CET Volume 86 DOI: 10.3303/CET2186160 Paper Received: 16 August 2020; Revised: 4 February 2021; Accepted: 10 April 2021 Please cite this article as: Bubpha C., Siemanond K.., 2021, Circular Integration of Crude Distillation Process, Chemical Engineering Transactions, 86, 955-960 DOI:10.3303/CET2186160 CHEMICAL ENGINEERING TRANSACTIONS VOL. 86, 2021 A publication of The Italian Association of Chemical Engineering Online at www.cetjournal.it Guest Editors: Sauro Pierucci, Jiří Jaromír Klemeš Copyright © 2021, AIDIC Servizi S.r.l. ISBN 978-88-95608-84-6; ISSN 2283-9216 Circular Integration of Crude Distillation Process Chanikan Bubpha, Kitipat Siemanond* The Petroleum and Petrochemical College, Chulalongkorn University, Bangkok, Thailand kitipat.s@chula.ac.th Circular integration (CI) of crude distillation unit (CDU) is introduced the closed loop to recover waste heat from hot product streams to heat the cold crude steams and reduce the external energy resulting in lower CO2 emission and utilities cost saving. The stagewise superstructure by Yee and Grossmann (1990) is applied to retrofit heat exchanger network (HEN) of crude preheat train supported by variable heat capacity as a function of temperature. The objective function for HENS is to maximize Net Present Value (NPV) of retrofitted HEN which relates to maximizing utilities cost saving. The optimization problem is solved by using General Algebraic Modelling System (GAMS) with DICOPT optimization solver. The HEN retrofit result shows that this model can reduce 65.44% of CO2 emission and achieve 12,584,244.74 $ of NPV. 1. Introduction CDU unit is the process in refinery and for separation of various crude fractions by their boiling points. The resource of this process is crude oil containing mixture of hydrocarbons with impurities of salts and water, which need to be removed by desalter before entering the distillation column to avoid corrosion of equipment in the crude preheat train (Dawe and Lucas, 2000). The furnace is required to heating up the desalted crude by heat of combustion. Therefore, this process requires external hot utilities for heating process causing large amount of direct CO2 emissions. The refinery is the third largest direct Greenhouse Gas (GHG) emissions reported about 177.6 million metric tons CO2e (EPA, 2019). The carbon emissions caused climate change and increasing of smog and air pollution (Osmanski, 2020) which cause humans respiratory disease. The energy efficiency improvement of crude preheat train reduces the energy consumption and CO2 emission at the furnace by heat recovery. The heat recovery can be achieved by reusing waste heat from the process hot streams of petroleum products to preheat process cold stream of crude. Heat and Water Recovery System (HWRS) proposed by Yoo et al. can achieve 4.348% reduction in CO2 emission and 20.29% reduction in fuel consumption by using hot flue gases as heat source to preheat water from water desalination (Yoo et al., 2019). Due to the increase of population, industrialization and standard of living, resulting in declining natural resources, global climate changes and risk to ecosystem, the sustainability plays the important role by balancing three principle including environmental protection, economic growth, and societal equity (El- Halwagi, 2017). These balancing also called circular integration which refers to Circular Economy (CE), Process Integration (PI), and Industrial Ecology (IE) (Walmsley et al., 2019). The Process Integration is developed to obtain the sustainable design focusing on the optimal design of HEN retrofit for reducing energy demand relating to the cost-efficiency, and the cost of new equipment and piping modifications (Smith, 2017). The retrofitting heat exchanger network can be achieved by relocation of heat exchanger, re-piping or adding new heat exchanger (Sieniutycz and Jezowski, 2018). Heat integration had been applied in crude distillation unit using HEN which the hot-product streams are used to heat up the cold-feed streams before entering distillation with decrease of utility consumption (Sieniutycz and Jezowski, 2018). The heat integration in pharmaceutical processing facility by transferring heat from hot to cold streams reduces hot utility from 4870 kW to 2620 kW and cold utility from 2300 kW to 50 kW (El-Halwagi, 2017). 955 2. Methodology This study aims to optimize the problem of HEN retrofitting for crude preheat train. The CDU is simulated by PRO/II simulation software to collect the stream data, like specific heat capacity (Cp), flowrate (F) and Temperature (T) of hot and cold process streams. These data are parameters used for Mixed Integer Non- Linear Programming (MINLP) optimization to retrofit crude preheat train. Stage-wise superstructure (Yee and Grossmann, 1990) is applied to the HEN retrofit model with variable heat capacity which is the function of temperature (Sreepathi and Rangaiah, 2015) to find the general form of heat transfer rate (Q) under variable heat capacity shown in Eq 1. Assume Cp(T) = eT2 + fT + g. Then, Q = F · Cp(T) dT = F · (eT2 + fT + g) T1 T2 dT = F · eT3 3 + fT2 2 + gT T1 T2 = F · eT133 + fT122 + gT1 +c - eT233 + fT222 + gT2 +c where e 3 = a, f 2 = aa and g = aaa Q = F · a·T1 3 +aa·T1 2 + aaa ·T1 - a·T23 +aa·T22 + aaa ·T2 (1) 2.1 Objective function for HEN retrofit This objective function is used to maximize the Net Present Value (NPV) as follows: Maximize NPV = ∑ ∑ CCUi (qcuoldi-qcui)+ ∑ CHUj (qhuoldj-qhuj) y 1+i y n y=1 - ∑ ∑ ∑ CFij× max 0, zijk-zoldijkkji - ∑ CFi,CU×max(0,(zcui-zcuoldi))i - ∑ CFj,HU×max(0,(zhuj-zhuoldj)j ) - ∑ ∑ ∑ Cijmax(0,(Areaijk-Areaoldijk))Bijkji - ∑ Ci,CUmax(0,(Areai-AreaColdi))Bi,CUi - ∑ Cj,HUmax(0,(Areaj-AreaHoldj))Bj,HUj Where n = lifetime in years, and I = annual interest rate zoldijk, zcuoldi, and zhuoldj are binary variables for base-case process, cold-utility, and hot-utility exchangers, respectively. Areaoldijk, AreaColdi, and AreaHoldj are area for base-case process, cold-utility, and hot-utility exchangers, respectively. 3. Result and discussion 3.1 Crude distillation unit as base case Base-case crude distillation unit was simulated by PRO/II (version 10) simulation containing 4 hot product streams (H1, H2, H3, and H4) and 2 cold crude streams (C1 and C2) with 4 process streams (E1, E2, E3, and E4), 3 coolers (CU1, CU, and CU3) and 2 Heater (HU1 and HU2) as shown in Figure 1 and 2. The stream and economic cost data of base-case of exchangers are shown in Table 1. From this base-case crude preheat train, the energy consumption of hot and cold utilities, and the area of heat exchanger and utilities are identified as base-case are shown in Table 2. 956 Figure 2: Existing process flow diagram for base-case crude preheat train by PRO/II Table 1: Process streams and economic cost data. Strea m TIN (°C) TOUT (°C) Flowrate (kg/s) Cp = aT2+bT+c (kJ/kg·°C) h (kW/m2·°C) Cost ($/kW· year) a b c Hot streams H1 199.21 40 25.3275 -2.2990E-06 4.4100E-03 1.7680 0.2 H2 266.05 40 27.0459 -2.0260E-06 4.4046E-03 1.6882 0.2 H3 330.08 45 16.1855 -1.7748E-06 4.3359E-03 1.6511 0.2 H4 358.52 45 105.789 -1.8025E-06 4.3126E-03 1.5840 0.2 Cold streams C1 29.51 94.71 237.398 -1.6601E-05 5.2186E-03 1.9368 0.2 C2 106.43 259.31 212.022 -3.5816E-06 4.4560E-03 1.7209 0.2 Utilities CU 15 28 - - - - 0.53 10 HU 400 400 - - - - 0.53 100 New heat exchanger cost ($): 250,000 + 550 ∆A Additional area cost for existing heat exchangers ($): 550∆A Project lifetime: 3 years Annual interest rate: 0% Figure 1 Existing grid diagram for base-case crude preheat train 957 3.2 Optimization and Validation results Figure 3: Optimal retrofitting HEN configuration by GAMS. Figure 4: Process flow diagram of retrofitted crude preheat train validated by PRO II. The stream and economic cost data were used to retrofit the base-case HEN with the maximum NPV through optimization in GAMS by using 4-stage superstructure model with variable Cp and exchanger minimum approach temperature (EMAT) of 3°C. The result of HEN retrofit model requires 2 new heat exchangers (E5 and E6) and additional area of three existing exchangers (E1, E3, and E4) as shown in Figure 3 and Table 2. The result shows the hot utility (HU) consumption reduces from 82,102.12 to 28,446.53 kW (65.35% saving in HU) and cold utility (CU) consumption reduces from 82,817.99 to 29,162.40 kW (67.79% saving in CU). NPV of this solution is $12,584,244.74 over 3 years with annual interest rate equal to 0 %. These results are validated by PRO/II simulation fixing hot and cold stream temperature of HEN and HU, as shown in Figure 4. The error between optimization and validation result is shown in Table 3. The difference between these two results comes from the difference of LMTD approximation and the inlet and outlet temperature of heat exchanger and utilities. In Table 4, the compared of total CO2 emission and total energy consumption for the CDU without HEN, the base case and retrofitting HEN show this model recover more heat to reduce external utilities consumption the utilities cost and CO2 emission by reducing the fuel consumption usage about 65.44% of CO2 emission (4.7612 to 1.6457 kg/s) as shown in Table 4. 958 Table 2: the energy consumption of hot and cold utilities, and the area of heat exchanger and utilities of base- case and retrofit HEN from GAMS. Unit Heat Duty (kW) Area (m2) Base Case Retrofit Base Case Retrofit E1 4,167.297 4,167.297 284.368 284.638 E2 10,073.725 10,073.725 2,361.281 2,361.281 E3 6,553.770 6,553.770 1,325.718 1,734.801 E4 9,104.049 9,104.049 1,802.724 1,833.564 E5 - 5,1642.71 - 7,601.185 E6 - 2,012.878 - 362.44 HU1 18,255.899 1,6243.02 390.268 349.261 HU2 63,846.224 12,203.513 2,248.617 554.873 CU1 4,039.571 2,026.693 564.554 384.598 CU2 307.033 307.033 74.023 74.023 CU3 78,471.388 26,828.678 4,478.999 2,579.414 Table 3: The Optimization and validation result of retrofitting HEN. Table 4: The result comparison of base case and retrofitting HEN. 4. Conclusion To make the model more realistic, variable heat capacity is used in stage-wise superstructure model to do HEN retrofit. Circular Integration (CI) is introduced to design the closed loop by using heat recovery that transfer heat between the hot product streams and cold crude feed streams which is also friendly to environment and health by reducing the CO2 emission via less fuel consumption in combustion reaction. The retrofitting HEN is optimized by using GAMS software and then this retrofitting HEN is validated by PRO/II simulation software. The installed exchangers for retrofitting HEN on CDU can reduce CO2 emission about 65.44% of CDU base case. This model can achieve NPV of $ 12,584,244.74. The model also has some validation error with PRO/II resulting from difference of LMTD approximation and the inlet and outlet temperature of heat exchanger and utilities. Unit GAMS PRO/II % Error Q (kW) Area (m2) Q (kW) Area (m2) Q A E1 4,167.297 284.638 4,165.97 285.208 0.03 0.20 E2 10,073.725 2,361.281 10,076.25 2,246.911 0.03 5.09 E3 6,553.770 1,734.801 6,556.80 1,496.064 0.05 15.96 E4 9,104.049 1,833.564 9,026.00 1,739.231 0.86 5.42 E5 51,642.710 7,601.185 51,645.48 7,323.709 0.01 3.79 E6 2,012.878 362.440 2,011.20 362.149 0.08 0.08 HU1 16,243.020 349.261 16,496.99 355.243 1.54 1.68 HU2 12,203.513 554.873 16,652.38 742.579 26.72 25.28 CU1 2,026.693 384.598 2,027.51 385.171 0.04 0.15 CU2 307.033 74.023 307.26 74.182 0.07 0.21 CU3 26,828.678 2,579.414 26,849.74 2,567.496 0.08 0.46 Base case Retrofit GAMS PRO/II GAMS PRO/II CO2 emission (fuel consumption) HU1 (Kg/s) - 1.2057 - 0.8190 HU2 (Kg/s) - 3.5555 - 0.8267 Total energy consumption HU (KW) 82,102.12 95,903.38 28,446.53 33,149.37 CU (KW) 82,817.99 82,847.09 29,162.40 29,184.51 NPV ($) (3 years) - - 12,584,244.74 15,537,729.81 959 Acknowledgements We would like to express our best gratitude to Chulalongkorn University’s Rachadapisaek Sompote Fund (2017) for supporting this research, and The Petroleum and Petrochemical College (PPC) for partially funding support. References Dawe, R.A., and Lucas, A.G. (2000). Modern petroleum technology, John Wiley & Sons Inc. El-Halwagi, M.M. (2017). Sustainable design through process integration: fundamentals and applications to industrial pollution prevention, resource conservation, and profitability enhancement, Butterworth- Heinemann. Elkamel, A., Ba-Shammakh, M., Douglas, P., and Croiset, E. (2008). An optimization approach for integrating planning and CO2 emission reduction in the petroleum refining industry. Industrial & Engineering Chemistry Research, 47(3), 760-776. EPA. (2019). Greenhouse Gas Reporting Program (GHGRP). Retrieved from www.epa.gov/ghgreporting/ghgrp-reported-data OSMANSKI, S. (2020). How Do Carbon Emissions Affect the Environment? Retrieved from www.greenmatters.com/p/how-do-carbon-emissions-affect-environment Roh, K., Lim, H., Yoo, H., Imran, H., Al-Hunaidy, A.S., and Lee, J.H. (2017). Technical and Strategic Approaches for CO. Saudi Aramco Journal of Technology, 41. Sieniutycz, S., and Jezowski, J. (2018). Energy optimization in process systems and fuel cells, Elsevier. Smith, R. (2017). Process Integration: Current Status and Future Challenges. Computer Aided Chemical Engineering, 40, 9-12. Sreepathi, B.K., and Rangaiah, G.P. (2015). Retrofitting of heat exchanger networks involving streams with variable heat capacity: Application of single and multi-objective optimization. Applied Thermal Engineering, 75, 677-684. Walmsley, T.G., Ong, B.H., Klemeš, J.J., Tan, R.R., and Varbanov, P.S. (2019). Circular Integration of processes, industries, and economies. Renewable and Sustainable Energy Reviews, 107, 507-515. Yee, T.F., and Grossmann, I.E. (1990). Simultaneous optimization models for heat integration—II. Heat exchanger network synthesis. Computers & Chemical Engineering, 14(10), 1165-1184. Yoo, H., Roh, K., Al Hunaidy, A.S., Imran, H., and Lee, J.H. (2019). Optimal design of heat and water recovery system utilizing waste flue gases for refinery CO2 reduction. Computers & Chemical Engineering, 124, 140-152. 960