Master Thesis
Configuration and Performance of Hydrogen Storage
System to Facilitate the Integration of Renewable
Energy Sources into the Electricity Grid
MSc Supply Chain Management, minor energy
University of Groningen, Faculty of Business and Economics
January 27, 2020
EVA VERTADIER
Student number: S3483053
E-mail: e.vertadier@student.rug.nl
Supervisor: dr. X. (Stuart, Xiang) Zhu
Co-assessor: dr. E. (Evrim) Ursavas
Acknowledgment: First of all my thanks go to my supervisor dr .X. Zhu and the co-assessor of my thesis dr .E. Ursavas. They provided helpful insights on mathematical modelling and energy systems. Finally, I would like to
2 Abstract
3
Table of Content
1 Introduction 5
2 Literature Review 7
2.1 Challenges and opportunities of integrating renewable energies into the
electricity grid 7
2.2 Hydrogen-storage system 8
2.3 Contribution to the literature 10
3 Methodology 11
3.1 Model development 11
3.2 Mathematical modelling 13
4 Case study 1: Study of a hydrogen-storage system supplied by wind energy
in Gelderland 18
4.1 Scenario details 18
4.2 Electricity demand and supply 19
4.3 Technical parameters of hydrogen technologies 19
4.4 Costs of hydrogen technologies 22
5 Results of Case Study 1 23
5.1 Optimal configurations of the hydrogen-storage system 24
5.2 Hydrogen-storage level 26
5.3 Costs analysis of the hydrogen-storage system 29 6 Case study 2: Study of a hydrogen-storage system supplied by solar energy
in Groningen 32
6.1 Scenario details 32
6.2 Electricity demand and supply 33
6.3 Technical parameters of hydrogen technologies 34
6.4 Costs of hydrogen technologies 35
7 Results of Case Study 2 38
7.1 Optimal configurations of the hydrogen-storage system 38
7.2 Hydrogen-storage level 40
7.3 Costs analysis of the hydrogen storage system 42
8 Discussion 45
9 Conclusion 48
REFERENCES 50
4
List of abbreviations
AEC Alkaline Electrolyser
PEMEC Polymer Electrolyte Membrane Electrolyser SOEC Solide Oxide Electrolyser
PHS Pumped Hydroelectric Storage CAES Compressed Air Hydrogen Storage PEMFC Polymer Electrolyte Membrane Fuel Fell AFC Alkaline Fuel Cell
5
1
Introduction
The share that renewable energy sources (RES) play in the world’s electricity production is rising. Renewable energy sources are gradually replacing fossil fuels because RES is carbon-free and are supported by national laws (Rashid, 2015). Meanwhile, the world’s electricity demand is increasing due to new sources of demand (e.g., plug-in hybrid vehicles). These new trends have a great impact on electricity grid. The grid needs to integrate the large amount of unpredictable electricity production from RES (like solar and wind) and simultaneously deliver large amounts of electricity (Roberts and Sandberg, 2011). This leads to grid-capacity issues. In some instances, the grid is overloaded and cannot integrate the electricity which has been produced. In other instances, it is under loaded and cannot deliver the electricity demand. In the Netherlands, energy-distribution companies have trouble finding solutions to cope with these issues. Expanding the electricity grid would take years, and it is very expensive (Enexis, 2018). An alternative solution is to store and generate electricity using hydrogen. Energy-storage systems using hydrogen have developed interest in the past few years. They provide long-term storage which can be used to manage intermittent supply and demand (Böttcher et al., 2017). Furthermore, when produced from RES, hydrogen is a low-emission fuel, which is good for the development of a decarbonized energy system. Apart from storing and generating electricity, hydrogen can be used in other applications. For example, hydrogen can be used as an energy carrier to transport and distribute energy (Amid, Mignard, and Wilkinson, 2016).
6
Studies have been conducted on the hydrogen-storage system. Ghenai and Bettayeb (2019) showed the costs ($92/MWh of energy) of a hydrogen-storage system independent of the grid used to deliver the electrical demand of a university building. In addition, Almansoori and Betancourt-Torcat (2016) built a mixed-integer linear-programming model (MILP) to determine the optimal configurations of a hydrogen-storage system used to supply hydrogen. However, there is no literature about the cost and efficiency of a hydrogen-storage system use to deliver electricity to the grid. Hydrogen-storage systems can help manage intermittency in supply and demand in an electricity grid and facilitate the integration of RES. Therefore, this present paper studies the cost and efficiency performance of an RES-based hydrogen-storage system connected to the electricity grid. The paper provides insight into optimal configurations of the hydrogen technologies used to produce, store and reconvert hydrogen into electricity. In addition, the paper provides an outlook to network operator regarding the viability of integrating such a system in the electricity grid.
The mathematical model in this paper takes its basis from the MILP network-optimization model of Won et al. (2017). They provide a model which gives the optimal configurations of a hydrogen supply system to minimize costs of the system. However, they only consider hydrogen production and storage but not hydrogen regeneration into electricity. Sanchez et al. (2014) provides a model that gives the optimal sizing of RES-based hydrogen-storage system independent of the grid. The model of this paper is a combination of the model of Won et al. (2017) and Sanchez et al. (2014) which considers hydrogen production, storage and regeneration into electricity, and which is connected to the grid. The objective of the model is to minimize the costs of the system and maximize its efficiency. It will optimize the hydrogen technology types and the amount of hydrogen that is produced, stored and regenerated into electricity.
7
2
Literature Review
This paper is about the configuration and performance of an RES-based hydrogen-storage system used to facilitate the integration of RES into the electricity grid. The hydrogen-storage system is supplied by renewable energy, and it is comprised of three hydrogen technologies: hydrogen-production technology, hydrogen-storage technology and hydrogen-regeneration technology. The first Chapter of this chapter considers the challenges and opportunities of integrating RES into the current electricity grid. The second Chapter of this chapter reviews the literature on storage systems. To provide a holistic view of the RES- based hydrogen-storage system in this paper, a model of the system is illustrated in Figure 2.1. The last Chapter discusses the contribution this paper makes to the literature
2.1 Challenges and opportunities of integrating renewable energies into the electricity grid
8
A promising way to store and generate electricity is to use hydrogen as an energy carrier. The conversion of electricity into hydrogen offers long-term, carbon-free energy storage (Kim and Kim, 2017). When the electricity grid is overloaded, the surplus of electricity can be converted into hydrogen and stored. When the electricity grid is underload, electricity shortages can be compensated for and released by reconverting the hydrogen into electricity (Böttcher et al., 2017). However, there are limits to implementing this process. The first limit is the low so-called round-trip efficiency (Lund, Lindgren, Mikkola, and Salpakari, 2015). Round-trip efficiency is defined as ‘the ability of the hydrogen technologies to convert electrical energy to hydrogen and the hydrogen back to electrical energy’ (Bernier et al., 2005). The second limit is the economic performance of such a process. Bernier et al. (2005) have argued that economic performance is strongly related to round-trip efficiency. We have studied the efficiency and cost performance of the round-trip efficiency of the process (i.e., an RES-based hydrogen-storage system) and, thus, the integration of renewable energies into the current electric grid. 2.2 Hydrogen-storage system
Production of hydrogen. The electricity generated by the RES is used for the production of hydrogen by using an electrolyser. The electrolyser technologies used in hydrogen-storage systems have been already studied. Won et al. (2017) studied the costs of a hydrogen-storage system used to supply hydrogen with an alkaline electrolyser (AEC). The authors showed that the operational costs of the electrolyser account for 20% of the total operational costs of the system. Zeng and Zhang (2010) studied the historical development and innovations of AEC. These authors showed that AEC is a simple technology and is currently less efficient than other types such as polymer electrolyte membrane electrolysers (PEMEC) and an electrolyser with a solid-oxide membrane (SOEC). The AEC is most widely used for large-scale production of hydrogen and has low capital cost, whereas the PEMEC is most widely used for small-scale production of hydrogen and has high capital costs (Kotowicz et al., 2017). Regarding SOEC, it is not yet commercialized but has been shown to be capable of high efficiency and low capital costs and therefore has large opportunities of development (Schmidt et al., 2017).
9
hydrogen is the preferred storage option when storing large quantities of hydrogen. More recently, Reub et al. (2017) showed that underground storage solutions, such as depleted gas fields and salt caverns, offer an economical solution to the storage of large amounts of hydrogen. Finally, in the case of cooperation with the electricity grid, the most favourable hydrogen-storage method is a pumped hydroelectric storage (PHS) in which hydrogen is stored as a liquid. Another favourable method is compressed-air hydrogen storage (CAES), in which hydrogen is stored as a gas (Kotowicz et al., 2017)).
Hydrogen regeneration into electricity. The hydrogen stored can either be used as a fuel or reconverted into electricity in a fuel cell. The fuel cell is one of the most promising technologies for producing electricity via hydrogen (U.S. Department of Energy). Ghenai and Bettayeb (2019) have demonstrated the economic feasibility ($92/MWh of hydrogen) of a hydrogen-storage system independent of the grid used to meet the electrical demand of a university building. In their study, the hydrogen was reconverted into electricity with a polymer electrolyte-membrane fuel cell (PEMFC) and delivered to the university building when needed. Similarly, Sanchez et al. (2014) analysed the costs of a hydrogen-storage system used to meet the electrical demand of a residential house (total annualized costs of their system is $3,654,580 which has been shown to be economically feasible). In addition, Baghaee et al. (2016) studied a hydrogen-storage system used to deliver electricity to a micro grid using a PEMF. The authors built a multi-objective model which aims to optimize the sizes of the hydrogen technologies to minimize the costs and maximize the efficiency of the hydrogen-storage system. Finally, Mekhilef et al. (2012) offer a comparison of different fuel-cell technologies: alkaline fuel cell (AFC), phosphoric-acid fuel cell (PAFC), solid-oxide fuel cell (SOFC) and molten-carbonate fuel cell (MCFC). The AFC shows the highest electrical efficiency (60%), but it is applied only for military applications because the electricity production needed is small in this domain. The PAFC, SOFC and MCFC have received interest because of their low operating temperatures and high efficiency for large electricity generation (U.S Department of Energy).
2.3 Contribution to the literature
10
hydrogen-storage system can be used to facilitate the integration of renewable energies into the electricity grid. In the following of the document a mathematical model is built and applied on two case studies in the Netherlands. The model provides insights into the optimal configurations of the hydrogen-storage system and considers which technology should be used (i.e., hydrogen technologies type and size) and which quantities of hydrogen should be stored (i.e., hydrogen technologies operations) to maximize the performance (in terms of costs and efficiency) of the RES-hydrogen-storage system.
11
3
Methodology
This chapter presents the mathematical model used to answer the study question. The model analyses the optimal configurations of hydrogen technologies (hydrogen production, hydrogen storage, hydrogen reconversion) in the RES-based hydrogen-storage system. First, we describe the model of the system based on the literature, and we provide information about the development of the mathematical model. Thereafter, we give the assumptions made for building the mathematical model and we explain the mathematical model.
3.1 Model development
In this section, we explain how we developed the model based on the literature reviewed previously and the observation in the reality. After that, we present the mathematical equations that have been combined to build the mathematical model. Finally, we explain how the model is run and validated.
12
fuel cell (PEMFC) and a molten-carbonate fuel cell (MCFC). They are consider because they are the most used for large-scale production of hydrogen.
For the purpose of this study, a combination of empirical mathematical equations from two sources have been used to obtain a mathematical model that considers the three hydrogen technologies in the RES-based hydrogen-storage. Won et al. (2017) provide the equations for hydrogen production and hydrogen storage in their paper. In this paper, we also consider hydrogen re-generation into electricity. Sanchez et al. (2014) provide an equation for hydrogen re-generation into electricity. In their paper, the electricity regenerated is used to meet the demand of an external load to the electricity grid. Therefore, in this paper we combine the equations of hydrogen production and hydrogen storage provided by Won et al. (2017) with the equation of hydrogen regeneration provided by Sanchez et al. (2014).
The model is applied to two case studies in the Netherlands in order to validate it and provide insight for its application in the reality. The model is formulated as a mixed-integer linear-programming (MILP) model and is solved using Python. Python is an open-source programming language in which constraints and parameters can be changed easily. The script of the program is written such that the input parameters of the model (i.e., technical, cost parameters of the technologies and supply-and-demand parameters) can be adapted to each scenario. For each scenario, the script will return the amount of hydrogen produce, store and reconvert into electricity and the type of each hydrogen technology to use to minimize cost and maximize efficiency.
13 3.2 Mathematical modelling
In this Chapter, we give the assumptions on which the mathematical model is build. Thereafter, we explain the mathematical model used to answer the study question: ‘What are the optimal configurations of the grid-connected system to maximize its performance?’
3.2.1 Assumptions
The mathematical model is built on the following assumptions: 1) the demand and supply of electricity in the hydrogen-storage system are known, 2) the entire amount of electricity produced by the renewable energy sources is supplied into the hydrogen-storage system, 3) the size of the electrolyser in kW is determined by the monthly peak of electricity input it needs to be able to handle, and 4) the size of the fuel cell in kW is determined by the monthly peak of hydrogen it needs to convert back into electricity.
Set notation
The model includes three types of electrolyser j (AEC, PEMEC and SOEC), two types of hydrogen-storage facilities k (salt cavern and depleted gas field) and three types of fuel cells l (PAFC, SOFC and MCFC).
-j ∈ J set of electrolysers: J1, J2, J3 stand for AEC, PEMEC and SOEC
respectively.
-k ∈ K set of hydrogen-storage facilities: K1 stands for salt cavern and
K2 stands for depleted gas field.
-l ∈ L set of fuel cells: L1, L2, L3 stand for PAFC, SOFC and MCFC
re-spectively.
-t ∈ T set of time periods; t=1,...,n.
Decision variables
The objective is to maximize the performance of the hydrogen-storage sys-tem. It is done by optimizing three decision variables: the amount of hydro-gen produce by electrolyser j in time period t (HPjt), the amount of hydrogen
store by storage facility k in time period t (HSkt), the amount of hydrogen
regenerate into electricity by fuel cell l in time period t (HRlt) and the binary
variables (Xj, Yk, Jl) to determine which type of electrolyser, storage facility
and fuel cell are used in the system.
-HPjt = Amount of hydrogen produce by electrolyser j in time period t.
-HSkt = Amount of hydrogen store by storage facility k in time period t.
-HRlt = Amount of hydrogen regenerate into electricity by fuel cell l in time
period t.
-Xj, Yk, Zl are binary variables:
. –Xj, Yk, Zl = 1 if hydrogen technologies: electrolyser j, storage
facil-ity k, and fuel cell l respectively are used in the system. . –Xj, Yk, Jl = 0 otherwise.
Parameters
First, the model considers the technical parameters of each electrolyser and fuel cell type. Technical parameters include efficiency and size. Second, the model considers the cost parameters of each electrolyser, hydrogen-storage facility and fuel cell type. Cost parameters include operational and capital costs. Finally, the model considers the electricity supply and demand in the hydrogen-storage system. Electricity supply is the amount of electricity pro-duce by the renewable energies in time period t. Electricity demand is the amount of electricity that the grid is unable to deliver.
Technical parameters
-Szj = Size of electrolyser j.
-Szl = Size of fuel cell l.
-nj = Efficiency of electrolyser j.
-nk = Efficiency of storage facility k.
-nl = Efficiency of fuel cell l.
-Bst = Hydrogen-storage level at the beginning of the year. Costs parameters
-Capexj = Capital cost of electrolyser j.
-Capexk = Capital cost of storage facility k.
-Capexl = Capital cost of fuel cell l.
-OCj = Operational cost of electrolyser j.
-OCk = Operational cost of storage facility k.
-OCl = Operational cost of fuel cell l.
Electricity supply and demand -Eex
t = Electricity supply: Amount of electricity produce by the renewable
energy sources. -Erel
t = Electricity demand: Amount of electricity that the electricity grid is
unable to deliver in time period t.
Constraints
X1+ X2+ X3 = 1 (6)
Y1 + Y2 = 1 (7)
Z1+ Z2+ Z3 = 1 (8)
HPjt, HSkt, HRlt≥ 0 ∀j ∈ J, k ∈ K, l ∈ L t ∈ T (9)
Objective functions
The model will use two objective functions in order to provide insights on the two performance indicators studied; cost and efficiency. Thus, the model will be run two times (one time for each objective function) for each sce-nario. The first objective is to minimize the Total Annualized Cost of System (TACS). TACS is the sum of the capital cost and the annualized operational cost. The second objective function is to maximize the round-trip efficiency of the hydrogen storage system in order to avoid energy losses. Based on the definition provided in the literature (c.f., Chapter 2.1), it is calculated by multiplying the efficiency of the electrolyser with the efficiency of the fuel cell. 1.Minimization of Total Annualized Cost
MIN TACS = Ccap+ COM
Ccap= X j CapexjXjSzj + X k CapexkYk+ X l CapexlZlSzl COM = X j X t OCjHPjtXj+ + X j X t OCkHSktYk+ + X j X t OClHRltZl
2.Maximization of round-trip efficiency MAX round-trip efficiency = Efj × Efl
Efficiency electrolyser = Efj = X
j
njXj
Efficiency fuel cell = Efl = X
l
nlXl
18
4
Case Study 1: Study of a hydrogen-storage system supplied by wind
energy in Gelderland
This chapter presents the first case study that is used as input into the mathematical model. An illustration of the case can be found in Appendix A. We used the mathematical model presented in Chapter 3 to analyse the performance of a wind turbine-based hydrogen-storage system which delivers electricity to the grid in Arnhem (region of Gelderland). We performed this case study because there is a large amount of greenhouse farms whose energy demand has risen exponentially these recent years. Liander, the Dutch network operator (DNO) in the region, is not always able to meet this demand. In some instances, the grid is under loaded and it can no longer supply the greenhouses consumption in Gelderland. Furthermore, there are wind turbines available and the municipally is planning to install more in the region (Provincie Gelderland Beleidslijn Windenergie, 2018). Liander is looking for solutions to facilitate the integration of the electricity produce by the wind turbines into the electricity grid (WES, 2019). In this case, the electricity produce by the wind turbines is stored as hydrogen in one of the existing hydrogen storage facility in Bergermeer or Zuidzendig. This hydrogen stored is converted to electricity and deliver to the electricity grid when it can no longer deliver electricity to the greenhouses. Thus, we will analyse the optimal configuration and performance of the hydrogen-storage system for different scenarios, taking account an increase of the demand (more greenhouses) and supply (more wind turbines). In this part, we describe the simulations and scenarios for this case, finally the input parameters in the mathematical model are given next.
4.1 Scenario details
Three simulations (I to III in Table 4.1) were run for this case. The first one simulates the current electricity demand in Gelderland with 10 greenhouses in use. The second and third simulate the electricity demand in Gelderland for 30 and 90 greenhouses (see Lane 1 in Table 4.1). The purpose of these simulations is to analyse the effects of increasing electricity demand on the optimal configurations, the storage level and the costs of the hydrogen-storage system.
hydrogen-19
storage system. The second and the third consider the electricity generation of 10 and 20 wind turbines, as the municipality is aiming to build northeast wind turbines (see Lane 4 in Table 4.1). The purpose of these scenarios is to analyse the effects of increasing electricity supply on the optimal configurations, the storage level and the costs of the hydrogen-storage system.
4.2 Electricity demand and supply
Regarding electricity demand, a greenhouse consumes 133 MWh per month (average value from van der Velden and Smit, 2018). It is assumed that 10% of each greenhouse’s electricity demand cannot be delivered by the grid per month and will therefore be delivered by the hydrogen-storage system. For the three simulations (demand of 10, 30 and 90 wind turbines), the hydrogen-storage system needs to release 133 MWh, 400 MWh and 1,200 MWh per month, respectively (Line 2 in Table 4.1).
Regarding the electricity supply into the hydrogen-storage system, the electricity generation of a standard wind turbine (Nordex N50/80) was used. Its technical characteristics are given in Appendix B. The electricity generated by the wind turbines per month was calculated with an online calculator provided by the Danish Wind Industry Association. In the present study, a wind turbine of 1 MW produces an average of 141 MWh per month. Thus, five, 10 and 20 wind turbines are expected to produce 703 MWh, 1,411 MWh and 2,812 MWh per month, respectively (line 5, Table 4.1).
4.3 Technical parameters of hydrogen technologies 4.3.1 Size of hydrogen technologies
20
In the same manner, we calculate that, for the electricity demands of 10 (133 MWh/month), 30 (400 MWh/month) and 90 (1,200 MWh/months) greenhouses, the fuel-cell sizes required are 0.3 (133*0.25/120), one (400*0.25/120) and five (1200*0.25/120) MW, respectively (Lane 3 in Table 4.1).
4.3.2 Efficiency of hydrogen technologies
Efficiencies of electrolysers and fuel cells depend on their type. The efficiencies required to produce hydrogen from electricity by the three types of electrolysers whose costs are studied thereafter (see section 4.4) are provided by IRENA. Efficiencies obtained from 2017 (Technology Outlook for The Energy Transition, 2018) are used as inputs for the model. The efficiencies of the three types of fuel cells whose costs are studied thereafter (see Section 4.4) to convert hydrogen into electricity are retrieved from Mekhilef et al. (2012). We can see in Table 4.2 that a combination of an electrolyser SOEC (efficiency of 70%) and a fuel cell MCFC (efficiency of 47%) will maximize the round-trip efficiency of the system.
4.3.3 Hydrogen-storage starting level
21
Table 4.1 Simulations and scenarios analysis for Case Study 1
Simulations: I II III
Scenarios: I.1 I.2 I.3 II.1 II.2 II.3 III.
1 III.2 III.3 1 D eman d number of greenhouses
10
30
90
2 Electricity demand (in MWh / month)13.3
400
1,200
3 Fuel cell size
22
Electrolyser Fuel cell
AEC PEMEC SOEC PAFC SOFC MCFC
Efficiency (%) 68 64 70 40 43 47
Table 4.2 Efficiencies of electrolysers and fuel cells in function of their type.
4.4 Costs of hydrogen technologies (including systems purchasing, hydrogen production by electrolyser, hydrogen storage and electricity production from hydrogen) In the present study, three types of electrolysers are studied: AEC, PEMEC and SOEC. The costs of electrolysers depend on their type and size. The capital cost for AEC and PEMEC were retrieved from the final report of the Industry Cluster Flanders (Roadmap for Flanders, 2016) and from Hendriksen et al. (2013). Operational costs for AEC and PEMEC were retrieved from Schmidt et al. (2017). Finally, data for capital and operational costs for SOEC were retrieved from Mathiesen et al. (2013). Table 4.3. below shows the cost of each electrolyser depending on its size for each scenario in the three simulations (Line 3 in Table 4.1).
Three types of fuel cells are studied: PAFC, SOFC and MCFC. Capital and operational costs for the PAFC and MCFC were retrieved from the U.S. Department of Energy (Combined Heat and Power Technology Fact Sheet Series, 2016). Capital and operational costs for the SOFC were retrieved from a report of Energinet, which is the Danish National Transmission Operator (Technology Data for Energy Plant, 2012). Table 4.3 below shows the cost of each fuel cell depending of its size for each scenario in the three simulations (see Lane 6 in Table 4.1).
23
the total operational costs of the salt cavern are 0.86 (0.005*170 + 0.0045) €/kWh of hydrogen stored. Table 4.4 shows the cost of each type of hydrogen storage used in the model.
Capital cost (€/kW/system) Operational cost (€/kWh of hydrogen produced)
Electrolyser size (MW)
AEC PEMEC SOEC AEC PEMEC SOEC
2 660 1,015 855 2.2 3.4 2.1 3 640 800 780 6 630 760 640 Fuel Cell size (MW)
Capital cost (€/kW/system) Operational cost (€/kWh of electricity produced)
PAFC SOFC MCFC PAFC SOFC MCFC
0.3 10,000 15,000 10,000 6.5 16 4.5
1 5,500 3,500 4,600 4 10 4
5 5,000 2,000 4,000 2 10 3
Table 4.3 Electrolysers and fuel cells costs in function of their type and size in Case 1.
Capital cost (M€) Capital cost (M€) Depleted gas field 700 0.75 Salt Cavern 40 0.86
24
5
Results of Case Study 1
This chapter presents the results of the first case study based on the information presented above. The aim of Case 1 is to study the application of the wind turbines-based hydrogen-storage system in Gelderland. The system is used to store the electricity generated by wind turbines as hydrogen and convert it to electricity when the grid cannot deliver the electricity consumption of the greenhouses. The objective is to provide insight to our research question: ‘What configurations of an RES-based hydrogen-storage system connected to the electricity grid would maximize its performance?’. Therefore, we present the optimal configurations (electrolyser type and fuel cell type) of the wind turbine-based hydrogen-storage system when increasing supply (number of wind turbines) and demand (number of greenhouses). Then, we analyse the effects of increasing supply or demand on the hydrogen-storage level. Finally, we analyse the effects of increasing supply or demand on the costs of the hydrogen-storage system.
5.1 Optimal configurations of the hydrogen-storage system when minimizing the costs and maximizing efficiency
In this part, we analyse the optimal configurations of the hydrogen-storage system when the electricity supply increases for different electricity demand. Therefore, optimal configurations of the hydrogen-storage system which allow for the minimization of annualized cost and maximization of efficiency of the system are presented. Furthermore, the energy losses due to minimization of system costs rather maximization of system efficiency are given. Raw data generated by the mathematical model is presented in Appendix C for each simulation (I.1 to III.3; Tables C.1 to C.9).
25
the system when combining these technologies. Indeed, In simulation I, only the hydrogen-storage system supplied by 20 wind turbines maximizes efficiency while minimizing cost because the same amount of hydrogen is left in the storage at the end of the year (20,212 MWh) (Figure 5.1). If the hydrogen-storage system is supplied by five and 10 wind turbines, there are hydrogen losses when the objective is to minimize costs. The energy losses are of 169 (2,500-2,331) and 538 (8,404-8,066) MWh, respectively, during hydrogen production. This is due to a slightly lower efficiency of AEC than SOEC (Table 4.4).
Simulations II.1, II.2 and II.3 analyse the optimal configurations of the hydrogen-storage system when there is a demand of 30 greenhouses and electricity supplied into the hydrogen-storage system increases (supplied by 5, 10 or 20 wind turbines). In the three scenarios of simulation II, as in Scenario I and III, only the hydrogen-storage system supplied by 20 wind turbines has the highest efficiency when minimizing costs (Table 4.4). Indeed, there is no energy loss in the hydrogen storage at the end of the year (13,402 MWh). However, 169 (691-522) MWh and 337 (1,594-1,257) MWh of hydrogen is lost when the hydrogen-storage system is supplied by the electricity generated by 5 and 10 wind turbines, respectively (figure 7.2). As before, this may be due to the hydrogen lost during hydrogen production, as AEC is less efficient than SOEC.
26 Scenarios 1 2 3 Electricity supply (number of wind turbines) 5 10 20 Simulations Electricity demand (number of greenhouses) Electrolyser size (MW): 2 MW 3 MW 6 MW
Fuel Cell size (MW):
I 10 0.3 MW AEC - MCFC AEC - MCFC SOEC - MCFC
II 30 1 MW AEC - MCFC AEC - MCFC SOEC - MCFC
III 90 5 MW AEC - MCFC AEC - MCFC SOEC - MCFC
Table 5.1 Optimal configurations of the hydrogen-storage system in each simulation to minimize total annualized costs.
5.2 Hydrogen-storage level
In this part, we analyse the evolution of the hydrogen-storage level when the electricity supply increases while the demand stays the same in the storage system. First, the hydrogen-storage system is used only to store energy when there is an electricity demand of 10 greenhouses. In this simulation, the hydrogen storage is filled along the year, and its level is of 2,500 MWh, 8,404 MWh and 2,021 MWh when the electricity generation of 5, 10 and 50 wind turbines respectively is supplied to the system (Figure 5.1).
Next, the hydrogen-storage system is mostly used to deliver electricity to the grid when the electricity demand is for 30 greenhouses and the electricity in the system is supplied by the electricity generation of five wind turbines. We can see from Figure 5.2 that the hydrogen-starting level needs to be filled at the beginning of the year for the system to be able to deliver the electricity demand. Second, the hydrogen-storage level is lower at the end of the year (522 MWh when the objective is to minimize cost) than at the beginning (5,000 MWh). However, the hydrogen-storage system which is supplied by either the electricity generated by 10 and 20 wind turbines has been used mostly used to produce and store hydrogen, as there is more hydrogen left in storage at the end of the year than there was in the starting hydrogen-storage level (1,257 MWh and 13,402 MWh, respectively, when minimizing costs in Figure 5.2).
27
greenhouses is very high. At the end of the year, there is almost no electricity left in the hydrogen storage when it is supplied by five wind turbines (97 MWh when the objective is to minimize costs of the system and 265 MWh when the objective is to maximize efficiency). There is more hydrogen left in storage when the electricity supplied by the wind turbines increases to 10 and 20 wind turbines. However, in both cases, the hydrogen-storage level is lower than the hydrogen-starting level, which means that the hydrogen-storage system has been mostly used to deliver electricity to the grid (in Figure 5.3).
Figure 5.1 Hydrogen-storage level at the end of the year for an electricity supply in the hydrogen-storage system of 5, 10 and 20 wind turbines and a demand of 10 greenhouses.
2331 8066 20212 2500 8404 20212 0 5000 10000 15000 20000 25000 5 WT (S1.1) 10 WT (S1.2) 20 WT (S1.3)
Hy
dr
og
en
st
or
ag
e
(M
W
h)
28
Figure 5.2 Hydrogen-storage level at the end of the year for an electricity supply in the hydrogen-storage system of 5, 10 and 20 wind turbines and a demand of 30 greenhouses.
Figure 5.3 Hydrogen-storage level at the end of the year for an electricity supply in the hydrogen-storage system of 5, 10 and 20 wind turbines and a demand of 90 greenhouses.
5000 0 0 522 1257 13402 691 1594 13402 0 2000 4000 6000 8000 10000 12000 14000 16000 5 WT (S2.1) 10 WT (S2.2) 20 WT (S2.3)
Hy
dr
og
en
st
or
ag
e
(M
W
h)
Hydrogen storage starting level Minimization of costs Maximization of efficiency
25000 20000 10000 97 831 2977 265 1169 2977 0 5000 10000 15000 20000 25000 30000 5 WT (S3.1) 10 WT (S3.2) 20 WT (S3.3) H yd ro ge n st or ag e (MW h)
29
5.3 Cost analysis of the hydrogen-storage system
In this section, we first analyse the effects of increasing electricity supply on the costs of the hydrogen-storage system. Then, we analyse the effects of increasing the electricity demand on the costs of the hydrogen-storage system.
30
Figure 5.4 Cost breakdown of the hydrogen-storage system for a demand of 10 greenhouses and a supply of 5, 10 and 20 wind turbines.
31
note that the capital costs of the fuel cell strongly increase for larger fuel cells (3 M€ for a fuel cell of 0.3 MW and 20 M€ for a fuel cell of 5 MW).
32
6
Case 2: Study of a hydrogen-storage system supplied by solar energy
in Groningen
This chapter presents the second case study that is used as input in the mathematical model. An illustration of the case can be found in Appendix D. We used the mathematical model presented in Chapter 3 to analyse the performance of a solar panel-based hydrogen-storage system which delivers electricity to the grid in Groningen. We performed this case study because incentives from the Dutch government regarding solar panels have led to exponential installations in this region. However, grid capacity is too small to handle the growing electricity production and the current grid is overloaded in some instances. These grid-capacity problems in the region become a threat to solar-energy development. Enexis, the local DNO, has to refuse or hold the connection of new solar-panel installations to the grid (Enexis Annual Report, 2018). In this case, the excess of electricity is stored in one of the existing hydrogen storage facility in Bergermeer or Zuidwendig. Furthermore, in some instances the grid is under loaded and it can no longer supply the electricity consumption in Groningen. Therefore, the shortage of electricity is retrieved from the hydrogen storage facilities when the grid can no longer supply the electricity consumption needed in Groningen. In this part, we describe the simulations and scenarios for this case, finally the input parameters in the mathematical model are given.
6.1 Scenario details
Two simulations (I and II in Table 6.1) are run for this case. The first one simulates an electricity demand considered to be low. The second simulates an electricity demand considered to be high (see Line 1 in Table 6.1). The purpose of these simulations is to analyse the effects of increasing electricity demand on the optimal configurations, the hydrogen-storage level and the costs of the hydrogen-storage system. In 2016, total annual consumption was 30,398 MWh in Groningen (which is considered low), and in 2018 the total annual consumption was 32,767 MWh in Groningen (which is considered high).
33
increasing electricity supply into the hydrogen-storage system on the optimal configurations, the hydrogen-storage level and the costs of the hydrogen-storage system.
6.2 Electricity supply and demand
To simulate the electricity demand that has to be delivered by the hydrogen-storage system, we contacted the team manager of System Operations at Enexis to gather the monthly electricity consumption in 2016 and 2018 in Groningen (Appendix E). Furthermore, he advised us to put a limit on the grid capacity of 2,400 MWh per month, above which the grid cannot deliver the whole electricity consumption of Groningen such that the hydrogen-storage system is required to deliver electricity.
Regarding the electricity generation of houses with solar panels in Groningen, there are 8,500 houses with solar panels installed in Groningen (Central Agency for Statistics, 2018). On average, each house has 12 solar panels with a capacity of 280 Wp (Solar Solution Int., 2018); thus, the total solar capacity installed on houses in Groningen is 28 MWp (8,500*12*280). Next, the solar park in Sappemeer has a solar capacity of 103 MWp (Powerfield, 2018) (see Appendix F). To simulate electricity supply by the solar panels into the hydrogen-storage system, the Photovoltaic Geographical Information System (PVGIS) tool from the European Commission was used. This tool gives monthly, location-dependent electricity-generation data for solar panels.
34
will keep increasing along the year, leading to increases in storage costs. Finally, we can observe that, with a solar capacity of 103 MWp, there is enough electricity produced by the solar park to meet the electricity demand in the hydrogen-storage system from January and February and from September to December, which is not the case with a solar capacity of 28 MWp. Therefore, the starting hydrogen level for each scenario of this case is set to 6 MWh to make sure that the hydrogen-storage system is always able to answer the demand over the year.
Figure 6.1 Electricity consumption in 2018 and electricity generation of the two types of solar capacity (28 and 103 MWp) in Groningen.
6.3 Technical parameters of hydrogen technologies 6.3.1 Size of hydrogen technologies
The sizes of the electrolysers and the fuel cells are based on their capacity (Hydrogen from Renewable Power, 2018). In our scenarios, in which the whole electricity produced by solar panels is converted into hydrogen for storage, the electrolyser capacity needed is based on the month during which the highest amount of electricity is produced. The electrolysers’ capacities were calculated in the same way as for Case Study 1. The maximum electricity that is generated by a solar capacity of 28 MWp is 3,581 MWh in May (c.f., Appendix E); thus, the electrolyser capacity required is 10 MW (3,581*0.25/120). The maximum electricity that is generated by a solar capacity of 103 MWp is 7,012 MWh in August (c.f., Appendix E); thus the electrolyser capacity required is 15 MW (7,012*0.25/120) (c.f., Line 6 in Table 6.1).
0,0 1000,0 2000,0 3000,0 4000,0 5000,0 6000,0 7000,0 8000,0 Janua ry Febr
uary March April May June July August
Septe mber Octob er Nove mber Dece mber MW h
35
We also calculated the capacity required of the fuel cell in 2016 and 2018. In 2016, the maximum electricity demand was 803 MWh in November; thus, the fuel-cell size required was 1.5 MW (803*0.25/120). In 2018, the maximum electricity demand was 1,356 MWh in December; thus, the fuel-cell size required was 3.5 (1356*0.25/120) MW (c.f., Line 3 in Table 6.1).
6.3.2 Efficiency of hydrogen technologies
The efficiencies of electrolysers and fuel cells depend on their type. The same efficiencies used for Case 1 for each electrolyser and fuel cell type is used for this case. An overview of the efficiencies can be found in Table 4.4 in Section 4. In this study, electrolyser SOEC (70%) and fuel-cell MCFC are the most efficient (47%) of electrolysers and fuel cells. Thus, a combination of electrolyser SOEC with fuel-cell MCFC should maximize the round-trip efficiency of the hydrogen-storage system.
Simulations: I (year 2016) II (year 2018)
Scenarios:
I.1 I.2 II.1 II.2
1
D
eman
d Demand: annual electricity consumption (in MWh) 30,398 32,767
2 Max electricity demand in a month (MWh) 803 1,356
3 Fuel cell size (MW) 1.5 3.5
4
Supply
Supply: Solar panel capacity (MWp) 28 103 28 103
5 Max electricity supply in a month (MWh) 3,581 7,012 3,581 7,012
6 Electrolyser size (MW) 10 15 10 15
7
S
tor
age
Hydrogen storage starting level in the beginning of the first year (MWh)
6,000
36
6.4 Costs of hydrogen technologies (including systems purchasing, hydrogen production by electrolyser, hydrogen storage and electricity production from hydrogen) The same types of electrolysers and fuel cells as were used for Case 1 are studied (AEC, PEMEC and SOEC). The costs of electrolysers and fuel cells depend on their type and size. Specific costs for each technology have been retrieved from the same sources as for Case 1 and are shown in Table 4.6 below. In addition, the same types of hydrogen-storage facilities as were used for Case 1 are studied (Table 6.3); thus, capital costs to store in the depleted gas field and salt cavern are the same as in Case 1. However, the operational costs for the depleted gas field are higher (0.8 €/kWh and 0.2 €/kWh, respectively) than for the salt cavern because the depleted gas field is much further away from the electricity grid than the salt cavern (see illustration in Appendix D); this distance increases transportation costs. Thus, the salt-cavern storage has both lower capital and operational costs.
Capital costs (€/kW for each system)
Operational costs (€/kWh of hydrogen produced) Electrolyser type/size
(MW) AEC PEMEC SOEC AEC PEMEC SOEC
10
580 740 350
1.9 2.4 0.9
15 1.7 2.1 0.6
Capital costs (€/kW for each system)
Operational costs (€/kWh of electricity produced) Fuel cell type/size
(MW) PAFC SOFC MCFC PAFC SOFC MCFC
1.5 690 1250 930 3 10 4
3.5 3,500 1,000 2,000 3 8 3
37 Capital costs (M€)
Operational costs (€/kWh of hydrogen stored) including transportation costs Depleted gas
field 700 0.80
Salt Cavern 40 0.20
38
7
Results of Case Study 2
This chapter presents the results of the second case study based on the information presented above. The aim of Case 2 is to study the application of the solar panels-based hydrogen-storage system in Groningen to facilitate the integration of the electricity generated by solar panels into the electricity grid. The system is used to store the electricity generated by solar panels as hydrogen and convert it to electricity when the grid cannot deliver the whole electricity consumption in Groningen. The objective is to provide insight to our research question: ‘What configurations of an RES-based hydrogen-storage system connected to the electricity grid would maximize its performance?’. Therefore, we analyse the optimal configurations (electrolyser size and type) of the solar panels-based hydrogen-storage system when the electricity supply (solar panel capacity) and demand increases (year 2016 and 2018). Then, we present the effects of intermittent supply and demand on the hydrogen-storage level over a year. Finally, we analyse the effects of increasing supply or demand on the costs of the hydrogen-storage system.
7.1 Optimal configurations of the hydrogen-storage system for minimizing costs and maximizing efficiency
In this part, we analyse the optimal configurations of the hydrogen-storage system when the electricity supply increases for different electricity demand (2016 and 2018). Therefore, optimal configurations of the hydrogen-storage system which allow for the minimization of annualized cost and maximization of efficiency of the system are presented. Furthermore, the energy losses due to minimization of system costs rather maximization of system efficiency are given. Raw data generated by the mathematical model is presented in Appendix H for each simulation (I.1 to II.2; Tables H.1 to H.9). Then, optimal configurations of the hydrogen-storage system which allow for the minimization of annualized cost are summarized in Table 7.1.
39
seen from the hydrogen-storage level in Figure 7.1, which is lower when minimizing costs than when maximizing efficiency along the year. The hydrogen losses are thus due to the fuel cell. Indeed, more hydrogen needs to be taken from the hydrogen storage and reconverted into electricity each month when using PAFC (efficiency=40%) than when using MCFC (efficiency=70%) to deliver the electricity demand.
Simulation II analyses the optimal configurations of the hydrogen-storage system in 2018 when the electricity supply into the hydrogen-storage system increases (28 MWp: houses with solar panels to 103 MWp: solar park). A combination of electrolyser SOEC (of 10 or 15 MW) with fuel-cell MCFC (of 3.5 MW) is optimal to minimize the total costs of the hydrogen-storage system when the electricity is supplied by either the houses with solar panels in Groningen (electrolyser SOEC of 10 MW) and when it is supplied by the solar park (electrolyser SOEC of 15 MW). This configuration also maximizes the efficiency of the hydrogen-storage system. Thus, it can be seen in Figure 7.2. that the hydrogen-storage level is the same along the year when minimizing costs and when maximizing efficiency.
Scenarios 1 2
Electricity supply: solar capacity (MWp) 28 103 Simulations Electricity demand: Annual electricity consumption (MWh)
Electrolyser/Fuel Cell size (MW) 10 MW 15 MW I 30,398 (in 2016) 1.5 MW SOEC - PAFC SOEC - PAFC II 32,767 (in 2018) 3.5 MW SOEC - MCFC SOEC - MCFC Table 7.1 Optimal configurations of the hydrogen-storage systems in each scenario to
40 7.2 Hydrogen-storage level
In this part, we present the effects of intermittent supply and demand on the hydrogen-storage level over a year. First, Scenario I.1 analyses the evolution of the hydrogen-storage level over the year 2016 when the electricity generated by the houses with solar panels in Groningen (28 MWp) is supplied to the hydrogen-storage system every month. The electricity supplied into the hydrogen-storage system by the houses with solar panels in Groningen is not able to meet the electricity demand in the beginning of the year in 2016. Hydrogen is taken from the hydrogen-starting level (6,000 MWh) until March (Figure 7.6). However, after March, the hydrogen-storage level increases strongly because of the high electricity generation from the solar panels in the summer and the lower electricity consumption. In October, the hydrogen-storage level decreases slightly because less electricity is generated, and demand becomes higher due to the winter season. Finally, we can see that, at the end of the year, much hydrogen is left in storage (19,108 MWh when minimizing the costs of the hydrogen-storage system and 20,087 MWh when maximizing its efficiency). In addition, in comparison with Scenario I.1, Scenario I.2 analyses the evolution of the hydrogen-storage level for the same electricity demand (in 2016) but for a higher electricity generation supplied to the hydrogen-storage system (103 MWp). The electricity generated by the solar park is directly able to deliver the electricity demand in January, which leads to an increase in hydrogen stored from January. At the end of the year, the hydrogen-storage level is of 46,463 MWh (Figure 7.7).
Secondly, in comparison with Scenario I.1, Scenario II.1 analyses the evolution of the storage level for the same electricity generation which is supplied to the hydrogen-storage system (28 MWp) but for a higher electricity demand (in 2018, electricity consumption was higher than in 2016). Electricity is taken from the hydrogen-storage starting level until April, and the hydrogen storage is used less over the year (Figure 7.3). Additionally, we can see a stronger decrease in hydrogen storage from October to December in Figure 7.2 (2018) than in Figure 7.1 (2016). This due to the higher electricity consumption in 2018 than 2016 during the winter. There is still hydrogen left in storage at the end of the year but less than in 2016— again due to the higher electricity consumption in 2018 (14,960 MWh in 2018 compared to 19,108 MWh in 2016).
41
Figure 7.1 Hydrogen-storage level of the hydrogen-storage system per month in 2016 when it is supplied by the houses with solar panels in Groningen (28 MWp).
42
Figure 7.3 Hydrogen-storage level of the hydrogen-storage system per month in 2018 when it is supplied by the houses with solar panels in Groningen (28 MWp).
7.3 Costs of the hydrogen-storage system
In this part, we first analyse the effects of increasing electricity supply on the costs of the hydrogen-storage system when the demand is considered low (in 2016). Then, we analyse the effects of increasing the electricity supply on the costs of the hydrogen-storage system when the demand is high (in 2018).
43
both scenarios). However, the increase in electricity supply does have an effect on the total operational costs of the system. The operational costs of the electrolyser are lower than the operational costs of the fuel cell (17.7 M€ and 19.7 M€, respectively) when the hydrogen-storage system is supplied by the solar panels on houses (28 MWp). They become higher than those of the fuel cell when the hydrogen-storage system is supplied by the solar park (103 MWp) (41.4 M€ and 19.7 M€, respectively). This is because more electricity is generated and converted into hydrogen in the electrolyser while the demand stays the same. In addition, we note that storage costs increase when electricity supply increases (33.2 M€ for a solar capacity of 28 MWp to 67.9 M€ for a solar capacity of 103 MWp) because more hydrogen is stored (Figure 7.4).
Figure 7.4. Cost breakdown of the hydrogen-storage system in 2016 when it is supplied by houses with solar panels in Groningen (28 MWp) and the solar park (103 MWp).
44
the investment in the salt cavern is still the highest investment of the system (40 M€). We note that investment cost in the fuel cell is almost twice as large in 2018 as in 2016 (14.4 M€ and 8.3 M€, respectively). This is because the electricity demand is higher and a larger fuel cell is required to convert the hydrogen into electricity. In both scenarios, fuel-cell investment costs contribute more than electrolyser-investment costs to the total capital cost of the system. In 2018, the increase in electricity supply has an effect on the total operational costs of the system. The operational costs of the electrolyser are lower than the operational costs of the fuel cell when the hydrogen-storage system is supplied by the solar panels on houses (28 MWp). The operational costs of the electrolyser become higher than the operational costs of the fuel cell when the electricity supply increases, and this even with a high electricity demand (in 2018).
45
8
Discussion
In this section we discuss the results of case study 1 and case study 2 in light of the previous literature. To begin, we discuss the optimal configurations of electrolyser and fuel cell in the RES-based hydrogen storage system. We then consider the costs of the system.
An important finding of this paper indicates that the optimal configuration to maximize round-trip efficiency of a RES-based hydrogen storage system connected to the electricity grid is an SOEC with an MCFC, with round-trip efficiency of 33%. However, this configuration does not always minimize the total annual costs of the system. The optimal configurations to minimize annual costs depend on the amount of hydrogen produced and the amount of hydrogen in the hydrogen-storage system that is reconverted to electricity.
First, regarding hydrogen production, this paper shows that an SOEC has the lowest cost for large-scale hydrogen production (electrolyser size of 6 to 15 MW; see figures 6.1 and 7.1). This finding supports the expert view of Schmidt et al. (2017), who in projecting to 2030 concluded that, when used in a hydrogen storage system, an SOEC should be preferred to an AEC or a PEMEC due to an expected decrease of costs because of higher hydrogen production by 2030. Furthermore, Kotowicz et al. (2017) pointed out that a PEMEC is more suitable for small-scale hydrogen production in an RES-based hydrogen-storage system used to directly deliver hydrogen. By contrast, in the present paper we show that an AEC has the lowest cost for small-scale hydrogen production in an RES-based hydrogen-storage system used to deliver electricity to the grid (electrolyser size of 2 to 3 MW; see figure 6.1).
RES-46
based hydrogen storage system connected to the electricity grid remains a very expensive alternative.
Installing a hydrogen storage system supplied by the electricity generated by a solar park (103 MWp) and used to meet the need for electricity that cannot otherwise be delivered by the grid in Groningen would cost 196.9 M€ in 2018 (including investment costs and annualized operational costs). Sanchez et al. (2014) showed that a hydrogen storage system used to deliver electricity to a residential house would cost 3.6 M€. Thus, the present paper shows that a RES based hydrogen-storage system connected to an electricity grid is currently very expensive and not feasible. The most significant costs are the hydrogen storage costs of the system. There are three reasons for this.
47
48
9
Conclusion
This paper presents a method for evaluating the cost and efficiency of an RES-based hydrogen-storage system which delivers electricity to the grid. The method involves using a mixed integer linear programming model to answer the research question. The model includes hydrogen production in an electrolyser, hydrogen storage in a hydrogen storage facility and hydrogen re-generation into electricity in a fuel cell. The results offer insight into the optimal configurations of hydrogen technologies in hydrogen storage systems to facilitate the integration of RESs into the electricity grid. Furthermore, the paper provides information on the costs and efficiency of hydrogen-storage systems.
Overall, we find some evidence that hydrogen-storage systems for facilitating the integration of RESs in the electricity grid are very expensive and therefore are not a solution in the short term.
There are two main reasons for this. First, the results indicate that an SOEC with an MCFC maximizes round-trip efficiency of the hydrogen-storage system and minimizes its costs when there is a large amount of hydrogen produced and reconverted into electricity. However, the SOEC is not yet commercialized, and although it is a fast-developing technology, it will be several years before it is commercialized (Schmidt et al., 2017). During this time, however, the grid should be expanded. An alternative would be to use an AEC instead, which is slightly less efficient (68% instead of 70%) but is less expensive than an SOEC for producing smaller amounts of hydrogen (electrolyser size less than 6 MW). The second reason is the high costs of hydrogen storage. In the two case studies in this paper, storage is the major capital cost component (4M€ for a salt cavern) of the hydrogen-storage system. Furthermore, operational costs for storing hydrogen are also high, which lead to high costs when the electricity supply is greater than the demand and much hydrogen has to be stored.
49
50 REFERENCES
Agbossou, K., Kolhe, M., Hamelin, J., & Bose, T. K. (2004). Performance of a stand-alone renewable energy system based on energy storage as hydrogen. IEEE Transactions on Energy Conversion,19(3), 633-640.
Almansoori, A., & Betancourt-Torcat, A. (2016). Design of optimization model for a hydrogen supply chain under emission constraints-A case study of Germany. Energy, 111, 414-429. Ansari, N., & Lo, C. H. (2016). U.S. Patent No. 9,246,334. Washington, DC: U.S. Patent and Trademark Office.
Amid, A., Mignard, D., & Wilkinson, M. (2016). Seasonal storage of hydrogen in a depleted natural gas reservoir. International Journal of Hydrogen Energy, 41(12), 5549-5558.
Baghaee, H. R., Mirsalim, M., Gharehpetian, G. B., & Talebi, H. A. (2016). Reliability/cost-based multi-objective Pareto optimal design of stand-alone wind/PV/FC generation microgrid system. Energy, 115, 1022-1041.
Ball, M., & Weeda, M. (2015). The hydrogen economy–Vision or reality? International Journal of Hydrogen Energy, 40(25), 7903-7919.
Barbir, F. (2005). PEM electrolysis for production of hydrogen from renewable energy sources. Solar energy, 78(5), 661-669.
Barbir, F., Lillis, M., Mitlitsky, F., & Molter, T. (2002). Regenerative fuel cell applications and design options. In Proceedings of 14th World Hydrogen Energy Conference, Montreal, Canada. As referenced in Ref (Vol. 35).
Barthelemy, H., Weber, M., & Barbier, F. (2017). Hydrogen storage: Recent improvements and industrial perspectives. International Journal of Hydrogen Energy, 42(11), 7254-7262.
Bernier, E., Hamelin, J., Agbossou, K., & Bose, T. K. (2005). Electric round-trip efficiency of hydrogen and oxygen-based energy storage. International Journal of Hydrogen Energy, 30(2), 105-111.
Blanco, H., & Faaij, A. (2018). A review at the role of storage in energy systems with a focus on Power to Gas and long-term storage. Renewable and Sustainable Energy Reviews, 81, 1049- 1086.
51
Bukar, A. L., & Tan, C. W. (2019). A review on stand-alone photovoltaic-wind energy system with fuel cell: System optimization and energy management strategy. Journal of cleaner production.
Bünger, U., Michalski, J., Crotogino, F., & Kruck, O. (2016). Large-scale underground storage of hydrogen for the grid integration of renewable energy and other applications. In Compendium of Hydrogen Energy (pp. 133-163). Woodhead Publishing.
CBS Statline. (2020). Retrieved 20 January 2020, from
https://opendata.cbs.nl/statline/#/CBS/en/dataset/82905ENG/table?from statweb
Chauhan, A., & Saini, R. P. (2014). A review on Integrated Renewable Energy System based power generation for stand-alone applications: Configurations, storage options, sizing methodologies and control. Renewable and Sustainable Energy Reviews, 38, 99-120.
Energinet (2012). Technology data for energy plant: Generation of Electricity and District Heating, Energy Storage and Energy Carrier Generation and Conversion. Denmark.
Enexis. (2018). 2017 annual report. Retrieved 6 January 2020 https://www.enexisgroep.com/investor-relations/publications/
Enexis. (2019). 2018 annual report. Retrieved 6 January 2020 https://www.enexisgroep.com/investor-relations/publications/
Fuel Cells and Hydrogen Joint Undertaking. (2019). HEAVENN. European Comission Proposal Submission Form. Brussel, CA: Belgium.
Foresight Climate & Energy. (2019). Hydrogen: The Northern Netherlands is ready. Retrieved 18 October 2019, from https://foresightdk.com/hydrogen-northern-netherlands-is ready/?fbclid=IwAR0As8bMr8Y_7fwrFfOtkDnYRTUlT5zy3aF_A0FrQb20zvwmdnG1EpA0 wiA
Ghenai, C., & Bettayeb, M. (2019). Modelling and performance analysis of a stand-alone hybrid solar PV/Fuel Cell/Diesel Generator power system for university building. Energy, 171, 180-189.
Hwang, J. J. (2013). Sustainability study of hydrogen pathways for fuel cell vehicle applications. Renewable and Sustainable Energy Reviews, 19, 220-229.
52
Solar Solutions International. (2018). de pers record aantal zonnepanelen geinstalleerd 2017 Retrieved 26 January 2020, from https://www.solarsolutions.nl/nieuws-2018/de-pers-record-aantal-zonnepanelen-geinstalleerd-2017/
Karlsson, C. (Ed.). (2016). Research methods for operations management. Routledge.
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.
Kotowicz, J., Bartela, Ł., Węcel, D., & Dubiel, K. (2017). Hydrogen generator characteristics for storage of renewably generated energy. Energy, 118, 156-171.
Kroniger, D., & Madlener, R. (2014). Hydrogen storage for wind parks: A real options evaluation for an optimal investment in more flexibility. Applied energy, 136, 931-946.
Lof, G. (2020). Power shortages in Dutch greenhouse horticulturists: Waiting another five years for a solution? | WES, Wind Energy Solutions. Retrieved 26 January 2020, from
https://windenergysolutions.nl/uncategorized/power-shortages-in-dutch-greenhouse-horticulturists-waiting-another-five-years-for-a-solution/.
Lund, P. D., Lindgren, J., Mikkola, J., & Salpakari, J. (2015). Review of energy system flexibility measures to enable high levels of variable renewable electricity. Renewable and Sustainable Energy Reviews, 45, 785-807.
Mekhilef, S., Saidur, R., & Safari, A. (2012). Comparative study of different fuel cell technologies. Renewable and Sustainable Energy Reviews, 16(1), 981-989.
Office Of Energy Efficiency And Renewable Energy (July 28, 2015). Hydrogen Storage. Retrieved 15 December 2019, from https://www.energy.gov/eere/fuelcells/hydrogen-storage Rashid, M. H. (2015). Electric Renewable Energy Systems. Academic Press.
Roberts, B. P., & Sandberg, C. (2011). The role of energy storage in development of smart grids. Proceedings of the IEEE, 99(6), 1139-1144.
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.
53
Schmidt, O., Gambhir, A., Staffell, I., Hawkes, A., Nelson, J., & Few, S. (2017). Future cost and performance of water electrolysis: An expert elicitation study. International journal of hydrogen energy, 42(52), 30470-3049.
Thomas, D., (Hydrogenics), Mertens, D., (Colruyt), Meeus, M., (Sustesco), Van der Laak, W., Francois I., (WaterstofNet). (2016). Power-to-Gas Roadmap for Flanders. Brussels.
Van der Velden, N., & Smit, P., (2018). Energiemonitor van de Nederlandse glastuinbouw 2017. Wageningen. Netherlands.
U.S. Department of Energy Efficency & Renewable Energy (2011). Combined Heat and Power Technology Fact Sheet Series. EERE Information Center. Washington.
Wolf, E., (2014). Electrochemical Energy Storage for Renewable Sources and Grid Balancing: Large-Scale Hydrogen Energy Storage. SIEMENS.
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.
Yang, C., & Ogden, J. (2007). Determining the lowest-cost hydrogen delivery mode. International Journal of Hydrogen Energy, 32(2), 268-286.
54 APPENDICES
Appendix A – Illustration of Case Study 1
Appendix B – Characteristics of the type of wind turbine used in Case Study 1
Wind turbine characteristics Nordex N50/800
Site data - Air density & Wind distribution Netherlands (Schipol) Wind Turbine size (kW) 1000
Rotor diameter (standard) (m) 50 Hub height (standard) (m) 46
Capacity factor (%) 25
Power output per year (kWh) 1686776 Power output per month (kWh) 140565
55 Appendix C – Results of Case Study 1
Month H_prod (min
costs) H_stor (min costs) H_rg (min costs) H_prod (max eff) H_stor (max eff) H_rg (max eff)
January 477904 194287 283617 491959 208342 283617 February 477904 388574 283617 491959 416684 283617 March 477904 582861 283617 491959 625026 283617 April 477904 777148 283617 491959 833368 283617 May 477904 971435 283617 491959 1041710 283617 June 477904 1165722 283617 491959 1250052 283617 July 477904 1360009 283617 491959 1458394 283617 August 477904 1554296 283617 491959 1666736 283617 Sept 477904 1748583 283617 491959 1875078 283617 Oct 477904 1942870 283617 491959 2083420 283617 Nov 477904 2137157 283617 491959 2291762 283617 Dec 477904 2331444 283617 491959 2500104 283617 Total 5734848 15154386 3403404 5903508 16250676 3403404 Table C.1 Hydrogen produced (H_prod), stored (H_stor) and re-generated (H_rg) into electricity for scenario 1
in simulation I when minimizing costs and when maximizing efficiency Month H_prod (min
costs) H_stor (min costs) H_rg (min costs) H_prod (max eff) H_stor (max eff) H_rg (max eff) January 955808 672191 283617 983919 700302 283617 February 955808 1344382 283617 983919 1400604 283617 March 955808 2016573 283617 983919 2100906 283617 April 955808 2688764 283617 983919 2801208 283617 May 955808 3360955 283617 983919 3501510 283617 June 955808 4033146 283617 983919 4201812 283617 July 955808 4705337 283617 983919 4902114 283617 August 955808 5377528 283617 983919 5602416 283617 Sept 955808 6049719 283617 983919 6302718 283617 Oct 955808 6721910 283617 983919 7003020 283617 Nov 955808 7394101 283617 983919 7703322 283617 Dec 955808 8066292 283617 983919 8403624 283617 Total 11469696 52430898 3403404 11807028 54623556 3403404 Table C.2 Hydrogen produced (H_prod), stored (H_stor) and re-generated (H_rg) into electricity for scenario 2
