Microsoft Word - 1426 Göllei 105.docx HUNGARIAN JOURNAL OF INDUSTRY AND CHEMISTRY VESZPRÉM Vol. 42(2) pp. 85–89 (2014) MEASUREMENT-BASED MODELLING AND SIMULATION OF A HYDROGEN- GENERATING DRY CELL FOR COMPLEX DOMESTIC RENEWABLE ENERGY SYSTEMS ATTILA GÖLLEI,* PÉTER GÖRBE, ATTILA MAGYAR, AND LÁSZLÓ NEUKIRCHNER Department of Electrical Engineering and Information Systems, University of Pannonia, Egyetem u. 10., Veszprém, 8200, HUNGARY *E-mail: golleia@almos.uni-pannon.hu Nowadays, the growing need for energy from renewable sources and growing revulsion towards fossil and nuclear fuels puts sustainable and green energy in the limelight. Producing (electrical) energy in domestic power plants from renewable sources (mainly solar and wind) hardly results in difficulties, but the storage of energy not consumed immediately is a great engineering challenge. In the present paper a complex model has been developed by investigating renewable energy sources, the surplus energy not actually consumed and stored in electrical vehicle (EV) batteries, the conversion to hydrogen for storage purposes and how the main grid is fed. A measurement-based model of a hydrogen generating cell developed for the simulation of complex energetic systems. The parameter estimation of the static model was based on the collected measurement data coming from the detailed examination of a built demonstration cell. The novel element of this work is the Matlab Simulink model for the hydrogen generation cell. Using this model, a dynamic simulator of a complex domestic power plant is made available using renewable energy sources and hydrogen generation cells. Hydrogen generation enables the lossless long-term storage of surplus electric energy collected, but not consumed or injected into the low voltage grid. The generated hydrogen can be consumed for transportation purposes in suitable vehicles or it can be applied in fuel cells generating direct electrical energy for energy-deficient low voltage network situations. Energetic situations potentially occurring in practice were simulated in our complex model. Simulations showed that the presented model is suitable for domestic scale low voltage complex energetic systems. Keywords: hydrogen generation, renewable energy sources, domestic power plants, modelling and simulation, measurement-based modelling Introduction Producing hydrogen gas (H2) from excess energy is not a new idea. This is an alternative way to store and convert renewable energy for further utilization. The produced hydrogen can be stored or used in power cells to be converted back to electric energy or in vehicles for hydrogen propulsion [1]. Although the described procedure is efficient and able to produce a high quantity of H2 it is not suitable for application in combination with domestic power plants. The most relevant from of H2 production is when the energy consumption and quantity of produced H2 are controlled. When power consumption and generation are continuous (and not necessarily deterministic) functions of time, the H2 production depends solely on the excess energy of the grid. The best solution is the usage of Supervisory Control and data Acquisition Systems (SCADA) of management [2]. The domestic applicability of this technology in the future depends on the cost of SCADA system installation. Producing hydrogen and oxygen gases from water using electricity in a laboratory is a simple electrochemical process that can be performed easily and in a very demonstrative way. Producing hydrogen on a large scale or in industrial quantities calls for an optimized or near-optimized cell model. In an energy demanding process only a few percent of variance in efficiency could mean a significant energy surplus or shortage [3]. The electrochemical parameters of a dry cell (Fig.1) that are used here are discussed to simulate hydrogen and oxygen gas production. Compared to wet oxyhydrogen (HHO) cells where the entire unit is underwater, the plates of dry cells are separated with rubber seals. These seals stop the water from leaking from the cell. The electrical connections and edges of Figure 1: The theoretical setup of a dry HHO block 86 the plates do not touch the electrolyte. These parts of the unit stay dry, thus the name dry cell. To make sure the gas made from the electrolyte gets out of the cell and the solution flows between the plates, there are holes on the top (for the gas) and bottom (for the electrolyte) on the metal slats (Fig.1). The application of dry HHO units has two main advantages. The surface of the dry cell plates enables one to use smaller amounts of electrolyte compared to with wet cells; therefore, the volume and weight of the cell is smaller. Furthermore, the connectors of dry cells remain dry, i.e. they do not corrode unlike to wet cells, where the connectors are underwater therefore their surface slowly corrodes [4]. Electrochemical Foundations Electrochemical cells can be considered as galvanic batteries where the electrochemical reactions are supported by an external current supply. They are composed of two electrodes and a conductive electrolyte fluid. If the electrode material does not participate directly in the electrode reaction, it is called an indifferent electrode (e.g. graphite). During electrolysis, if there is more than one possible type of electrochemical reaction, then a simple anion will detach from the positive anode (e.g. chloride), without this anion, HO- will be created by water splitting. The dissolution voltage of water is 1.23 V at 25 °C, the temperature coefficient is -0.85 mV/K, which means that at 100 °C this voltage decreases to 1.17 V. Therefore, in the light of these data, the specific energy demand to make hydrogen via electrolysis at 25°C can be calculated from Eqs.(1–4). The amount of charge needed to evolve 1 kg of H2 gas is q = zFM = 2⋅96487⋅0.5 = 96487 A s mol-1 = = 26801 Ah kg-1 (1) wH2 = qEMF = 26801⋅1.23 = 32966 Wh kg -1 (2) Since the volume of 1 kg of standard state H2 is 12474 dm3, the amount of energy required to produce 1 dm3 of H2 gas is: wH2 = 32966 12474 = 2.64 Wh dm−3 (3) To generate 1 dm3 of hydrogen gas, 1.5 dm3 of HHO gas is needed and thus the energy demand of producing 1 ldm3 of HHO gas (0.667 dm3 H2) is: wH2(HHO) = 0.667⋅2.64 = 1.76 Wh dm -3 (4) The unit has been measured at 10 different electrolyte concentrations, using different currents. At the same time, the voltage on the plates and amount of gas produced by electrolysis has also been measured. The HHO Cell Unit The setup of one block of the unit is shown in Fig.2. Usually five cells make up one block giving one gas- producing block. The block’s electrical connections are on the ends of two plates (Fig.1). Four of the six electrode plates are neutral electrodes, as there is no voltage connected to them. The potential is divided between the neutral plates according to voltage division in series connections. It means that the voltage between two electrodes is one fifth of the voltage on one whole block. In the experiment, a unit with three blocks connected in parallel has been used. Besides the HHO cell, a water reserve tank to infuse the electrolyte into the cell was necessary. A tube between the gas outlet and the tank has also been installed since due to bubbling, electrolyte comes out of the tube that needs to be recycled back into the system. Then, as the electrolyte drips back into the tank, the gas can escape into the bottle through another hose. The produced H2 volume and the production speed are measured with this bottle. A power supply (Manson SPS9600) has been connected to the electrical connections of the HHO unit, in this way the input current was controlled (Table 1). Matlab Model of the Dry Cell The model of the dry cell considered was implemented in Matlab Simulink using the SimPowerSystems Toolbox. Two unknown functional relationships between the generated H2 volume, the cell current and the KOH concentration and between cell voltage, cell current and KOH concentration were approximated using fourth and third order polynomials, respectively Figure 2: The setup of a HHO gas generator cell block Table 1: Experimental results Electrolyte concentration, g dm-3 MMWa, cm3 min-1 W-1 Gas production, dm3 min-1 Power of unit, W 1 2.13 0.20 10.8 2 2.66 0.75 34.4 3 2.66 1.37 55.8 4 2.59 1.51 82.2 5 2.72 1.90 90.6 6 2.63 2.52 119.5 7 2.67 2.96 140.0 8 2.65 2.76 125.0 9 2.46 2.28 105.6 10 1.82 2.15 103.2 a millilitres per minute per watt 87 using the Matlab Surface Fitting Tool. As the fitted polynomials do not have a physical connection to the given device, the model is applicable to any similar electrochemical H2 generation device with an electric two-pole system. In the different linear and non-linear physical and chemical models different coefficients become dominant. The voltage relationship is given by Eq.(5), where icell denotes the cell current and cKOH stands for the KOH concentration. Parameters can be found in Table 2. ucell(icell,cKOH)= p00 + p10 icell + p01 cKOH + p20 icell 2 + p11 icellcKOH + p02cKOH 2 + p30icell 3 + p21icell 2 cKOH (5) The volume of the generated H2 is given by Eq.(6). H2(icell, cKOH ) = p00 + p10 icell + p01 cKOH + p20 icell 2 + p11 icellcKOH + p02 cKOH 2 + p30 icell 3 + p21 icell 2 cKOH + p12 icellcKOH 2 + p03 cKOH 3 + p40 icell 4 + p31 icell 3 cKOH +p22 icell 2 cKOH 2 + p13 icellcKOH 3 (6) Table 3 and Figs.3-4 show representative results for the model. As expected, the H2 generation speed decreases and the cell finally stops working as the amount of water decreases and the KOH concentration increases. A Simulink block scheme of the cell model is depicted in Fig.5. This Simulink model was validated by considering a system with the same parameters as the layout of the experimental cell. In this layout, we ran a simulation for 24 h using this model, decreasing water and increasing KOH concentrations. The results of this simulation can be seen in Fig.6. It can be seen that the hydrogen gas generated is reduced because of the rising KOH concentration. The exact values are in good agreement with our measurements. Dry Cell Model in Complex Energetic Systems The model for H2 generating cells described in the previous section was investigated in the Matlab Simulink simulation environment that studies the energy flow conditions of a complex energetic system consisting of a renewable source with a grid- Table 2: Coefficients of the polynomial relationship describing the cell voltage Value Value Value p00 1.429 p10 0.2548 p01 -0.1226 p20 -0.008571 p11 -0.01191 p02 0.008257 p30 0.0001141 p21 -8.76e-05 p12 0.0009697 Table 3: Coefficients of the polynomial relationship for the generated H2 gas Value Value Value p00 -0.1695 p10 0.1687 p01 -0.01765 p20 -0.007486 p11 -0.03234 p02 0.03446 p30 -0.0001077 p21 0.00412 p12 -0.004094 p03 -0.004061 p40 -4.269e-06 p13 6.169e-05 p22 -0.0005518 p13 0.0009544 Figure 3: Simulation of the cell model with a constant current of 5 A for 1 day Figure 4: Generated H2 as a function of KOH concentration and dry cell current Figure 5: Matlab Simulink model of the HHO cell. The functional blocks implementing Eqs.(5) and (6) are denoted by different background colours Figure 6: Simulation of a cell model with a constant current of 5 A for 1 day 88 synchronized inverter, a low voltage grid, an intermediate voltage controller [3,5] and a lithium ion battery. We replaced the lithium ion battery in this cell model, which reduces the potential energy flow modes, because this cell can only adsorb current for storing energy in hydrogen production. It cannot reverse the electrochemical process for electrical energy generation from hydrogen gas. The structure of the system can be seen in Fig.7, where it is apparent that the cell model is connected directly only to the grid-synchronized inverter module of the system. The system depicted in Fig.7 operates in different discrete states according to the energy flow direction. Four cases can be defined: • Normal inverter mode: The energy flows from the renewable source to the grid only (Fig.8A). • Normal inverter and hydrogen generation mode: The energy flows from the renewable source to both the dry cell and the grid (Fig.8B). • Hydrogen generation only mode: The energy flows from the grid to the dry cell only (Fig.8C). • Distortion reduction only mode: The energy flows from the grid into the intermediate capacitance and from the intermediate capacitance into the grid. The energy balance is zero for a period, and the active power is zero (Fig.8D). Model verification was performed by changing the energy flow modes in subsequent time intervals, and this was implemented by changing the energy balance of the system with outer current loads (Iouter load). The different values for the simulations as parameters can be Figure 7: Simulink model of a complex energetic system with an HHO cell model inside Figure 8: Complex energetic system energy flow modes: A: normal inverter mode, B: inverter and hydrogen generator mode, C: hydrogen generation only mode, D: distortion reduction only mode 89 seen in Table 4. The simulation results are shown in Fig.9, where Uconn is the effective value of the voltage at the connection point, IHHO is the current value of the dry cell, VH2 is the volume of generated hydrogen gas, cKOH is the KOH concentration of the electrolyte and Vwater is the volume of water inside the cell system. These values are plotted as a function of time. The results of the simulation show that the behaviour of the simulated electronic two-pole system is identical to that of the measured database. Conclusion We developed a complex model to investigate renewable energy sources, for the conversion of surplus energy to hydrogen gas for storage and to represent how a main grid is fed. We built a measurement-based model of a hydrogen-generating cell for the simulation of complex energy systems in MATLAB SIMULINK environment. We estimated the parameters of the model based on measurements collected during the detailed examination of a demonstration cell. We carried out a series of experiments on a HHO gas producing dry cell to find the optimal electrolyte concentration, current value, etc. or change the setup by altering the distance between the plates with KOH electrolyte solution. We monitored the experimental setup in several regards, for example cell voltage, and gas production. The novel element is the temperature and concentration dependent Matlab Simulink model of the hydrogen generation cell, which was found to be suitable for simulation purposes. We tested it in a simulation of a complex domestic power plant using a renewable energy source and hydrogen generation cell. Hydrogen generation enables the long-term storage of surplus electrical energy collected, but not consumed or injected into the low voltage grid. The generated hydrogen can be consumed by vehicles for transportation purposes or it can be applied in fuel cells generating direct electrical energy for energy-deficient low voltage network situations. We simulated all the potential energetic situations in this complex model of an energetic system. The simulations showed that the presented model of a hydrogen- generating cell performed well. ACKNOWLEDGEMENT We acknowledge the financial support of this work by the Hungarian State and the European Union under the TAMOP-4.2.2.A-11/1/ KONV-2012-0072 project. REFERENCES [1] KOUTSONIKOLASA D.E., KALDISA, S.P., PANTOLEONTOSB, G.T., ZASPALISAC, V.T., SAKELLAROPOULOSC, G.P.: Techno-Economic Assessment of Polymeric, Ceramic and Metallic Membranes Integration in an Advanced IGCC Process for H2 Production and CO2 Capture, Chem. Engng. Trans. 2013, 35, 715-720 [2] ZIOGOUA C., ELMASIDESB C., PAPADOPOULOUCA S., VOUTETAKISA S.: Supervisory Control and Unattended Operation of an Off-Grid Hybrid Power Generation Station with Hydrogen Storage Chem. Engng. Trans. 2013, 35, 529-534 [3] GÖRBE P., MAGYAR A., HANGOS K.M.: Line Conditioning with Grid Synchronized Inverter's Power Injection of Renewable Sources in Nonlinear Distorted Mains, Proc. 10th International PhD Workshop on Systems and Control, 2009, 978-980 [4] AL-ROUSAN A.A.: Reduction of fuel consumption in gasoline engines by introducing HHO gas into intake manifold, Int. J. Hydrogen Energy, 2010, 35, 12930-12935 [5] GÖRBE P., MAGYAR A., HANGOS K.M.: THD Reduction with Grid Synchronized Inverter’s Power Injection of Renewable Sources, Proc. 20th Int. Symp. Power Electronics, Electrical Drives, Automation and Motion (SPEEDAM) 2010, 1381- 1386 Figure 9: Simulation results of a complex energetic system using a cell model short time range (5 sec)