CHEMICAL ENGINEERING TRANSACTIONS VOL. 78, 2020 A publication of The Italian Association of Chemical Engineering Online at www.cetjournal.it Guest Editors: Jeng Shiun Lim, Nor Alafiza Yunus, Jiří Jaromír Klemeš Copyright © 2020, AIDIC Servizi S.r.l. ISBN 978-88-95608-76-1; ISSN 2283-9216 Optimization of Hydrogen Supply Chain: A Case Study in Malaysia Angel Xin Yee Maha, Wai Shin Hoa,*, Mimi H. Hassima, Haslenda Hashima, Peng Yen Liewb, Umi Aisah Aslia, Zarina Ab. Muisa, Gabriel Hoh Teck Lingc aSchool of Chemical and Energy Engineering (SCEE), Universiti Teknologi Malaysia (UTM), 81310 UTM Johor Bahru, Johor, Malaysia. bDepartment of Environmental Engineering and Green Technology, Malaysia-Japan International Institute of Technology (MJIIT), Universiti Teknologi Malaysia (UTM), Jalan Sultan Yahya Petra, 53100, Kuala Lumpur, Malaysia. cDepartment of Urban and Regional Planning, Faculty of Built Environment and Surveying, Universiti Teknologi Malaysia, 81310 UTM Johor Bahru, Johor, Malaysia hwshin@utm.my Hydrogen is regarded as the fuel of future by having greater heating value than the conventional fuels and with zero carbon emission. Most of the previous supply chain studies only consider the application of hydrogen as transportation fuel. Taking Johor as a case study, this paper aims to develop a holistic optimization model that exploits the use of hydrogen for vehicle fueling and electricity generation. Oil palm biomass and solar energy are used as the energy sources to produce hydrogen and electricity to satisfy the local energy demand. Through this study, the optimal configuration of hydrogen supply chain in Johor has been identified and the associated cost is found to be 3,644,800 USD/d. 1. Introduction The world energy demand has been elevating as a result of economic growth and development of global markets (Sáez-Martínez et al., 2016). Concurrently, the over-reliance on fossil fuels for energy generation has raised concerns regarding energy security and climate change. Thus, the transition of energy production from fossil fuels to renewable sources is crucial for a sustainable ecosystem. Numerous studies have been conducted to reduce the use of fossil resources by substituting with renewable energy supply, i.e. biomethane injection into natural gas distribution grid (Hoo et al., 2018) and biomass co-firing in coal-fired power plant (Mohd Idris, 2018). Hydrogen is an energy carrier that can be synthesized from both fossil resources and renewable resources (Palma et al., 2018). With its clean emission during combustion (Haron et al.,2017), hydrogen could serve as the fuel for fuel cell electric vehicles, fuel for combustion as well as electrical energy storage that stores the excess electricity generated from intermittent renewable sources (Mah et al., 2019). For smooth operation of hydrogen-based energy system, the hydrogen supply chain (HSC) must be well planned. In recent studies, Kim and Kim (2016) developed an optimization model to design and analyze the hydrogen supply system using onshore and offshore wind energy. The onshore wind farms were more preferable than offshore farms due to lower capital investment, while centralized hydrogen production system was more superior over the distributed system owing to economy of scales. Reuß et al. (2017) investigated the use of liquid organic hydrogen carriers (LOHC) as seasonal hydrogen storage to balance the fluctuating renewable electricity generation and fuel demand. Through their study, LOHC would be a promising option if salt cavern storage for hydrogen remains uncompetitive. Won et al. (2017) formulated a MILP model by integrating multiple resources and technologies in which hydrogen is only produced from renewable energy sources (RES). The model was validated through a case study on Jeju Island and results showed that capital cost contributed the most significant part to the expenditure of RES-based HSC. While most of the previous studies focus on the application of hydrogen in the transportation sector, this paper aims to develop a holistic optimization model that considers the use of hydrogen for electricity generation in addition to the transportation fuel. DOI: 10.3303/CET2078015 Paper Received: 11/04/2019; Revised: 06/09/2019; Accepted: 17/10/2019 Please cite this article as: Mah A.X.Y., Ho W.S., Hassim M.H., Hashim H., Liew P.Y., Asli U.A., Ab Muis Z., Ling G.H.T., 2020, Optimization of Hydrogen Supply Chain: A Case Study in Malaysia, Chemical Engineering Transactions, 78, 85-90 DOI:10.3303/CET2078015 85 2. Methodology 2.1 Hydrogen supply chain superstructure The superstructure of hydrogen supply chain is illustrated in Figure 1, where the empty fruit bunches (EFB) and palm kernel shells (PKS) generated from palm oil mills serve as the raw materials of gasification process to produce hydrogen. Solar radiation will be collected through the photovoltaic (PV) panel and converted to electrical energy in the form of direct current. Inverter is used to convert electricity from direct current form into alternating current form. The electricity produced from solar radiation can be used to satisfy electricity demand of the district or converted into hydrogen via electrolysis. The hydrogen produced from gasification and electrolysis processes is stored in liquid form, which is then used to fulfill local hydrogen fuel demand or generate electricity through fuel cell to cater to the local electricity demand. Excess H2 produced in a district can be exported to other districts that have insufficient energy supply. Figure 1: Superstructure of integrated hydrogen supply chain 2.2 Mathematical model The mixed-integer linear programming (MILP) model in this work is extended from the model proposed by Almaraz et al. (2014), where PV system and fuel cell system have been incorporated into the model to consider the electricity generation from hydrogen and the interconversion between electricity and hydrogen. 2.2.1 Production constraints Eq(1) represents the mass balance of hydrogen product in each district: 𝑃𝑇𝑖𝑔 = ∑ (𝑄𝑖𝑙𝑔𝑔′ − 𝑄𝑖𝑙𝑔′𝑔 )𝑙,𝑔′ + (𝐻2𝑖𝑔 𝑡𝑜𝑒 − 𝐻2𝑖𝑔 𝑓𝑟𝑜𝑚𝑒 ) + 𝐷𝑇𝑖𝑔 ∀𝑖, 𝑔; 𝑔 ≠ 𝑔′ (1) where PTig is production rate of hydrogen form i in district g, Qilgg’ is amount of hydrogen form i delivered from district g to district g’ through transportation mode l , Qilg’g is amount of hydrogen form i delivered from district g’ to district g through transportation mode l and DTig is the hydrogen fuel demand in form i in district g. As shown in Eq(2), the total hydrogen production from biomass, PTig is equal to the sum of hydrogen produced from all types of production plants that utilize biomass as feedstock. Eq(3) illustrates that the capacity of each production plant should be within the lower and upper limits. 𝑃𝑇𝑖𝑔 = ∑ 𝑃𝑅𝑝𝑖𝑔𝑝 ∀𝑖, 𝑔 (2) 𝑃𝐶𝑎𝑝𝑝𝑖 𝑚𝑖𝑛𝑁𝑃𝑝𝑖𝑔 ≤ 𝑃𝑅𝑝𝑖𝑔 ≤ 𝑃𝐶𝑎𝑝𝑝𝑖 𝑚𝑎𝑥𝑁𝑃𝑝𝑖𝑔 ∀𝑝, 𝑖, 𝑔 (3) where PRpig is the production capacity of plant p producing hydrogen form i in district g, PCapminpi is the minimum capacity of plant p producing hydrogen form i, PCapmaxpi is the maximum capacity of plant p producing hydrogen form i and NPpig is the number of plant p producing hydrogen form i in district g. 2.2.2 Transportation constraints Eq(4) indicates the amount of hydrogen transported between districts should not violate the allowable limits. Eq(5) shows a district should not be exporting and importing hydrogen to/from the same district at a time. 𝑄𝑖𝑙 𝑚𝑖𝑛 𝑋𝑖𝑙𝑔𝑔′ ≤ 𝑄𝑖𝑙𝑔𝑔′ ≤ 𝑄𝑖𝑙 𝑚𝑎𝑥 𝑋𝑖𝑙𝑔𝑔′ ∀𝑖, 𝑙, 𝑔, 𝑔′; 𝑔 ≠ 𝑔′ (4) 𝑋𝑖𝑙𝑔𝑔′ + 𝑋𝑖𝑙𝑔′𝑔 ≤ 1 ∀𝑖, 𝑙, 𝑔, 𝑔′; 𝑔 ≠ 𝑔′ (5) 86 where Qminil is the minimum amount of hydrogen form i transported via transportation mode l, Qmaxil is the maximum amount of hydrogen form i transported via transportation mode l, Xilgg’ is a binary determinant for the transportation of hydrogen form i via transportation mode l from district g to district g’, and Xilg’g is a binary determinant for the transportation of hydrogen form i via transportation mode l from district g’ to district g. 2.2.3 Storage constraints Eq(6) shows the storage requirement of product and Eq(7) indicates the hydrogen product can be stored in tanks of different capacities. Eq(8) depicts the hydrogen stored in tanks should be within the allowable limits. 𝑆𝑇𝑖𝑔 = 𝛽(𝐷𝑇𝑖𝑔 + 𝐻2𝑖𝑔 𝑡𝑜𝑒 ) ∀𝑖, 𝑔 (6) 𝑆𝑇𝑖𝑔 = ∑ 𝑆𝐼𝑠𝑖𝑔𝑠 ∀𝑖, 𝑔 (7) 𝑆𝐶𝑎𝑝𝑠𝑖 𝑚𝑖𝑛𝑁𝑆𝑠𝑖𝑔 ≤ 𝑆𝐼𝑠𝑖𝑔 ≤ 𝑆𝐶𝑎𝑝𝑠𝑖 𝑚𝑎𝑥𝑁𝑆𝑠𝑖𝑔 ∀𝑖, 𝑔 (8) where STig is the total inventory of product i in district g, β is the storage holding period, SCapminsi is the minimum storage capacity of hydrogen form i in storage unit s, SCapmaxsi is the maximum storage capacity of hydrogen form i in storage unit s, SIsig is the hydrogen stored in form i in storage unit s in district g and NSsig is the number of storage unit s storing hydrogen form i in district g. 2.2.4 PV system The electricity produced from solar radiation is given in Eq(9). Eq(10) indicates the electricity generated from solar radiation can be used to fulfil the local electricity demand or converted to H2. Eq(11) depicts the total electricity demand in a district can be satisfied by electricity produced from solar radiation and hydrogen. 𝐸𝑔 𝑆 = 𝑅𝑔 𝑆 𝑃𝑉𝑔 𝐴 𝑃𝑉𝐸𝐹𝐹 𝐼𝑁𝑉 𝐸𝐹𝐹 ∀𝑔 (9) 𝐸𝑔 𝑆 = 𝐸𝐷𝑔 + 𝐸𝑔 𝑡𝑜𝐻2 ∀𝑔 (10) 𝐷𝐸𝑇𝑔 = 𝐸𝐷𝑔 + 𝐸𝑔 𝑓𝑟𝑜𝑚𝐻2 ∀𝑔 (11) where ESg denotes the electricity produced from solar radiation in district g, RSg is the solar radiation in district g, PVAg is the area of PV panel in district g, PVEFF is the efficiency of PV panel, INVEFF is the efficiency of inverter, EDg is the amount of electricity transmitted to the demand site in district g, EtoH2g is the amount of electricity converted to H2 through electrolysis in district g, DETg is the total electricity demand in district g, and EfromH2g is the amount of electricity produced from H2 through fuel cell in district g. 2.2.5 Fuel cell system and electrolysis plant The amount of electricity converted from H2 is given by Eq(12). Eq(13) computes the yield of hydrogen in electrolysis process. The sum of hydrogen produced from each type of electrolysis plant should balance the total amount of hydrogen generated from electricity as illustrated in Eq(14). Eq(15) shows the production capacity of electrolysis plant is bounded by its lower and upper production limits. 𝐸𝑔 𝑓𝑟𝑜𝑚𝐻2 = ∑ 𝐻2𝑖𝑔 𝑡𝑜𝐸 𝑌𝐻2𝑡𝑜𝐸 𝐼𝑁𝑉𝐸𝐹𝐹 𝑖 ∀𝑔 (12) ∑ 𝐻2𝑖𝑔 𝑓𝑟𝑜𝑚𝑒 = 𝐸𝑔 𝑡𝑜𝐻2 𝑌𝐸𝑡𝑜𝐻2 ∀𝑔𝑖 (13) 𝐻2𝑖𝑔 𝑓𝑟𝑜𝑚𝑒 = ∑ 𝑃𝐸𝑒𝑖𝑔𝑒 ∀𝑖, 𝑔 (14) 𝐸𝐶𝑎𝑝𝑒𝑖 𝑚𝑖𝑛𝑁𝐸𝑒𝑖𝑔 ≤ 𝑃𝐸𝑒𝑖𝑔 ≤ 𝐸𝐶𝑎𝑝𝑒𝑖 𝑚𝑎𝑥𝑁𝐸𝑒𝑖𝑔 ∀𝑒, 𝑖, 𝑔 (15) where H2toEig is the amount of hydrogen form i converted to electricity in district g and YH2toE is yield of electricity per unit hydrogen reacted in fuel cell, YEtoH2 is yield of H2 per unit electricity consumed and PEeig is production rate of hydrogen form i at electrolysis plant e in district g, ECapminei is the minimum capacity of electrolysis plant e producing hydrogen form i, ECapmaxei is maximum capacity of electrolysis plant e producing hydrogen form i and NEeig is the number of electrolysis plant e producing hydrogen form i in district g. 87 Eq(16) and (17) indicate that the capacity of inverter is dependent on the amount electricity from PV panel or fuel cell, whichever is greater. Meanwhile, the capacity of fuel cell is dependent on the electricity throughput as shown in Eq(18). 𝐼𝑁𝑉𝑔 𝐶𝐴𝑃 ≥ 𝐸𝑔 𝑆 𝐻𝑆 𝐼𝑁𝑉 𝐸𝐹𝐹 ∀𝑔 (16) 𝐼𝑁𝑉𝑔 𝐶𝐴𝑃 ≥ 𝐸𝑔 𝑓𝑟𝑜𝑚𝐻2 𝐻𝑆 𝐼𝑁𝑉 𝐸𝐹𝐹 ∀𝑔 (17) 𝐹𝐶𝑔 𝐶𝐴𝑃 ≥ 𝐸𝑔 𝑓𝑟𝑜𝑚𝐻2 𝐻𝑆 𝐹𝐶 𝐸𝐹𝐹𝐼𝑁𝑉 𝐸𝐹𝐹 ∀𝑔 (18) where INVCAPg is inverter capacity in district g, HS is peak sun hours, FCCAPg is fuel cell capacity in district g. 2.2.6 Objective function The objective function of this model is to minimize the total daily cost of HSC which is defined in Eq(19). Eq(20) and (21) represents the mathematical expression for facility capital cost and operating cost. 𝑇𝐷𝐶 = 𝐹𝐶𝐶+𝑇𝐶𝐶 𝛼∙𝐶𝐶𝐹 + 𝐹𝑂𝐶 + 𝑇𝑂𝐶 (19) 𝐹𝐶𝐶 = ∑ [∑ (∑ 𝑃𝐶𝐶𝑝𝑖 𝑁𝑃𝑝𝑖𝑔 + ∑ 𝑆𝐶𝐶𝑠𝑖 𝑁𝑆𝑠𝑖𝑔𝑠 + ∑ 𝐸𝐶𝐶𝑒𝑖 𝑁𝐸𝑒𝑖𝑔𝑒𝑝 ) + 𝑃𝑉𝑔 𝐴 𝑃𝑉𝐸𝐹𝐹 𝑃𝑉𝐶𝐶 + 𝐹𝐶𝑔 𝐶𝐴𝑃 𝐹𝐶𝐶𝐶𝑖 +𝑔 𝐼𝑁𝑉𝑔 𝐶𝐴𝑃 𝐼𝑁𝑉𝐶𝐶] (20) 𝐹𝑂𝐶 = ∑ [∑ (∑ 𝑈𝑃𝐶𝑝𝑖 𝑃𝑅𝑝𝑖𝑔 + ∑ 𝑈𝑆𝐶𝑠𝑖 𝑆𝐼𝑠𝑖𝑔𝑠 + ∑ 𝑈𝐸𝐶𝑒𝑖 𝑃𝐸𝑒𝑖𝑔𝑒𝑝 ) + 𝑃𝑉𝑔 𝐴 𝑃𝑉𝐸𝐹𝐹 𝑃𝑉𝑈𝐶 + 𝐹𝐶𝑔 𝐶𝐴𝑃 𝐹𝐶𝑈𝐶𝑖 ]𝑔 (21) where TDC is total daily cost, FCC is facility capital cost, TCC is transportation capital cost, FOC is facility operating cost, TOC is transportation operating cost, α is network operating period, CCF is capital change factor, PCCpi is capital cost of production technology p producing hydrogen form i, SCCsi is capital cost of storage unit s for hydrogen form i, ECCei is capital cost of electrolysis plant e producing hydrogen form i, PVCC is unit capital cost of PV system, FCCC is unit capital cost of fuel cell, INVCC is unit capital cost of inverter, UPCpi is unit production cost of production plant p producing hydrogen form i, USCsi is unit storage cost of storage unit s for hydrogen form i, UECei is unit electrolysis cost for electrolysis plant e producing hydrogen form i, PVUC is unit operating cost of PV system, FCUC is unit operating cost of fuel cell system. Eq. (22) and (23) represent the mathematical expressions for transportation capital cost: 𝑁𝑇𝑈𝑔𝑟𝑖𝑑𝑖𝑙𝑔𝑔′ = [ 𝑄𝑖𝑙𝑔𝑔′ 𝑇𝑀𝐴𝑙𝑇𝐶𝑎𝑝𝑖𝑙 ( 2𝐴𝐷𝑔𝑔′ 𝑆𝑃𝑙 + 𝐿𝑈𝑇𝑙 )] ∀𝑖, 𝑙, 𝑔, 𝑔 ′ (22) 𝑇𝐶𝐶 = ∑ 𝑁𝑇𝑈𝑔𝑟𝑖𝑑𝑖𝑙𝑔𝑔′𝑖𝑙𝑔𝑔′ TMCil (23) where NTUgridilgg’ is the number of transportation unit carrying hydrogen form i via transportation mode l from district g to district g’, TMAl is availability of transportation mode l, TCapil is capacity of transportation mode l carrying hydrogen form i, ADgg’ is the average distance between district g and g’, SPl is the mean speed of transportation mode l, LUTl is the load and unload time for transportation mode l, and TMCil is the cost of establishing transportation mode l transporting hydrogen form i. Eq (24) to (28) show the calculations for transportation operating cost: 𝐹𝐶 = ∑ 𝐹𝑃𝑙𝑖𝑙𝑔𝑔′ ( 2𝐴𝐷𝑔𝑔′𝑄𝑖𝑙𝑔𝑔′ 𝐹𝐸𝑙𝑇𝐶𝑎𝑝𝑖𝑙 ) (24) 𝐿𝐶 = ∑ 𝐷𝑊𝑙𝑖𝑙𝑔𝑔′ [ 𝑄𝑖𝑙𝑔𝑔 𝑇𝐶𝑎𝑝𝑖𝑙 ( 2𝐴𝐷𝑔𝑔 ′ 𝑆𝑃𝑙 + 𝐿𝑈𝑇𝑙 )] (25) 𝑀𝐶 = ∑ 𝑀𝐸𝑙𝑖𝑙𝑔𝑔′ ( 2𝐴𝐷𝑔𝑔′𝑄𝑖𝑙𝑔𝑔′ 𝑇𝐶𝑎𝑝𝑖𝑙 ) (26) 𝐺𝐶 = ∑ 𝐺𝐸𝑙𝑖𝑙𝑔𝑔′ [ 𝑄𝑖𝑙𝑔𝑔 ′ 𝑇𝑀𝐴𝑙𝑇𝐶𝑎𝑝𝑖𝑙 ( 2𝐴𝐷𝑔𝑔 ′ 𝑆𝑃𝑙 + 𝐿𝑈𝑇𝑙 )] (27) 𝑇𝑂𝐶 = 𝐹𝐶 + 𝐿𝐶 + 𝑀𝐶 + 𝐺𝐶 (28) 88 where FC is fuel cost, LC is labour cost, MC is maintenance cost, GC is general cost, FPl is the fuel price of transportation mode l, FEl is the fuel economy for transportation mode l, DWl is the driver wage of transportation mode l, MEl is the maintenance expenses for transportation mode l, GEl is the general expenses for transportation mode l. 3. Case study In this work, the optimization of HSC is performed on Johor region which is geographically divided into 10 districts. Table 1 shows the availability of energy resources and energy demand of the districts, where the fuel and electricity of each district are estimated using its population as illustrated in Eq(29) and (30). The optimal HSC configuration determined from the optimization result is illustrated in Figure 2. 𝐸𝑙𝑒𝑐𝑡𝑟𝑐𝑖𝑡𝑦 𝑑𝑒𝑚𝑎𝑛𝑑 𝑜𝑓 𝑑𝑖𝑠𝑡𝑟𝑖𝑐𝑡 = 𝑁𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑒𝑙𝑒𝑐𝑡𝑟𝑖𝑐𝑖𝑡𝑦 𝑑𝑒𝑚𝑎𝑛𝑑 × 𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑖𝑛 𝑑𝑖𝑠𝑡𝑟𝑖𝑐𝑡 𝑁𝑎𝑡𝑖𝑜𝑛𝑎𝑙 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 (29) 𝐷𝑖𝑠𝑡𝑟𝑖𝑐𝑡 𝑓𝑢𝑒𝑙 𝑑𝑒𝑚𝑎𝑛𝑑 = 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑣𝑒ℎ𝑖𝑐𝑙𝑒 𝑖𝑛 𝐽𝑜ℎ𝑜𝑟 ×𝐷𝑖𝑠𝑡𝑟𝑖𝑐𝑡 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝐽𝑜ℎ𝑜𝑟 𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 × 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑑𝑖𝑠𝑡𝑎𝑛𝑐𝑒 𝑡𝑟𝑎𝑣𝑒𝑙𝑙𝑒𝑑 × 𝐹𝑢𝑒𝑙 𝑒𝑐𝑜𝑛𝑜𝑚𝑦 𝑜𝑓 ℎ𝑦𝑑𝑟𝑜𝑔𝑒𝑛 (30) The vehicle number in Johor and the average distance travelled is obtained from Shabadin et al. (2014), the national electricity demand is obtained from Malaysia Energy Commission (2018), the national and regional population is retrieved from Department of Statistics Malaysia (2010), and the fuel economy of fuel cell vehicle is assumed to be 0.76 kg/100 km, the same as Toyota Mirai (Toyota Motor Europe, 2015). For solar system, the information for PV panel and inverter is obtained from Elshurafa et al. (2019), the average solar radiation in the districts is extracted from Meteoblue (2019) and peak sun hours is assumed to be 4 hours per day. The capital and operating cost of fuel cell is 434 USD/kW and 0.05 USD/kW·d (Steward et al., 2009). Information for production, storage, transportation and capital change factor is adapted from Almaraz et al (2014). Table 1: Transportation fuel and electricity demand of the districts No. District Solar radiation (kWh/d) EFB supply (kg/d) PKS supply (kg/d) Fuel demand (kg H2/d) Electricity demand (kWh/d) 1 Batu Pahat 4.29 424,110 96,389 48,979 4,917,749 2 Johor Bahru 4.89 29,589 6,725 162,604 16,325,378 3 Kluang 4.29 1,692,493 384,658 35,146 3,528,477 4 Kota Tinggi 5.03 1,147,068 260,697 22,894 2,298,250 5 Mersing 4.51 230,538 52,395 8,415 844,640 6 Muar 4.35 345,205 78,456 29,140 2,924,780 7 Pontian 4.89 24,658 5,604 18,280 1,834,670 8 Segamat 4.58 1,010,959 229,763 22,311 2,239,039 9 Ledang 4.55 36,986 8,406 16,082 1,613,831 10 Kulaijaya 4.89 355,068 80,697 29,901 3,001,464 Figure 2: Optimal HSC configuration The optimization model in GAMS was run in Windows 10 Pro 64-bit operating system with Intel(R) Core(TM) i5-6200U CPU @ 2.30GHz 2.40GHz and 8GB installed RAM. The CPU time taken to solve the model with 51 89 variables was 43s. Through GAMS optimization, the optimal cost of HSC is determined to be 3,644,800 USD/d. From Figure 2, it can be observed that all districts will employ PV system and 3 of the districts will be having electrolysis plant. No gasification plant is used meaning that the all hydrogen will be produced from electrolysis process. By having higher intensity of solar radiation and lower energy demand, Kota Tinggi could potentially be a large exporter of hydrogen followed by Pontian, while other districts will be the hydrogen importer. Fuel cell system is only found in Johor Bahru, which is the district with the greatest energy demand. 4. Conclusions The model proposed in this paper extends the use of hydrogen to fulfil the fuel and electricity demand, and the interconversion between hydrogen and electricity is being modelled. This work is significant in exploring the roles of hydrogen in future energy system, and future study should consider the spatial planning of facilities. Acknowledgments The authors would like to acknowledge the research grants from Universiti Teknologi Malaysia with the grant no. Q.J130000.3551.06G47, R.J130000.7351.4J362 and Q.J130000.2546.19H05. References Almaraz S.D.L., Azzaro-Pantel C., Montastruc L., Domenech S., 2014, Hydrogen supply chain optimization for deployment scenarios in the Midi-Pyrénées region, France, International Journal of Hydrogen Energy, 39(23), 11831-11845. Department of Statistics Malaysia, 2010, Population and housing census of Malaysia accessed 14 October 2019. Elshurafa A.M., Alsubaie A.M., Alabduljabbar A.A., Al-Hsaien S.A., 2019, Solar PV on mosque rooftops: Results from a pilot study in Saudi Arabia, Journal of Building Engineering, 25, 100809. Haron R., Yun H.A.H., Mat R., Mohammed M., 2017, Overview of biodiesel wastes utilization for hydrogen production, Chemical Engineering Transactions, 56, 391-396. Hoo P.Y., Hashim H., Ho W.S., 2018, Opportunities and challenges: Landfill gas to biomethane injection into natural gas distribution grid through pipeline, Journal of Cleaner Production, 175, 409-419. Idris M.N.M., Hashim H., Razak N.H., 2018, Spatial optimisation of oil palm biomass co-firing for emissions reduction in coal-fired power plant, Journal of Cleaner Production, 172, 3428-3447. Kim M., Kim J., 2017, An integrated decision support model for design and operation of a wind-based hydrogen supply system, International Journal of Hydrogen Energy, 42(7), 3899–3915. Mah A. X. Y., Ho W. S., Bong C. P. C., Hassim M. H., Liew P. Y., Asli U. A., Kamaruddin M. J., Chemmangattuvalappil N. G., 2019, Review of hydrogen economy in Malaysia and its way forward, International Journal of Hydrogen Energy, 44(12), 5661–5675. Malaysia Energy Commission, 2018, National Energy Balance 2016, Energy Comission accessed 26 August 2019. Meteoblue, 2019, Historical Weather Report accessed 27 August 2019. Palma V., Barba D., Vaiano V., Colozzi M., Palo E., Barbato L., Cortese S., Miccio M., 2018, Catalytic Oxidative Decomposition of H2S for Hydrogen Production, Chemical Engineering Transactions, 70, 325- 330. Reuß M., Grube T., Robinius M., Preuster P., Wasserscheid P., Stolten D., 2017, Seasonal storage and alternative carriers: A flexible hydrogen supply chain model, Applied Energy, 200, 290-302. Sáez-Martínez F., Lefebvre G., Hernández J., Clark J., 2016, Drivers of sustainable cleaner production and sustainable energy options, Journal of Cleaner Production, 138, 1-7. Shabadin A., Megat Johari N., Mohamed Jamil H., 2014, Car annual vehicle kilometer travelled estimated from car manufacturer data–an improved method, Pertanika, 25(1), 171-180. Steward D., Saur G., Penev M., Ramsden T., 2009, Lifecycle cost analysis of hydrogen versus other technologies for electrical energy storage, National Renewable Energy Lab. Toyota Motor Europe, 2015, Toyota Motor Europe accessed 26 August 2019. Won W., Kwon H., Han J. H., Kim J., 2017, Design and operation of renewable energy sources based hydrogen supply system: Technology integration and optimization, Renewable Energy, 103, 226–238. 90