/10.1016/j.fuel.2022.124285 rudzki, k., gomulka, p., hoang, a.t., 2022. optimization model to manage ship fuel consumption and navigation time. polish marit. res. 29, 141–153. https://doi.org/10.2478/pomr-20220034 sajith, v., sobhan, c.b., peterson, g.p., 2010. experimental investigations on the effects of cerium oxide nanoparticle fuel additives on biodiesel. adv. mech. eng. https://doi.org/10.1155/2010/581407 sakthivel, g., nagarajan, g., ilangkumaran, m., gaikwad, a.b., 2014. comparative analysis of performance, emission and combustion parameters of diesel engine fuelled with ethyl ester of fish oil and its diesel blends. fuel 132, 116–124. https://doi.org/10.1016/j.fuel.2014.04.059 sakthivel, r., ramesh, k., purnachandran, r., mohamed shameer, p., 2018. a review on the properties, performance and emission aspects of the third generation biodiesels. renew. sustain. energy rev. 82, 2970–2992. https://doi.org/10.1016/j.rser.2017.10.037 sani, s., kaisan, m.u., kulla, d.m., obi, a.i., jibrin, a., ashok, b., 2018. determination of physico chemical properties of biodiesel from citrullus lanatus seeds oil and diesel blends. ind. crops prod. 122, 702–708. https://doi.org/10.1016/j.indcrop.2018.06.002 saravanan, s., nagarajan, g., sampath, s., 2014. a correlation for the ignition delay of a ci engine fuelled with diesel and biodiesel. int. j. green energy 11, 542–557. https://doi.org/10.1080/15435075.2013.777906 sarin, a., 2012. biodiesel: production and properties. royal society of chemistry. sathish, t., ağbulut, ü., george, s.m., ramesh, k., saravanan, r., roberts, k.l., sharma, p., asif, m., hoang, a.t., 2023. waste to fuel: synergetic effect of hybrid nanoparticle usage for the improvement of ci engine characteristics fuelled with waste fish oils. energy 275, 127397. https://doi.org/10.1016/j.energy.2023.127397 semwal, s., raj, t., patel, a.k., arora, a.k., badoni, r.p., singhania, r.r., 2022. synthesis of ca–fe-based heterogeneous catalyst from waste shells and their application for transesterification of jatropha oil. syst. microbiol. biomanufacturing. https://doi.org/10.1007/s43393-022-00123-6 seraç, m.r., aydın, s., yılmaz, a., şevik, s., 2020. evaluation of comparative combustion, performance, and emission of soybean-based alternative biodiesel fuel blends in a ci engine. renew. energy 148, 1065–1073. https://doi.org/10.1016/j.renene.2019.10.090 serbin, s., diasamidze, b., gorbov, v., kowalski, j., 2021. investigations of the emission characteristics of a dual-fuel gas turbine combustion chamber operating simultaneously on liquid and gaseous fuels. polish marit. res. 28, 85–95. https://doi.org/10.2478/pomr-2021-0025 shaafi, t., sairam, k., gopinath, a., kumaresan, g., velraj, r., 2015. effect of dispersion of various nanoadditives on the performance and emission characteristics of a ci engine fuelled with diesel, biodiesel and blends—a review. renew. sustain. energy rev. 49, 563–573. shafiee, s., topal, e., 2009. when will fossil fuel reserves be diminished? energy policy 37, 181–189. https://doi.org/10.1016/j.enpol.2008.08.016 shah, p.r., 2015. study the effects of rice bran oil methyl ester on performance and emission characteristics of agriculture diesel engine. int. res. j. eng. technol. shahabuddin, m., liaquat, a.m., masjuki, h.h., kalam, m.a., mofijur, m., 2013. ignition delay, combustion and emission characteristics of diesel engine fueled with biodiesel. renew. sustain. energy rev. 21, 623–632. https://doi.org/10.1016/j.rser.2013.01.019 shaisundaram, v.s., chandrasekaran, m., shanmugam, m., padmanabhan, s., muraliraja, r., karikalan, l., 2021. investigation of momordica charantia seed biodiesel with cerium oxide nanoparticle on ci engine. int. j. ambient energy 42, 1615– 1619. https://doi.org/10.1080/01430750.2019.1611657 sharma, p., balasubramanian, d., thanh khai, c., papla venugopal, i., alruqi, m., josephin js, f., sonthalia, a., geo varuvel, e., khalife, e., ravikumar, r., wae-hayee, m., 2023. enhancing the performance of renewable biogas powered engine employing oxyhydrogen: optimization with desirability and d-optimal design. fuel 341, 127575. https://doi.org/10.1016/j.fuel.2023.127575 sharma, p., chhillar, a., said, z., huang, z., nguyen, v.n., nguyen, p.q.p., nguyen, x.p., 2022. experimental investigations on efficiency and instability of combustion process in a diesel engine fueled with ternary blends of hydrogen peroxide additive/biodiesel/diesel. energy sources, part a recover. util. environ. eff. 44, 5929–5950. https://doi.org/10.1080/15567036.2022.2091692 shrivastava, p., verma, t.n., pugazhendhi, a., 2019. an experimental evaluation of engine performance and emisssion characteristics of ci engine operated with roselle and karanja biodiesel. fuel 254, 115652. https://doi.org/10.1016/j.fuel.2019.115652 silitonga, a.s., hassan, m.h., ong, h.c., kusumo, f., 2017. analysis of the performance, emission and combustion characteristics of a turbocharged diesel engine fuelled with jatropha curcas biodiesel-diesel blends using kernel-based extreme learning machine. environ. sci. pollut. res. 24, 25383–25405. https://doi.org/10.1007/s11356-017-0141-9/figures/13 silviana, s., anggoro, d.d., hadiyanto, h., salsabila, c.a., aprilio, k., utami, a.w., sa’adah, a.n., dalanta, f., 2022. a review on the recent breakthrough methods and influential parameters in the biodiesel synthesis and purification. int. j. renew. energy dev. 11, 1012–1036. https://doi.org/10.14710/ijred.2022.43147 singh, d., sharma, d., soni, s.l., inda, c.s., sharma, s., sharma, p.k., jhalani, a., 2021. a comprehensive review of physicochemical properties, production process, performance and emissions https://doi.org/10.1016/j.fuel.2019.02.106 https://doi.org/10.1016/j.matpr.2017.11.190 https://doi.org/10.1080/15435070701665446 https://doi.org/10.1080/15567036.2018.1503756 https://doi.org/10.1016/j.fuel.2006.02.001 https://doi.org/10.1016/j.rineng.2023.100916 https://doi.org/10.1016/j.rser.2019.01.056 https://doi.org/10.1016/j.energy.2019.04.185 https://doi.org/10.1016/j.fuel.2022.124285 https://doi.org/10.2478/pomr-2022-0034 https://doi.org/10.2478/pomr-2022-0034 https://doi.org/10.1155/2010/581407 https://doi.org/10.1016/j.fuel.2014.04.059 https://doi.org/10.1016/j.rser.2017.10.037 https://doi.org/10.1016/j.indcrop.2018.06.002 https://doi.org/10.1080/15435075.2013.777906 https://doi.org/10.1016/j.energy.2023.127397 https://doi.org/10.1007/s43393-022-00123-6 https://doi.org/10.1016/j.renene.2019.10.090 https://doi.org/10.2478/pomr-2021-0025 https://doi.org/10.1016/j.enpol.2008.08.016 https://doi.org/10.1016/j.rser.2013.01.019 https://doi.org/10.1080/01430750.2019.1611657 https://doi.org/10.1016/j.fuel.2023.127575 https://doi.org/10.1080/15567036.2022.2091692 https://doi.org/10.1016/j.fuel.2019.115652 https://doi.org/10.1007/s11356-017-0141-9/figures/13 https://doi.org/10.14710/ijred.2022.43147 v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 739 issn: 2252-4940/© 2023. the author(s). published by cbiore characteristics of 2nd generation biodiesel feedstock: jatropha curcas. fuel. https://doi.org/10.1016/j.fuel.2020.119110 singh, d., sharma, d., soni, s.l., sharma, s., kumar sharma, p., jhalani, a., 2020. a review on feedstocks, production processes, and yield for different generations of biodiesel. fuel 262, 116553. https://doi.org/10.1016/j.fuel.2019.116553 singh, d., sharma, d., soni, s.l., sharma, s., kumari, d., 2019. chemical compositions, properties, and standards for different generation biodiesels: a review. fuel 253, 60–71. https://doi.org/10.1016/j.fuel.2019.04.174 sinha, d., murugavelh, s., 2016. biodiesel production from waste cotton seed oil using low cost catalyst: engine performance and emission characteristics. perspect. sci. 8, 237–240. https://doi.org/10.1016/j.pisc.2016.04.038 sivaramakrishnan, k., ravikumar, p., 2014. optimization of operational parameters on performance and emissions of a diesel engine using biodiesel. int. j. environ. sci. technol. 11, 949–958. https://doi.org/10.1007/s13762-013-0273-5 speight, j.g., 2011. an introduction to petroleum technology, economics, and politics. john wiley & sons. stelmasiak, z., larisch, j., pielecha, j., pietras, d., 2017. particulate matter emission from dual fuel diesel engine fuelled with natural gas. polish marit. res. 24, 96–104. https://doi.org/10.1515/pomr-2017-0055 su, b., wang, l., xue, y., yan, j., dong, z., lin, h., han, s., 2021. effect of pour point depressants combined with dispersants on the cold flow properties of biodiesel‐diesel blends. j. am. oil chem. soc. 98, 163–172. https://doi.org/10.1002/aocs.12456 sung, y.j., lee, j.s., yoon, h.k., ko, h., sim, s.j., 2021. outdoor cultivation of microalgae in a coal-fired power plant for conversion of flue gas co2 into microalgal direct combustion fuels. syst. microbiol. biomanufacturing 1, 90–99. sunil, s., chandra prasad, b.s., kakkeri, s., suresha, 2021. studies on titanium oxide nanoparticles as fuel additive for improving performance and combustion parameters of ci engine fueled with biodiesel blends. mater. today proc. 44, 489–499. https://doi.org/10.1016/j.matpr.2020.10.200 suresh, m., jawahar, c.p., richard, a., 2018. a review on biodiesel production, combustion, performance, and emission characteristics of non-edible oils in variable compression ratio diesel engine using biodiesel and its blends. renew. sustain. energy rev. 92, 38–49. tamilselvan, p., nallusamy, n., rajkumar, s., 2017. a comprehensive review on performance, combustion and emission characteristics of biodiesel fuelled diesel engines. renew. sustain. energy rev. 79, 1134–1159. https://doi.org/10.1016/j.rser.2017.05.176 temizer, i̇., cihan, ö., eskici, b., 2020. numerical and experimental investigation of the effect of biodiesel/diesel fuel on combustion characteristics in ci engine. fuel 270, 117523. https://doi.org/10.1016/j.fuel.2020.117523 tesfa, b., mishra, r., gu, f., powles, n., 2010. prediction models for density and viscosity of biodiesel and their effects on fuel supply system in ci engines. renew. energy 35, 2752–2760. https://doi.org/10.1016/j.renene.2010.04.026 thangavelu, s.k., ahmed, a.s., ani, f.n., 2016. impact of metals on corrosive behavior of biodiesel–diesel–ethanol (bde) alternative fuel. renew. energy 94, 1–9. https://doi.org/10.1016/j.renene.2016.03.015 tomar, m., kumar, n., 2020. influence of nanoadditives on the performance and emission characteristics of a ci engine fuelled with diesel, biodiesel, and blends–a review. energy sources, part a recover. util. environ. eff. https://doi.org/10.1080/15567036.2019.1623347 truong, t.t., nguyen, x.p., pham, v.v., le, v.v., le, a.t., bui, v.t., 2021. effect of alcohol additives on diesel engine performance: a review. energy sources, part a recover. util. environ. eff. 1–25. https://doi.org/10.1080/15567036.2021.2011490 tuan hoang, a., nižetić, s., chyuan ong, h., tarelko, w., viet pham, v., hieu le, t., quang chau, m., phuong nguyen, x., 2021. a review on application of artificial neural network (ann) for performance and emission characteristics of diesel engine fueled with biodiesel-based fuels. sustain. energy technol. assessments 47, 101416. https://doi.org/10.1016/j.seta.2021.101416 uyumaz, a., 2018. combustion, performance and emission characteristics of a di diesel engine fueled with mustard oil biodiesel fuel blends at different engine loads. fuel 212, 256–267. https://doi.org/10.1016/j.fuel.2017.09.005 uyumaz, a., solmaz, h., yilmaz, e., yamik, h., polat, s., 2014. experimental examination of the effects of military aviation fuel jp-8 and biodiesel fuel blends on the engine performance, exhaust emissions and combustion in a direct injection engine. fuel process. technol. 128, 158–165. https://doi.org/10.1016/j.fuproc.2014.07.013 vali, r.h., hoang, a.t., wani, m.m., pali, h.s., balasubramanian, d., arıcı, m., said, z., xuan phuong nguyen, 2022. optimization of variable compression ratio diesel engine fueled with zinc oxide nanoparticles and biodiesel emulsion using response surface methodology. fuel 323, 124290. https://doi.org/10.1016/j.fuel.2022.124290 vani, m.v., basha, p.o., rajesh, n., riazunnisa, k., 2022. development of chlorella pyrenoidosa ems mutants with enhanced biomass and lipid content for biofuel production. syst. microbiol. biomanufacturing. https://doi.org/10.1007/s43393-022-00153-0 varatharajan, k., cheralathan, m., 2012. influence of fuel properties and composition on nox emissions from biodiesel powered diesel engines: a review. renew. sustain. energy rev. 16, 3702–3710. vellaiyan, s., partheeban, c.m.a., 2018. emission analysis of diesel engine fueled with soybean biodiesel and its water blends. energy sources, part a recover. util. environ. eff. 40, 1956– 1965. https://doi.org/10.1080/15567036.2018.1489911 venu, h., madhavan, v., 2017. influence of diethyl ether (dee) addition in ethanol-biodiesel-diesel (ebd) and methanol-biodiesel-diesel (mbd) blends in a diesel engine. fuel 189, 377–390. https://doi.org/10.1016/j.fuel.2016.10.101 venu, h., madhavan, v., 2016. effect of nano additives (titanium and zirconium oxides) and diethyl ether on biodiesel-ethanol fuelled ci engine. j. mech. sci. technol. 2016 305 30, 2361–2368. https://doi.org/10.1007/s12206-016-0446-5 venugopal, i.p., balasubramanian, d., rajarajan, a., 2021. potential improvement in conventional diesel combustion mode on a common rail direct injection diesel engine with pode/wco blend as a high reactive fuel to achieve effective soot-nox tradeoff. j. clean. prod. 327, 129495. https://doi.org/10.1016/j.jclepro.2021.129495 venugopal, i.p., balasubramanian, d., rajarajan, a., suresh, k., 2023. quantification of φ-operating range with impact of exhaust gas recirculation under low-temperature combustion mode with polyoxymethylene dimethyl ether: a perspective study. j. clean. prod. 411, 137298. https://doi.org/10.1016/j.jclepro.2023.137298 verma, p., sharma, m.p., 2015. performance and emission characteristics of biodiesel fuelled diesel engines. int. j. renew. energy res. 5, 245–250. https://doi.org/10.20508/ijrer.v5i1.1963.g6491 veza, i., karaoglan, a.d., ileri, e., kaulani, s.a., tamaldin, n., latiff, z.a., muhamad said, m.f., hoang, a.t., yatish, k.v., idris, m., 2022. grasshopper optimization algorithm for diesel engine fuelled with ethanol-biodiesel-diesel blends. case stud. therm. eng. 31, 101817. https://doi.org/10.1016/j.csite.2022.101817 wei, l., cheung, c.s., ning, z., 2018. effects of biodiesel-ethanol and biodiesel-butanol blends on the combustion, performance and emissions of a diesel engine. energy 155, 957–970. https://doi.org/10.1016/j.energy.2018.05.049 widayat, w., maheswari, n.t., fitriani, w., buchori, l., satriadi, h., kusmiyati, k., ngadi, n., 2023. preparation of mgo-cao/sio2 catalyst from dolomite and geothermal solid waste for biodiesel production. int. j. renew. energy dev. 12, 541–549. https://doi.org/10.14710/ijred.2023.51573 yang, z., tan, q., geng, p., 2019. combustion and emissions investigation on low-speed two-stroke marine diesel engine with low sulfur diesel fuel. polish marit. res. 26, 153–161. https://doi.org/10.2478/pomr-2019-0017 yilmaz, n., atmanli, a., 2017. experimental assessment of a diesel engine fueled with diesel-biodiesel-1-pentanol blends. fuel 191, 190–197. https://doi.org/10.1016/j.fuel.2016.11.065 yilmaz, n., ileri, e., atmanli, a., 2016. performance of biodiesel/higher alcohols blends in a diesel engine. int. j. energy res. 40, 1134– 1143. https://doi.org/10.1002/er.3513 yilmaz, n., vigil, f.m., benalil, k., davis, s.m., calva, a., 2014. effect of https://doi.org/10.1016/j.fuel.2020.119110 https://doi.org/10.1016/j.fuel.2019.116553 https://doi.org/10.1016/j.fuel.2019.04.174 https://doi.org/10.1016/j.pisc.2016.04.038 https://doi.org/10.1007/s13762-013-0273-5 https://doi.org/10.1515/pomr-2017-0055 https://doi.org/10.1002/aocs.12456 https://doi.org/10.1016/j.matpr.2020.10.200 https://doi.org/10.1016/j.rser.2017.05.176 https://doi.org/10.1016/j.fuel.2020.117523 https://doi.org/10.1016/j.renene.2010.04.026 https://doi.org/10.1016/j.renene.2016.03.015 https://doi.org/10.1080/15567036.2019.1623347 https://doi.org/10.1080/15567036.2021.2011490 https://doi.org/10.1016/j.seta.2021.101416 https://doi.org/10.1016/j.fuel.2017.09.005 https://doi.org/10.1016/j.fuproc.2014.07.013 https://doi.org/10.1016/j.fuel.2022.124290 https://doi.org/10.1007/s43393-022-00153-0 https://doi.org/10.1080/15567036.2018.1489911 https://doi.org/10.1016/j.fuel.2016.10.101 https://doi.org/10.1007/s12206-016-0446-5 https://doi.org/10.1016/j.jclepro.2021.129495 https://doi.org/10.1016/j.jclepro.2023.137298 https://doi.org/10.20508/ijrer.v5i1.1963.g6491 https://doi.org/10.1016/j.csite.2022.101817 https://doi.org/10.1016/j.energy.2018.05.049 https://doi.org/10.14710/ijred.2023.51573 https://doi.org/10.2478/pomr-2019-0017 https://doi.org/10.1016/j.fuel.2016.11.065 https://doi.org/10.1002/er.3513 v.g.nguyen et al int. j. renew. energy dev 2023, 12(4),720-740 | 740 issn: 2252-4940/© 2023. the author(s). published by cbiore biodiesel–butanol fuel blends on emissions and performance characteristics of a diesel engine. fuel 135, 46–50. https://doi.org/10.1016/j.fuel.2014.06.022 yubaidah, s., 2023. cold start ability test for diesel passenger cars using. aip conf. proc. 2646, 050014. yuvarajan, d., dinesh babu, m., beemkumar, n., amith kishore, p., 2018. experimental investigation on the influence of titanium dioxide nanofluid on emission pattern of biodiesel in a diesel engine. atmos. pollut. res. https://doi.org/10.1016/j.apr.2017.06.003 zare, a., nabi, m.n., bodisco, t.a., hossain, f.m., rahman, m.m., chu van, t., ristovski, z.d., brown, r.j., 2017. diesel engine emissions with oxygenated fuels: a comparative study into coldstart and hot-start operation. j. clean. prod. 162, 997–1008. https://doi.org/10.1016/j.jclepro.2017.06.052 zare, a., stevanovic, s., jafari, m., verma, p., babaie, m., yang, l., rahman, m.m., ristovski, z.d., brown, r.j., bodisco, t.a., 2021. analysis of cold-start no2 and nox emissions, and the no2/nox ratio in a diesel engine powered with different dieselbiodiesel blends. environ. pollut. 290, 118052. https://doi.org/10.1016/j.envpol.2021.118052 zarrinkolah, m.t., hosseini, v., 2022. detailed analysis of the effects of biodiesel fraction increase on the combustion stability and characteristics of a reactivity-controlled compression ignition diesel-biodiesel/natural gas engine. energies 2022, vol. 15, page 1094 15, 1094. https://doi.org/10.3390/en15031094 zhang, w., zhang, z., ma, x., awad, o.i., li, y., shuai, s., xu, h., 2020. impact of injector tip deposits on gasoline direct injection engine combustion, fuel economy and emissions. appl. energy 262, 114538. https://doi.org/10.1016/j.apenergy.2020.114538 zhang, z., lv, j., xie, g., wang, s., ye, y., huang, g., tan, d., 2022. effect of assisted hydrogen on combustion and emission characteristics of a diesel engine fueled with biodiesel. energy 254, 124269. https://doi.org/10.1016/j.energy.2022.124269 zullaikah, s., putra, a.k., fachrudin, f.h., naulina, r.y., utami, s., herminanto, r.p., rachmaniah, o., ju, y.h., 2021. experimental investigation and optimization of non-catalytic in-situ biodiesel production from rice bran using response surface methodology historical data design. int. j. renew. energy dev. 10, 803–810. https://doi.org/10.14710/ijred.2021.34138 zulqarnain, yusoff, m.h.m., ayoub, m., jusoh, n., abdullah, a.z., 2020. the challenges of a biodiesel implementation program in malaysia. processes 8, 1–18. https://doi.org/10.3390/pr8101244 zuorro, a., garcía-martínez, j.b., barajas-solano, a.f., 2020. the application of catalytic processes on the production of algaebased biofuels: a review. catalysts 11, 22. © 2023. the author(s). this article is an open access article distributed under the terms and conditions of the creative commons attribution-sharealike 4.0 (cc by-sa) international license (http://creativecommons.org/licenses/by-sa/4.0/) https://doi.org/10.1016/j.fuel.2014.06.022 https://doi.org/10.1016/j.apr.2017.06.003 https://doi.org/10.1016/j.jclepro.2017.06.052 https://doi.org/10.1016/j.envpol.2021.118052 https://doi.org/10.3390/en15031094 https://doi.org/10.1016/j.apenergy.2020.114538 https://doi.org/10.1016/j.energy.2022.124269 https://doi.org/10.14710/ijred.2021.34138 https://doi.org/10.3390/pr8101244 http://creativecommons.org/licenses/by-sa/4.0/