1
Creating a network design for local green
hydrogen initiatives
Faculty of Economics and Business
Master Supply Chain thesis
by
Marnix Hiemstra
S3258041
m.a.hiemstra.1@student.rug.nl
Supervisors:
dr. Zhu & dr. Ursavas
Word count:
15,624
2
Abstract
The current global fossil energy system is unsustainable and requires rigorous changing. Hydrogen is one of the main candidates to replace fossil fuels in a post fossil-fuel age due to its abundance and non-polluting aspects (if made from a renewable source). Abundant research indicates the potential use of hydrogen and multiple models and design have been created for large scale hydrogen implementation. Few models or designs, however, focus on small local green hydrogen initiatives. This research proposes a network design for these initiatives including production method, costs, and spatial requirements.
The proposed model is used to simulate several plausible scenarios for the city and province of Groningen in the North of the Netherlands. Additionally, a hydrogen initiative case, located in the city of Groningen, is qualitatively reviewed and compared to the proposed model. Though the model is able to provide solutions by using the case based parameters, the case differs from figures found in literature. These differences are due to several factors that are probably unique in this setting.
The green hydrogen initiative located in Groningen has broken through the chicken and egg problem by connected different parties and thus creating demand and supply simultaneously. They are also able to produce hydrogen at near competitive cost and will continue to grow in the coming years.
Based on the scenarios and the case it can be concluded that there are several aspects which are important when creating a green hydrogen initiative. The social aspect connected to the usage of renewable energy will remain important in the coming years, flexibility and scalability are important factors when creating hydrogen initiative, making use of opportunities (as found in the case) can provide (important) advantages for a local green hydrogen initiative, and solving the chicken and egg problem requires several involved parties.
3
Acknowledgments
This master thesis marks the end of my ‘awfully’ long time as a student. Where I started as a Sport, Health and Management bachelor student at the Hanze University of applied sciences, I will end as a Supply Chain Management master, graduated at the University of Groningen. The educational path I followed as a student could be labeled as challenging, where I combined several boards and committee functions during my last bachelor and all master years. Looking back on this period I feel great pride that I got so much further than I would ever have believed years ago.
I, however, would never have made it this far without the help of my parents, girlfriend, friends, and fellow committee and board members who supported, inspired, and endured me throughout my entire time as a student. You deserve gratitude that I will never be able to express in words. The overarching theme of this master thesis is the struggle mankind is in against its self-made environmental problem. A problem which I personally really have taken to heart. This was the reason that I decided to pursue the energy certificate and specialize in environmental issues. I hope my experiences during the creation of this master thesis will help me in my future career, in which I hope to daily contribute to decreasing climate change.
4
Table of content
1. Introduction ...6
2. Theoretical background ...8
2.1 Hydrogen production ...8
2.2 Negative aspects of hydrogen ...9
2.3 Hydrogen storage ...9 2.4 Modeling hydrogen ... 10 2.5 Knowledge gap ... 10 3. Model description ... 12 3.1 Nomenclature ... 12 3.2 Main assumptions ... 12 3.3 Parameters... 14
3.3.1 Demand, budget, and total available space ... 14
3.3.2 Production capacity ... 14 3.3.3 Production cost ... 15 3.3.4 Spatial requirements ... 16 3.3.5 Social factor ... 16 3.3.6 Storage ... 17 3.4 Constraints ... 19 3.5 Objective function ... 20 4. Numerical study ... 21
4.1 Study design (control scenarios) ... 21
4.1.1 Control scenario 1 ... 21 4.1.2 Control scenario 2 ... 22 4.1.3 Control scenario 3 ... 23 4.2 Solar energy ... 25 4.3 Wind energy ... 29 4.4 Biomass energy ... 33
5. Case study (Groningen) ... 36
5.1 Background of the case ... 36
5.2 Case characteristics ... 36
5.3 Comparison case and model ... 37
5.4 Future scenario ... 40
6. Discussion ... 42
7. Conclusion ... 43
5
7.2 Hydrogen production ... 43
7.3 Problems concerning hydrogen usage ... 44
7.4 Hydrogen storage ... 44
7.5 Summarized conclusions, theoretical contribution, and managerial implications ... 45
7.6 Future research ... 45
References ... 46
Table of figures ... 51
Appendix i Linear programming model ... 52
6
1. Introduction
3rd of July 2018, a local paper in the North of the Netherlands heads an article with “Groningen sacredly believes in hydrogen future” (DVHN, 2018). The article expresses the desire of the city and municipality of Groningen to invest in projects concerning hydrogen to strive toward a carbon neutral future. “Groningen is one of the frontrunners of Europe concerning the application of hydrogen” which is displayed by its functioning local (green) hydrogen initiative. Despite these local hydrogen initiatives, on a national level hydrogen has not made the great changes connected to hydrogens’ potential. No country in the world has been able to create a sustainable hydrogen economy on a large scale, even though there is a dire need for structurally changing the global energy structure (Hanley, Deane, & Gallachóir, 2018). According to Hanley et al. there are multiple reasons for changing the global energy system (the endeavor to achieve a carbon neutral future) (Hanley et al., 2018). These reasons include the fact that greenhouse gas (GHG) emissions need to be reduced to mitigate climate change (global warming), and also energy security (linked to the finiteness of fossil fuel), affordability of energy and growing energy demand needs to be considered (Dincer & Acar, 2018; Hanley et al., 2018; Hübert, Boon-Brett, Black, & Banach, 2011). GHG emissions, energy (in)security, increasing energy demand, and energy affordability combined have realized a global ‘energy crisis’. To achieve the desired change mentioned by Hanley et al., new green fuels or energy carriers will have to be implemented globally, perhaps even by starting locally mirroring the city of Groningen.
Already in 1973, hydrogen was mentioned as a promising new ‘energy’ for storage and transmission (Russell, Nuttall, & Fickett, 1973). Continuing the years the usage of hydrogen (in small initiatives) grew, and many studies continued to recognize the potential of hydrogen (Hanley et al., 2018; Hirose, 2011; Melaina, 2003; Sharma & Ghoshal, 2015). The damaging effects of the usage of fossil fuel (Armaroli & Balzani, 2011b, 2011a; Balat & Balat, 2009) received more attention throughout the 00’s eventually leading to global accords like the recent Paris Agreements. In the Paris Agreements the long term goal of keeping the increase in global temperature to well below 2° C above pre-industrial levels was set (Paris Agreement, 2015). The possibilities and potential of hydrogen have been discussed and studied abundantly. These studies however focused mainly on the ‘big picture’. Many studies focus on continental (mostly EU) or national plans (Hanley et al., 2018) for large scale hydrogen usage. Kuby et al. (2009) investigate how hydrogen stations can be optimized ‘locally’ in Florida. However the scope of the study is still quite broad (Florida is more than four times the size of the Netherlands) (Kuby et al., 2009). Literature concerning the design of hydrogen networks also focuses on the broad application or industrial use of hydrogen (Zhou et al., 2013). Parker, Fan, and Ogden created a network design (including a linear model) for the production of hydrogen from biomass and several distribution methods (Parker, Fan, & Ogden, 2010). This, however, was based on a northern California case which is still many times larger than the Netherlands, and thus fairly incomparable to a city as Groningen.
7
cities/municipalities, it can be difficult to make apt decisions how to decrease overall pollution considering sustainability, renewability, innovations (such as hydrogen production/usage), etc. This research attempts to provide a practical tool for easing such decisions. The research question connected to this research is as follows:
What is a suitable network design of a local green hydrogen initiative?
While research is scarcely performed on small hydrogen implementation, these could prove essential in achieving a non-polluting energy system. Dincer & Acar signify the importance of small scale hydrogen systems by stating that “Large scale central H2 generation has higher capital investments costs, so small scale distributed systems could have a key part during the transition to widespread use of innovative H2 energy systems” (Dincer & Acar, 2017a) P.14854. By starting small and locally, capital costs remain low and thus less risk exists for investors and more possibilities for subsidies from governmental bodies. The produced hydrogen will be consumed locally by pre-determined goal such as public transportation, where the hydrogen-fueled transport market is expected to develop most prominently (Dincer & Acar, 2018). Knowledge concerning small green hydrogen initiatives is limited. though it could play a big role in changing the global energy system. This study aims to add insight in this area to stimulate the necessary energy transition from fossil fuels to green energy.
8
2. Theoretical background
The usage of hydrogen in the city of Groningen follows the global desire and necessity for changing the current (fossil) energy system. Of all new fuels and energy carriers, hydrogen is one of the most promising, when considering alternative (non-polluting) energy carriers (Hosseini & Wahid, 2016). Hydrogen is an energy carrier due to the fact that it has to be produced (Armaroli & Balzani, 2011a; Bhandari, Trudewind, & Zapp, 2014; de Bruijn, 2008). An advantage of hydrogen, compared to other renewables, is that it can relatively easily be stored and ‘supplied’ when there is demand (Zhang, Zhao, Niu, & Maddy, 2016). It is unrealistic to believe that hydrogen will replace fossil fuel energy supply overnight, therefore the only implementation method is to start small, by locally supplying hydrogen to satisfy energy demand. Perhaps slowly, but steadily, replacing energy supplied from fossil fuels to energy supplied from (locally produced) hydrogen. Dincer and Acar (2017) state that “the detailed structure for a well-developed H2 economy in the long term is not clear yet. However,
it is expected that different combinations or existing sources, systems, and structures will play a major role” (Dincer & Acar, 2017a) P.14855. These different combinations can, for example, be large or small implementations of hydrogen.
2.1 Hydrogen production
Hydrogen can be produced in many different ways depending on the techniques that are used (Bhandari et al., 2014; Dincer & Acar, 2017a). Even though fossil fuels are the major source of hydrogen production today (Dufour, Serrano, Gálvez, Moreno, & González, 2011), to solve the energy crisis carbon-neutral energy is needed, which for hydrogen means that it has to be produced with renewables (Dincer & Acar, 2017a). Geothermal, hydropower, solar, and wind are considered renewables because the consumption rate is lower than the rate at which their reserves are restored (Dincer & Acar, 2015). Bhandari et al. (2014) state that from an ecological perspective wind and hydro-based electrolytic hydrogen production are considered most suitable, which really specifies the most renewable method (Bhandari et al., 2014). Due to their unsustainable nature, this study will not review hydrogen production by other non-renewable means (coal, natural gas, oil) (Dincer & Acar, 2017a). It should, however, be stated that gas will likely be used in the transition from fossil fuel to hydrogen from fossil fuel to green hydrogen (Ismail & Bahnemann, 2014). A transition in the energy sector simply cannot be done overnight and thus transition fuels will be required. Also, a focus on small scale/locally implementable production methods will be required to stay within the boundaries of this research.
Currently, only one commercially available hydrogen production method from renewable resources exists, namely electrolysis (Ozbilen, Dincer, & Rosen, 2012). In electrolysis, electricity is used to split H2O (water) in H2 (hydrogen) and O2 (oxygen) (Wang, Wang, Gong,
9
Photonic hydrogen production, which used (sun) light energy to H2O in H2 and O2, could also
be convenient for local hydrogen initiatives. This method, however, is “currently in very early stages of research, these processes offer long-term potential for sustainable production with low environmental impact” (Dincer & Acar, 2017a) P.14847. Hydrogen produced by sunlight has great potential however also requires more research (Bicer, Dincer, & Zamfirescu, 2016).
2.2 Negative aspects of hydrogen
Though there is a lot of optimism about the usage of hydrogen there is also grounded skepticism. Armaroli and Balzani (2011a) mention several problems currently connected to the (considerable) use of hydrogen such as a large amount of electricity needed to produce hydrogen (Armaroli & Balzani, 2011a). Additionally, Sharma and Ghoshal (2015) state that usage of hydrogen as an energy carrier to avoid the emission of harmful gases is highly debatable (Sharma & Ghoshal, 2015). This is conjointly due to the fact that hydrogen has to be produced, which requires (renewable) electricity which makes hydrogen production expensive (Dincer & Acar, 2017b). Concluding on the skepticism, Mackay (2009) takes it even further by calling hydrogen a “hyped up bandwagon” (David & Mackay, 2009) P.129, explained by the problems concerning (electric) hydrogen production, hydrogens’ backlog on electricity, and hydrogen’s bulk (volume). Concerning small hydrogen initiatives, Edwards, Kuznetsov, David, and Brandon state that small initiatives will not be able to match the volumes of global hydrogen requirement (Edwards, Kuznetsov, David, & Brandon, 2015). Simply due to the huge demand for energy in transportation and households, several large hydrogen production initiatives will be needed too. The low (current) usage of hydrogen in transportation and households is (partly) due to the chicken and egg problem as explained by Melaina (2003) and Kuby et al. (2009) essentially stating that hydrogen is not supplied due to the low demand and not demanded due to the low supply (Kuby et al., 2009; Melaina, 2003). Especially when reviewing countries or states the chicken and egg problem creates a big problem for the implementation of hydrogen. Large investments are required in both the supply and demand sides to implement hydrogen on a large scale.
2.3 Hydrogen storage
One of the advantages of hydrogen is that it can be stored after production until it is needed for consumption (Zhang et al., 2016). This is especially useful in combination with other renewable energy sources which fluctuate in their power output due to their dependability on the weather (intermittency) (Dincer & Acar, 2017a). Subsequently, this fluctuation in output makes it difficult to align power demand and supply. For example solar energy is logically produced more during sunny days, however on these days, the temperature is higher and thus less energy is used for example for heating. On less sunny (colder) days, less energy is produced while more energy is needed for heating. It is, however, not possible to use the excess solar energy produced during sunny days seeing that this energy could not be stored. If the excess energy was used to produce hydrogen, which can be stored, it could then be used during colder days. Dincer & Acar (2018) underline this by stating that “Hydrogen has a high-energy density, can be stored for long periods and is easy to transport. It is therefore well-suited to serve as an energy buffer and for strategic reserves of energy” (Dincer & Acar, 2018) P.8585.
10
(Dalebrook, Gan, Grasemann, Moret, & Laurenczy, 2013). Though effective hydrogen storage is possible Bhandari et al. (2014) finds that more innovation is still required before hydrogen can be successfully commercialized (Bhandari et al., 2014). Additionally, Dincer and Acar (2017) find that hydrogen storage is, due to the low volumetric energy density, expensive and in storage, there is the potential for leaking (Dincer & Acar, 2017a). The costs of hydrogen are high due facets like production and storage (Pudukudy, Yaakob, Mohammad, Narayanan, & Sopian, 2014). According to Pudukudy et al. (2014) producing hydrogen affordably, effectively and non-damaging to the environment is not possible (Pudukudy et al., 2014).
2.4 Modeling hydrogen
In the article by Samsatli and Samsatli (2018) a MILP approach is used for the design and operation of integrated urban energy systems, which comes close to the goal of this research. Their research is one of many using optimization models to model urban energy systems. Samsatli and Samsatli (2018) illustrate their model with an eco-town in which electricity and heat demand are met with a variety of energy resources. Their model purposely does not focus on one specific energy or feedstock because they aim to create a generic model applicable to different situations. Samsatli and Samsatli (2018)) find that their case (an eco-town) can be supplied heat and electricity by using one centralized biomass CHP plant with a backup boiler and conclude that their presented MILP model is useful for modeling energy systems in urban and rural areas. According to Samsatli and Samsatli (2018)) many models in this field do not account for storage and many neglect spatial aspects which is an interesting research gap (Samsatli & Samsatli, 2018). Optimizations models can have different scales such as systems, districts, cities, and countries. On a country scale, Kim and Moon (2008) use a MILP approach to build the future hydrogen supply chain of Korea. Their model includes cost efficiency and safety which are important factors for large scale usage of hydrogen. They conclude that their proposed model is suitable to support strategic decision making concerning the future of hydrogen usage throughout Korea (Kim & Moon, 2008). Though this indicates the potential of modeling energy-related issues the scale of the model by Kim and Moon (2008) is very large and also situated in a non-European country which makes the applicability for a local setting questionable. Other well-known models include MARKAL and TIMES which can describe the evolution of specific energy systems over long periods of time. These models are frequently applied to solve energy-related matters as in the article by Sgobbi et al. (2015) in which the role of hydrogen in medium and long-term decarbonization of Europe is investigated (Sgobbi et al., 2016). Comparable to the objective of a TIMES model (concisely using linear programming to satisfy energy demand while minimizing e.g. costs) (Sgobbi et al., 2016) the goal of the model proposed in this research will also focus on supplying energy while minimizing costs.
2.5 Knowledge gap
11
12
3. Model description
The goal of this study is to qualitatively review the usage of green hydrogen in a local setting. To support this qualitative review a simplified model is proposed considering hydrogen production from several renewable resources, cost of hydrogen production, spatial requirements, a social factor to account for (local) public sentiment connected to renewables, and the storage of hydrogen. The model will follow a specific set of demands and will be formulated as an optimization model regarding how to produce green hydrogen. The four renewables resources that are considered are wind (1), solar (2,3), geothermal (4) and biomass (5). These renewable resources were chosen based on the fact that the can be applied in small to large settings and are thus fitting for a local hydrogen initiative.
3.1 Nomenclature
Xi = amount of green hydrogen produced via method i Ii = number of installations per different production method i D = demand for hydrogen
B = budget
TS = total available space
PCi = production capacity of method i (per installation) Ci = cost of producing hydrogen via method i
SRi = spatial requirements of method i SFi = social factor
DSF = desired social factor
SCA = storage capacity (of hydrogen) SCO = storage costs (of hydrogen)
SSR = storage spatial requirement (of hydrogen) 1 = Wind
2 = Solar (per solar installation) 3 = Solar (per solar panel) 4 = Geothermal
5 = Biomass
3.2 Main assumptions
13
The second assumption is that de budget for a local hydrogen initiative is limited. The model uses a fairly simple approach towards the costs of producing hydrogen and the budget available. For this study it is assumed that the first hydrogen initiatives will not be commercialized but, as mentioned at the demand assumption, will have a specific (municipal) goal. Therefore it is assumable that hydrogen will not be ‘sold’ but the focus will lie on the cost of producing hydrogen and the available budget.
The third assumption is that the space for a local hydrogen initiative is available, shapeable, and limited. To include the spatial requirements it is assumed that the required space is available (no costs, distance to other facilities or other factors are considered), that the space can have any shape whichever fits the selected production method, (for example for wind turbines a circular area is best suited due to the distance that is required from wind turbines to other buildings), and that the provided space is limited precisely to that what is provided (due to other buildings, permits, etc.).
The fourth assumptions concern the storage of hydrogen. This assumption is connected to the consumption of the produced hydrogen. If the produced hydrogen can be consumed immediately there is no need for storage. However, if the produced hydrogen cannot be consumed immediately (which is likely), a storage facility is required at the local hydrogen initiative. In this case, it will also be possible to produce more than demand however limited to the storage capacity. It is further assumed that the cost of storage, the storage capacity, and the spatial requirements are the same for each production method. Hydrogen produced via solar energy or biomass does not have different properties and thus these factors can be assumed to be similar.
The fifth assumption/limit is that the model focusses on the creation (start) of a local hydrogen initiative. When the initiative is operational, some changes to the model have to occur to account for the stored hydrogen and other operational aspects. Stored hydrogen could, for example, be sold or consumed while production is lowered to below demand. The model does not allow this at first to guarantee that demand will be met. No municipality would want to have bought for example 10 hydrogen vehicles to find out that only 8 can be fueled.
14
3.3 Parameters
In the linear model, several parameters are used. The figures used as parameters are based on articles and/or are deducted from statistical sources. For some figures, some logical deduction had to take place in order to make the figures usable. The goal of this model is not to provide precise calculations but to indicate general directions for hydrogen production. To do this some freedom in figures based on logical reasoning and (simple) calculations is required. Table 1 is used throughout this study to (re)calculate figures into the used metrics.
Conversion factors for hydrogen units
To: From: KG GJ kWh
KG 1 0.12 33.33
GJ 8.333 1 277.8
kWh 0.03 0.0036 1
Table 1 Conversion factors for hydrogen units (adopted from Lemus et al. 2010 and adjusted for this study) 3.3.1 Demand, budget, and total available space
The three parameters that are adjustable based on the investigated situation are demand, budget, and the total available space. For example, a local hydrogen initiative could have a demand of 100 MWH, a budget of 1000 Euros and 5000m2 of available space. The model will determine
the best renewable resource ‘set-up’ to satisfy these demands. Additionally, the desired social factor can be changed depending on the situation. The desired social factor is, however, a more qualitative factor and its importance can differ per situation. It should, therefore, be seen as a guiding factor and not a quantitatively decisive one.
3.3.2 Production capacity
PC1 was calculated based on information provided by CBS Statline (Dutch statistics bureau). In 2017, 6869 million kilowatt-hours (kWh) of electricity was produced by wind turbines on land. At the beginning of 2017, 1857 wind turbines were operational on land and at the end of 2017, 1981 were operational on land, averaging 1919 wind turbines in 2017. Based on these figures the average electricity production per wind turbine was calculated and converted to hydrogen in MWh.
PC2 was calculated based on information provided by CBS Statline. In 2017, 2149 million kWh of electricity was produced by solar energy installation of which 5.13% (110.22 million kWh) had a commercial goal (the other +/- 95% was produced on housing, farms, etc.). This 5.13% was generated by 318 commercial installations (with the goal of generating electricity for ‘the net’). Based on these figures the average electricity production per (commercial) solar panel installation was calculated and converted to hydrogen in MWh.
15
included. This selection provided that in 2017 on average 9836 (beginning 2017:9860, end of 2017: 9812) geothermal heat pumps were operational generating 3838 TJ of heat. The TJ of heat were re-calculated into kWh and, by using the law of thermodynamics, the amount of generated electricity was calculated.
PC5 could not be based on information provided by CBS Statline due to the fact that the number of biomass electricity generating plants (or other usages of biomass which could be feasible for a local hydrogen initiative) was not available. There is statistical information concerning the total production of electricity via biomass however by how many biomass installations (centralized and/or decentralized) this electricity was generated was not present. There are however qualitative examples of biomass-based electricity production in the Netherlands. For example, the biomass electricity generating plant at Balkbrug is a 10.2 MWh plant which is a feasible option for a local hydrogen initiative.
For each method initially, the electricity production was calculated. This study, however, focusses on hydrogen and thus the produced electricity had to be converted to hydrogen by using several factors. Badwal, Giddey, and Munnings (2018) describe several commercial solid-state electrolyzers that are in operation in several countries. The electrolyzer used in Frankfurt Germany for power-to-gas and refueling was estimated to be most comparable for this study and uses 53.5 kWh to produce one KG of hydrogen (Badwal, Giddey, & Munnings, 2018). Lemus and Martínez Duart (2010) provide a conversion table for hydrogen units in which the conversion factor from KG to kWh is 33.33 (Lemus & Martínez Duart, 2010).
PCi Hydrogen production capacity MWH Period Per
PC1 2,228.0 Year Wind turbine
PC2 215.7 Year Energy provision solar installation
PC3 0.2 Year Solar panel
PC4 23.6 Year Geothermal heat pump
PC5 63.5 Year Biomass electricity provision plant
Table 2 Production capacity per method i 3.3.3 Production cost
C1, C2, and C5 are based on the work of Nikolaidis and Poullikkas (2017) who compare different hydrogen production methods including the cost of producing hydrogen in USD/kg per production method (Nikolaidis & Poullikkas, 2017). The selected data for this study were converted from USD to EUR by using the average exchange rate of 2017. Data dating from before 2017 was corrected for inflation. C3 was based on C2 however a factor of 1.15 was used to compensate for the fact that purchasing single solar panels does not allow economy of scale advantages.
C4 was based on the work of Lemus and Martínez Duart (2010) who provide a comparison of hydrogen production methods including production costs in USD/GJ (Lemus & Martínez Duart, 2010). Similar to C1, C2, and C4 all data was converted to EUR and corrected for inflation.
Ci Costs Per Costs Per
C1 €0.20 kWh €200 MWH
C2 €0.33 kWh €330 MWH
C3 €0.38 kWh €380 MWH
C4 €0.21 kWh €210.5 MWH
C5 €0.08 kWh €80 MWH
16
3.3.4 Spatial requirements
SR1 was calculated based on the minimal distance to a wind turbine is required for housing. Regarding the fact that there are varying opinions concerning this subject a ‘general’ distance of 1000m was assumed. This results in a (circular) spatial need of 3,141,593 m2 (10002*Π) or 3.14 km2. Naturally, SR1 can vary depending on the local circumstances. When considering multiple wind turbines the framework by Schallenberg-Rodriguez (2013) should be used to determine wind turbine spacing, array (wind turbines) efficiency and eventually spatial requirements for wind turbines (Schallenberg-Rodriguez, 2013).
SR2 was calculated by dividing P2 by the average solar panel yield per year (225 kWh). This provided the number of solar panels per installation (here the term installation is similar to the explanation of Pi) and by adopting the average size of a solar panel (1m by 1.6m thus 1.6 m2)
SR2 could be calculated. Larger solar panel installations, however, do not entirely consist of solar panels but also of walking paths between them. To account for this a factor of 1.05 was used. For SR 3 the size of one solar panel (1.6 m2) was adopted.
SR 4 was calculated based on the work by Li et al. (2015) who state that geothermal installations require 18-74 km2 to generate 1 TWh (Li, Bian, Liu, Zhang, & Yang, 2015). For this study, it is assumed that for a local hydrogen initiative the smallest possible geothermal installation is most suited and thus the 18 km2 required for 1 TWH is used for calculations. By taking the
smallest geothermal installation the average energy generation per m2 increases, which could lead to higher (investment) costs. In the article by Li et al. (2014) no clarification is provided for the large spatial range of the geothermal installations and thus multiple reasons, besides costs, could be considered (Li et al., 2015). Therefore the potential increase in cost due to a smaller (spatially more efficient) geothermal facility is ignored. Additionally for geothermal installations logically a substantial part of the installation is located underground. To account for this it is assumed that at most 20 percent of the installation is above ground. Additionally, we assume that the ground used for the other 80 percent becomes unavailable for a building. For the model, 20 percent of the spatial requirement (based on 18 km2 for 1tWh) for an
installation of 23.6 MWh is used, with the side note that the land used for the underground portion of the heat pump is thus unavailable for a building.
SR5 was based on the biomass electricity generating plant in Balkbrug. The biomass plant and connected warehouses for biomass take approximately 16000 m2.
SRi Spatial requirement m2 Per
SR1 3,141,593 Wind turbine
SR2 2,588 Energy provision solar installation
SR3 1.6 Solar panel
SR4 136,581 Geothermal heat pump
SR5 16,000 Biomass electricity provision plant
Table 4 Spatial requirement per method i 3.3.5 Social factor
17
could be found providing legitimate negative social aspects of the generation of solar energy. Though visual effects could be seen as a negative aspect (though possibly depending on the size of the solar panel field), compared to the visual effect of wind turbines this is considered negligible. Therefore SF2 and SF3 are set on 1.
Li et al. (2014) mention several qualitative social impacts of multiple renewable energy sources. Though they mention potential seismic activity due to the usage of geothermal energy, this would only occur in areas where seismic activity is already a threat and could be influenced due to the drilling for geothermal energy (Li et al., 2015). On geothermal heat pumps no negative social aspects could be found and thus SF4 is set on 1.
According to Herbert and Krishnan (2016), there are several negative aspects connected to electricity production from biomass (Joselin Herbert & Unni Krishnan, 2016). Most of these negative aspects, however, concentrate on feedstock production and labor requirements which, concerning a local hydrogen initiative, do not necessarily suffice as social factors (Joselin Herbert & Unni Krishnan, 2016). This would depend on whether the feedstock is produced locally and if it should be included in the model. Evans et al. (2010) mention food competition (energy crops competing with food crops for valuable agriculture land) as the key social factor (Evans, Strezov, & Evans, 2010). Though this would also depend on how local these crops are produced, SF5 is set on 0.9.
SFi Social factor SF1 0.8
SF2 1
SF3 1
SF4 1
SF5 0.9
Table 5 Social Factor per method i 3.3.6 Storage
To account for storage in the model three parameters were created. Storage capacity (SCA), storage cost (SCO) and storage spatial requirements (SSR). Storing hydrogen as compressed gas or in a liquefied state are the most mature technologies (Barthelemy, Weber, & Barbier, 2017; Zhang et al., 2016) and thus these were considered as potential storage methods. Storing hydrogen as compressed gas is globally the most widely adopted method. This combined with the fact that liquefied hydrogen is often more expensive, and also mainly used to transport large volumes of hydrogen, favors compressed gaseous hydrogen for a local hydrogen initiative (Zhang et al., 2016).
18
According to Zhang et al. (2016), the projected cost per kWh is 17 USD and 13 USD for compressed hydrogen 700 bar and 350 bar respectively. For the local hydrogen initiative, it is assumed that highly compressed hydrogen is not needed and thus 13 USD per kWh was selected as a parameter. This is due to the fact that highly compressed hydrogen is mainly used in transportation (seeing that the fuel tanks for cars have to be small and light) and not in general storage (Zhang et al., 2016). Similar to the cost of hydrogen production, the cost of hydrogen was corrected for inflation and exchanged to Euros.
Hydrogen storage tanks come in different sizes and with different pressures. High-pressure is usually used when space for the tank is limited (for example in cars or trucks). High-pressure tanks can contain more hydrogen however also faces higher cost due to the usage of compressors, high-pressure valves, etc. Additionally, the volumetric density of hydrogen cost does not increase proportionally to the applied pressure. Increasing pressure to store more hydrogen does work however is fairly ineffective. For stationary hydrogen storage, low costs is usually a primary goal and thus low pressure and ‘bulky’ solutions are preferred (Zhang et al., 2016). An 850 L storage tank can store 4.2 KG of hydrogen at a pressure of 60 bar. This tank measures 0.84M by 1.87M which results in 1.57 m2. These measurements are however without the support for the tank. To account for this a factor of fifteen percent is used which results in a rounded 1.81 m2. One KG of hydrogen can provide 33.3 kWh and thus this tank can
contain 139.86 in kWh. For this parameter, it is assumed that the size of the storage tank cannot simply be expanded or contracted without consequences and thus the storage spatial requirement is considered per 850 L tank. Additionally, seeing that the total required storage capacity is connected to the demand, the spatial requirement will also be calculated based on 10% of demand. This means that even for example with a demand of 100 kWh, a storage capacity of 10 kWh will be created even though the actual production is 105. The storage tank will then simply not be completely filled.
3.3.7 Overview of parameters Xi Production capacity (PCi) (MWh) Costs (Ci) Spatial requirement (Sri) (m2) Social factor (SFi) Storage capacity (SCA) Storage costs (SCO) (kWh) Storage spatial requirements (SSR) (m2) X1 2,228.0 €200 3,141,593 0.8 10%D €11.8 1.81 X2 215.7 €330 2,465 1 10%D €11.8 1.81 X3 0.16 €380 1.6 1 10%D €11.8 1.81 X4 23.6 €210.5 136,581 1 10%D €11.8 1.81 X5 63.5 €80 16,000 0.9 10%D €11.8 1.81
Table 6 Overview of parameters per method i 3.3.8 Decision variable
19
3.4 Constraints
In this section, the constraints of the model are provided. ∑5 𝐼𝑖 𝑃𝐶𝑖
𝑖=1 ≥ D (1)
Constraint (1) ensures that the production of hydrogen is similar to or higher than demand. This constraint is needed to ensure that production always meets demand. Here it is assumed that there is no stored hydrogen available to solve any shortages between production and demand. ∑5𝑖=1𝐼𝑖 𝑃𝐶𝑖 – D ≤ SCA (2)
Constraint (2) ensures that the production of hydrogen does not exceed the capacity in which hydrogen can be used or stored. Production can exceed demand however it cannot exceed demand plus storage capacity or, as mentioned above, production minus demand cannot exceed storage capacity.
∑5𝑖=1𝐼𝑖 𝑃𝐶𝑖 𝐶𝑖 + (∑5𝑖=1𝐼𝑖𝑃𝐶𝑖) − 𝐷 𝑆𝐶𝑂 ≤ B (3)
Constraint (3) ensures that the sum of the hydrogen production costs plus the sum of the hydrogen storage costs does not exceed budget. The production costs per method are represented by Ci which is multiplied by Ii and PCi to account for total production. The storage costs do not vary per production method (for example hydrogen produced by wind energy or biomass energy have the same storage costs) and thus the storage cost for all hydrogen is represented by SCO. This is multiplied by the hydrogen that is stored (production minus demand).
∑5𝑖=1𝑆𝑅𝑖 𝐼𝑖+ 𝑆𝑆𝑅 ≤ TS (4)
Constraint (5) ensures that the spatial requirements of the used method per installation plus the storage spatial requirements do not exceed the total amount of space available for the local hydrogen initiative. The hydrogen initiative will thus exist of two parts (production and storage) which both need to ‘fit’ in the available space.
SFi ≥ DSF Si (5)
The social factor of a used production method should be higher or similar to the set desired social factor. In this way, no ‘socially unwanted’ production methods are used in the production of hydrogen.
Ii ≤ Si M (6)
The number of installations should be smaller than SiM. If Si is 1 than any amount of installations can be used. If Si is 0 than method Ii cannot be used due to the fact that Ii will then have to be below zero which is impossible due to constraint (7).
Ii ≥ 0, Ii ∈ ℕ (7)
20
3.5 Objective function
Max ∑5𝑖=1𝐼𝑖𝑃𝐶𝑖 (8)
The objective of the model is to maximize the amount of hydrogen that can be produced while considering budget and space limitations and serving a certain demand.
21
4. Numerical study
In this section, several scenarios will be discussed and reviewed with the help of the proposed linear model. Considering a local area such as Groningen many different scenarios can be derived from recent trends. These trends originate in different areas, such as local political changes, resulting in new policies and regulations or commercial changes, resulting in entrepreneurs pursuing monetary gain. The following scenarios are mainly based on such trends as observed in the city and province of Groningen.
4.1 Study design (control scenarios)
To test the model and provide an overview of the possible outcomes, firstly several ‘control’ scenarios were performed. These control scenarios will be used throughout the numerical study as a reference to the ‘real-life’ based scenarios. To apply the model, the used demand, budget, total available space and desired social factor have to be determined first. It is good to point out that demand can originate from different usage, which can influence the used figures for demand, budget, space, and social factor for this control and other (applied) scenarios. Hydrogen can, for example, be used in transportation, in households and industries. The different usages of hydrogen create 3 scenarios, transportation (scenario 1), households (scenario 2), and industry (scenario 3). Though the transportation sector is a huge consumer of energy, the average usage per vehicle will be rather low (especially compared to households) and thus scenario 1 can be considered the ‘small’ scenario. Households use more energy than individual vehicles, and perhaps more than industry together, however, no households are completely powered by hydrogen (yet) and thus scenario 2 is considered the ‘medium’ scenario. Industry is the most developed user of hydrogen and thus demand (or relative usage) will be highest when compared to transportation and households and thus scenario 3 is considered the ‘large’ scenario (Dincer & Acar, 2017a; Parthasarathy & Narayanan, 2014; Yolcular, 2009). The used input parameters (demand, budget, space and social factor) are considered per scenario and are calculated in this context. In this way, the scenarios represent different usages of hydrogen, test the model in different aspects, and provide proper control scenarios to compare to the ‘real-life’ based scenarios
4.1.1 Control scenario 1
To determine the demand (in hydrogen) for scenario 1, the number of used vehicles, their efficiency, and covered distance had to be determined. According to Eberle et al. (2012), to cover a distance of 500 km, 5 kg of hydrogen is required, which would be stored in a 70 MPa pressure tank (Eberle, Müller, & von Helmolt, 2012). For a small hydrogen initiative, a max of 15 vehicles is assumed which cover a distance of 25 km a day for 200 days a year. Based on this assumption 50 kg of hydrogen is required per vehicle and thus a total of 750 kg for the 15 cars. By using the factors provided by Lemus et al. (2010) this results in a 24997.5 kWh or 25 MWh of demand for scenario 1. Budget is set on €7,500 (€500 per vehicle), total space 0.2 km2
22 Scenario 1 as control Input Demand 24,997.5 kWh Budget € 7,500 Available space 200,000 m2 Output (production) I3PC3 1,457 kWh I4PC4 23,614 kWh Total production 25,071 kWh Storage 71 kWh Output (costs) C3 € 553 C4 € 4,970 Storage € 871 Total costs € 6,391
Output (spatial requirement)
SR3 14 m2
SR4 136,581 m2
Storage 33 m2
Total required space 136,628 m2 Table 7 Model results control scenario 1
Table 7 provides a summaryof the results of scenario 1. Based on the used demand, budget and available space the model determined that geothermal combined with loose solar panels is the best option to maximize hydrogen production. For ‘small’ usage of hydrogen, these two production methods are best suited. Due to the fact that the DSF was put on 1, I1 (wind) and I5 (biomass) were unavailable as production methods seeing that their SF was lower than 1. In scenario 1 the type of vehicle is completely ignored. Vehicles used by for example a municipality can naturally variate in their size and purpose, and thus also in the amount of hydrogen they require to properly function. The scenario is designed to indicate the capabilities of the model and thus such small inaccuracies can be ignored in this stage. Additionally, as mentioned before, the model is designed to provide estimations, not exact figures.
4.1.2 Control scenario 2
This scenario simulates the usage of hydrogen in households. According to CBS Statline, in 2017 the average electricity usage per household in the province of Groningen is 2640 kWh per year which is a little lower than the 2860 average of the Netherlands. The average natural gas usage per household is 1,480 m3, which is higher than the average of the Netherlands (1,240 m3), or 14,458 kWh. The total power usage per household is thus 17,098 kWh. In this scenario, it is assumed that, if a household is powered by hydrogen, electricity, and gas usage are replaced by hydrogen. Considering how far the overall usage of hydrogen (especially in households) is, a local hydrogen initiative powering 50 households is already quite enthusiastic/utopian. However, seeing that Groningen has expressed its belief in the applications of hydrogen 10 households is considered viable (in the near future). The demand is hence put on a total of 170,981 kWh or 171 MWh. For the budget €2,500 per household (€208 per month) is assumed totaling €25,000 and the available space is set on 0.4 km2. The desired social factor is set on
0.9 seeing that, to power several households in a non-pollution manner, some negative social aspects might be acceptable. This is in line with the fact that many (partly undesired) solar/wind
23
power fields are being built to generate electricity for households. With these parameters, the model provides a solution which is summarized in table 8. A complete overview of the output can be found in appendix ii.
Scenario 2 as control Input Demand 170,981 kWh Budget € 25,000 Available space 400,000 m2 Output (production) I3PC3 20,559 kWh I4PC4 23,614 kWh I5PC5 126,976 kWh Total production 171,149 kWh Storage 167.6 kWh Output (costs) C3 € 7,802 C4 € 4,970 C5 € 10,158 Storage € 1,972 Total costs € 24,902
Output (spatial requirement)
SR3 203 m2
SR4 136,581 m2
SR5 32,000 m2
Storage 221 m2
Total required space 169,005 m2
Table 8 Model results control scenario 2
Table 8 provides a summary of the results of scenario 1. Based on the used demand, budget and available space the model determined that biomass combined with loose solar panels is the best option to maximize hydrogen production. Also, due to the fact that DSF was set on 0.9, wind energy I1 (wind) was unavailable as a production method. From the results of scenario 2, it can already be ascertained that the model is most suited for small(er) hydrogen initiatives. The model proposes the usage of a geothermal heat pump, several loose solar panels, and multiple biomass installations. Completely replacing gas and electricity in households is not necessarily impossible, however, will, in reality, most likely not be solved by the proposed method of the model. Therefore here the model deviates from reality however does provide an overview of potential options to supply households with hydrogen. Additionally, the budget per household is quite high however it should be kept in mind that this would be completely green energy which has been transformed into hydrogen which are both costly aspects.
4.1.3 Control scenario 3
This scenario simulates the use of hydrogen in industry. The demand, budget, and available space for this scenario depend on the size of the modeled industry. For this scenario, it is assumed that these factors would be (significantly) higher than scenario one and two. This already implies that this scenario is not entirely in line with the goal of this study, which is to create a model for (small) local hydrogen initiatives. The only purpose for such a scenario would be the demonstration of the functioning of the model with larger numbers. When a large(r) hydrogen initiative is simulated by the model, the results are similar to scenario two
24
25
4.2 Solar energy
In the city and province of Groningen, a large increase in photovoltaic power stations (solar parks) is planned for the coming years. In recent years two big projects in the city (50,000 solar panels) and 7 projects in surrounding villages (approx. 320,000 solar panels) have been completed. This includes one of the biggest solar parks in the Netherlands. Furthermore, 25 new solar parks ranging from around 1,500 solar panels to 85,000 are planned and have either already received permits, have petitioned for permits or are in preparation/negotiation phase. It is safe to say that solar energy is on the rise in Groningen, solar energy that could be used for hydrogen production. Currently, the established solar parks are destined to provide energy for nearby households. However, due to the intermittency in energy demand, during low energy demand excess energy is run back in the electricity net. This excess energy could also be used to create hydrogen to provide the nearby population with a non-polluting fuel.
Figure 1 Solar field near the city of Groningen housing 7777 solar panels (source Grunneger power & Hier opgewekt)
The large increase in solar energy in Groningen can have multiple consequences for a local hydrogen initiative. The increase in solar energy in the city of Groningen and its surrounding province could have a commercially positive effect. The high demand for solar panels could increase competition and could cause prices to fall. This could result in a lower cost to create hydrogen from solar energy which can be interesting for local hydrogen initiatives. Even before all the solar parks (mentioned above) are actually operational, there is already a lot of resistance due to the visual side effect of solar parks. Due to this resistance, the emphasis lies on community support before permitting new solar parks. This could result in a limitation on the size of new solar panel fields which can also influence new hydrogen initiatives.
Simulated solar scenarios
26
Scenario 1 with 10% decrease in solar energy costs Scenario 1 with 10% decrease in solar energy costs Scenario 1 with 10% decrease in solar energy costs
Input Input Input
Demand 24,997.5 kWh Demand 24,997.5 kWh Demand 24,997.5 kWh
Budget € 7500 Budget € 7500 Budget € 7500
Available space 200,000 m2 Available space 200,000 m2 Available space 200,000 m2
Output (production) Output (production) Output (production)
I3PC3 1,457 kWh I3PC3 1,457 kWh I3PC3 25,091 kWh
I4PC4 23,614 kWh I4PC4 23,614 kWh I4PC4
Total production 25,071 kWh Total production 25,071 kWh Total production 25,091 kWh
Storage 74 kWh Storage 74 kWh Storage 91 kWh
Output (costs) Output (costs) Output (costs)
C3 € 498 C3 € 415 C3 € 4761
C4 € 4,970 C4 € 4,970 C4 €
Storage € 871 Storage € 871 Storage € 1101
Total costs € 6,336 Total costs € 6,253 Total costs € 5,862
Output (spatial requirement) Output (spatial requirement) Output (spatial requirement)
SR3 14 m2 SR3 14 m2 SR3 248 m2
SR4 136,581 m2 SR4 136,581 m2 SR4
Storage 33 m2 Storage 33 m2 Storage 33 m2
Total required space 169,005 m2 Total required space 169,005 m2 Total required space 281 m2
27
Scenario 2 with 10% decrease in solar energy costs Scenario 2 with 25% decrease in solar energy costs Scenario 2 with 50% decrease in solar energy costs
Input Input Input
Demand 170,981 kWh Demand 170,981 kWh Demand 170,981 kWh
Budget € 25,000 Budget € 25,000 Budget € 25,000
Available space 400,000 m2 Available space 400,000 m2 Available space 400,000 m2
Output (production) Output (production) Output (production)
I3PC3 20,559 kWh I3PC3 20,720 kWh I3PC3 44,517 kWh
I4PC4 23,614 kWh I4PC4 23,614 kWh
I5PC5 126,976 kWh I5PC5 126,976 kWh I5PC5 126,976 kWh
Total production 171,149 kWh Total production 171,311 kWh Total production 171,492 kWh
Storage 168 kWh Storage 330 kWh Storage 511,2 kWh
Output (costs) Output (costs) Output (costs)
C3 € 7,022 C3 € 5989 C3 € 8,447
C4 € 4,970 C4 € 4,970 C4
C5 € 10,158 C5 € 10,158 C5 € 10,158
Storage € 1,972 Storage € 3876 Storage € 6,014
Total costs € 24,122 Total costs € 24,902 Total costs € 24,619
Output (spatial requirement) Output (spatial requirement) Output (spatial requirement)
SR3 203 m2 SR3 205 m2 SR3 440 m2
SR4 136,581 m2 SR4 136,581 m2 SR4
SR5 32,000 m2 SR5 32,000 m2 SR5 32,000 m2
Storage 221 m2 Storage 221 m2 Storage 221 m2
Total required space 169,005 m2 Total required space 169,006 m2 Total required space 32,661 m2
28
The results of the modeled scenarios show that the decrease in solar energy costs only causes serious changes to the control scenario when a decrease of 50 percent is simulated. At that point, geothermal energy is replaced by loose solar panels. This accounts for both the small and medium scenario, respectively scenario one and two. A 50 percent decrease in cost is however unlikely. Even if competition would increase heavily, a 50 percent drop is too steep to expect to happen in a short time period. Therefore it can be said that for small and medium initiatives geothermal electricity production is preferred.
29
4.3 Wind energy
Changes in the usage of wind energy in the city and province of Groningen is a plausible scenario, which is fairly similar to the solar energy scenario. Currently, there are already several wind farms in the province of Groningen, and several more (expansions) are planned. Two of the highest wind turbines in the Netherlands and the largest operational wind farm are located in the province of Groningen. Plans for additional wind parks are already approved which will yield another 441 MWh of electricity. Similar to solar energy, electricity from wind turbines could be used to produce hydrogen. For example during periods when there is excess wind-generated electricity.
Figure 2 Source: RTV Noord. Wind farm near Delfzijl.
Location-wise the wind turbines are more limited than solar panels. Wind turbines have a higher visual impact on the area and thus they are not placed near cities or (larger) villages. Also wind speeds and consistency influence where the wind turbines are built. This is also distinguishable in the figure below. Most wind turbines (or plans for wind turbines) are located in the North of Groningen near the sea or near low populated areas and highways.
Figure 3 Source: RTV Noord. Existing and planned wind parks in Groningen
30
turbines around Groningen is the appeal for placing wind turbines off-shore where a higher efficiency can be realized. The current wind energy potential and the planned developments can greatly influence the usage of wind energy in local green hydrogen initiatives. Similarly to the solar energy scenario the social factor of wind energy can decrease due to the social resistance towards wind turbines. This could cause a limit on the number of wind turbines to be built in the near future, a reduction on subsidies for wind energy, etc. It could be said that the increase of more wind turbines could also lead to a decrease in cost, similar to the solar scenario. However, it is assumed that wind turbines will not promptly be mass produced in and around Groningen to lead to a significant cost advantage.
Simulated wind scenarios
In the control scenarios one and two, wind energy is not used as a feedstock for hydrogen production. This is mainly due to the large electricity generating capacity of the selected wind turbine (average wind turbine in the Netherlands) and due to its large spatial requirement (where 1 km distance to the wind turbine is assumed). To ‘involve’ wind energy in the hydrogen production model the parameters ‘production capacity’ (PCi) and the ‘spatial requirement’(Sri) have to be altered. For a local hydrogen initiative, not all electricity generated by a wind turbine has to be used for producing hydrogen. Let’s assume that one percent of the generated electricity of a wind turbine is used for hydrogen production. There are varying opinions about the minimal distance of housing to a wind turbine. In the scenarios one and two, a distance of 1km is assumed resulting in a huge spatial requirement of over 3km2. If however the wind turbine is located remotely from houses the spatial requirement could change. The actual surface required for a wind turbine is limited due to the small foundation. Here a 1000 m2 is assumed, which is still spacious for the foundation alone, however, some safety distance to the structure is included. With these simulated parameters both scenario one and two do not use wind energy in its solution. This is due to the fact that geothermal hydrogen production has similar parameters to the simulated wind hydrogen production, and better matches the demand and budget. To ‘overtake’ geothermal as the main production method the cost of producing hydrogen via wind energy has to be reduced with at least 10 percent. These adjustments (lower production capacity, other spatial requirements, and lower costs) to wind scenarios were simulated based on the control scenarios. With these parameters, the model provides several solutions which are summarized in table 11 and 12. A complete overview of the output can be found in appendix ii.
31
Scenario 1 (control) Scenario 1 with simulated wind-energy parameters
Input Input
Demand 24,997.5 kWh Demand 24,997.5 kWh
Budget € 7,500 Budget € 7,500
Available space 200,000 m2 Available space 200,000 m2
Output (production) Output (production)
I3PC3 1,457 kWh I1PC1 22,280 kWh
I4PC4 23,614 kWh I3PC3 2,914 kWh
Total production 25,071 kWh Total production 25,194 kWh
Storage 71 kWh Storage 194 kWh
Output (costs) Output (costs)
C3 € 553 C1 € 4,010
C4 € 4,970 C3 € 1,106
Storage € 871 Storage € 2,306
Total costs € 6,391 Total costs € 7,422
Output (spatial requirement) Output (spatial requirement)
SR3 14 m2 SR1 1,000 m2
SR4 136,581 m2 SR3 29 m2
Storage 33 m2 Storage 33 m2
Total required space 136,628 m2 Total required space 1,061 m2 Table 11 Comparison scenario 1 as a control to scenario 1 with adjusted 'wind' parameters
32
Scenario 2 (control) Scenario 2 with simulated wind-energy parameters
Input Input
Demand 170,981 kWh Demand 24,997.5 kWh
Budget € 25,000 Budget € 7,500
Available space 400,000 m2 Available space 200,000 m2
Output (production)
I3PC3 20,559 kWh I1PC1 44,559 kWh
I4PC4 23,614 kWh I5PC5 126,976 kWh
I5PC5 126,976 kWh Total production 172,089 kWh
Total production 171,149 kWh Storage 554 kWh
Storage 167.6 kWh
Output (costs)
C3 € 7,802 C1 € 8,021
C4 € 4,970 C5 € 10,158
C5 € 10,158 Storage € 6,518
Storage € 1,972 Total costs € 24,697
Total costs € 24,902
Output (spatial requirement)
SR3 203 m2 SR1 2,000 m2
SR4 136,581 m2 SR5 32,000 m2
SR5 32,000 m2 Storage 221 m2
Storage 221 m2 Total required space 34,221 m2
Total required space 169,005 m2
Table 12 Comparison scenario2 as control to scenario 2 with adjusted 'wind' parameters
33
4.4 Biomass energy
Out of the twelve provinces of the Netherlands, Groningen is ranked second on the list of the relative highest GHG emissions per province. This is mainly due to the fact that Groningen houses two out the 15 biggest polluters (fossil fuel users) of the Netherlands. The biggest polluter of the Netherlands is the RWE power plant in Eemshaven which produces electricity for three and a half million households (1,560 mWh). The power plant is coal-based and burns 4 million tons of coal each year. Ranked thirteenth, the ENGIE Eemscentrale also contributes to the pollution in the province of Groningen by producing electricity from natural gas.
Figure 4 Coal based RWE power plant in Eemshaven. Source RTV Noord
One of the controversies surrounding the RWE power plant is that it became operational in 2015 while the Netherlands (and Groningen as a province) were already struggling to achieve the set climate goals. Current national policy concerning coal power plants is that they will be closed or converted to a non-polluting plant by the end of 2030. RWE is following this policy by slowly replacing coal with biomass. Starting in 2019, the power plant will function on 15 percent of biomass. In the near future, the factory should have completely substituted coal for biomass, at which point it will require 6 million tons per year. At ENGIE Eemscentrale a transformation to green(er) energy is taking place. Already two out of five gas based-electricity generators have shut down (however for economic reasons) and multiple wind turbines and a solar panel field has been placed on the nearby terrain.
The developments in this scenario could have several effects on local hydrogen initiatives. First of all the transition to biomass by the RWE power plant could create a vast amount of carbon-neutral energy which can be used for hydrogen production. The sheer quantity could cause biomass-based hydrogen cost to decline. If however the energy produced by the RWE power plant will be reserved for households only, the high demand for biomass feedstock could increase prices making local biomass-based hydrogen production unattractive.
Simulated biomass scenarios
34
plant will probably siphon electricity to the local hydrogen initiative based on its demand. The production capacity would thus precisely fit the demand which, combined with the low price of biomass, will make biomass the best possibility for hydrogen production by far. If however there is no possibility to use the electricity generated by the RWE plant, the local hydrogen initiative might have to compete with the RWE plant for its biomass feedstock. Not only will the RWE plant require a large amount of feedstock, but its requirement will also grow for the coming years making it an interesting and competing market. To compensate for this effect a 10, 20 and 50 percent costs increase is simulated for scenario 2. Scenario 1 is not re-simulated seeing that scenario 1 did not use biomass as a production method. With these parameters, the model provides several solutions which are summarized in table 13. A complete overview of the output can be found in appendix ii.
35
Scenario 2 with 10% increase in biomass energy costs Scenario 2 with 20% increase in biomass energy costs Scenario 2 with 50% increase in biomass energy cost
Input Input Input
Demand 170,981 kWh Demand 170,981 kWh Demand 170,981 kWh
Budget € 25,000 Budget € 25,000 Budget € 25,000
Available space 400,000 m2 Available space 400,000 m2 Available space 400,000 m2
Output (production) Output (production) Output (production)
I3PC3 20,397 kWh I3PC3 20,397 kWh
I4PC4 23,614 kWh I4PC4 23,614 kWh
I5PC5 126,976 kWh I5PC5 126,976 kWh
Total production 170,987 kWh Total production 6 kWh Total production No outcome
Storage 6 kWh Storage 170,987 kWh Storage
Output (costs) Output (costs) Output (costs)
C3 € 7,741 C3 € 7,741
C4 € 4,970 C4 € 4,970
C5 € 11,174 C5 € 12,190
Storage € 1,972 Storage € 1,972 Storage No outcome
Total costs €23,952 Total costs €24,968 Total costs
Output (spatial requirement) Output (spatial requirement) Output (spatial requirement)
SR3 SR3
SR4 SR4
SR5 SR5
Storage Storage Storage No outcome
Total required space 169,003 m2 Total required space 169,003 m2 Total required space
36
5. Case study (Groningen)
As befitting for a city that “believes in the hydrogen future”, Groningen already has a working green hydrogen initiative. Powered by one of the larger solar fields in the province of Groningen a hydrogen initiative has been active for a year now. The rest of this chapter is based on information provided by employees at Holthausen Groep and the municipality of Groningen.
Figure 5 Solar field near the city of Groningen containing 46,000 solar panels and is connected to a hydrogen station (Source:Groenleven)
5.1 Background of the case
The hydrogen initiative in Groningen is a cooperation between the municipality, Holthausen groep and the owners of the solar panel field (Groen-leven). The municipality of Groningen has decided that all logistics (cleaning, garbage removal, etc.) performed by the municipality in the city center has to become free of emissions within a few years. For that reason, it was determined that several hydrogen vehicles would be purchased/build creating a (local) demand for hydrogen. Around this time Holthausen groep was contacted by the municipality and asked if they could supply the hydrogen the municipality required. After some trouble with permits and locations, a hydrogen pump was built by Holhausen groep on an area provided by the municipality. So though the ground is owned by the municipality, the hydrogen station is owned by Holthausen Groep. The solar panel field ‘Woldjerspoor’ was built next to the terrain on which the hydrogen initiative was placed which provided the perfect opportunity to produce green hydrogen. Holthausen groep, in consultation with Groen-Leven, was able to directly connect the hydrogen initiative to the solar panel grid providing direct renewable energy. Due to the fact that producing hydrogen (by electrolyzer), compressing hydrogen and refueling vehicles takes very long, the municipality invested in a pressurized storage container which made high-pressure refueling possible. The project is thus a cooperation between three parties.
5.2 Case characteristics
37
store 140 KG of hydrogen. Additionally, the excess hydrogen is pressurized in smaller pressure tanks, which are sold in the commercial sector to small stations, industry, and laboratories. The hydrogen is sold for €10 per KG and, the cost of producing a KG is around €7. About a (tested) 100 KM can be driven on 1 KG of hydrogen and thus for a consumer the costs per KM are €0.10. Any negative social aspect of this hydrogen initiative is negligible due to the fact that the solar field is located on an old garbage dump on the outskirts of the city. There are no nearby households and other negative social aspects such as noise or smell are basically non-existent. There were however difficulties with permits which eventually led to the situation in which the municipality provided an area which was owned by them and had enough distance to households.
5.3 Comparison case and model
The case of the hydrogen initiative in the city of Groningen is fairly comparable to control scenario one in which the usage of hydrogen for vehicles was simulated. To test how the proposed model compares to a real life case the parameters of the case will be put into the model. After some recalculations, the case provides the following parameters.
Parameter Used metric Explanation
Demand 333,000 kWh 50 KG per day for 200 days a year (based on demand, not supply)
Budget €70,000 €7,00 per KG (based on costs)
Space 0.5 km2 Based on the size of control scenario 2
DSF 0.8 The hydrogen initiative is located far from households and thus some negative social aspects could be acceptable. Table 14 Parameters for case hydrogen initiative Groningen
