The potential of minimizing curtailment by implementing a hybrid energy system

The potential of minimizing

curtailment by implementing a

hybrid energy system

Analysis about the applicability of a hydrogen-battery storage systems on

local scale by minimizing curtailment and shortages.

Thijs Kuipers

1 Introduction

Moving towards alternative methods of generating electricity is inevitable as a result of fossil fuels reaching eventually to its end (Moka et al., 2014). Different methods for generating renewable electricity, such as solar and wind energy, are currently heavily investigated (Turconi, Boldrin, & Astrup, 2013). Renewable energy consists of two crucial characteristics due to its weather dependency. These two characteristics are the variability in accessibility and the intermittent generation of electricity, and consequently causing electricity surplus (Mazloomi & Gomes, 2012; Turconi et al., 2013). Hence, solar and wind energy are called variable renewable energy (VRE) sources. The issue with electricity surplus is that when it is not utilized it has to be shelved and results in curtailment (Chen et al., 2009).

An example of the extent that electricity surplus could have is shown in the research of Guti´errez-Mart´ın, Confente, and Guerra (2010). This study investigated a Spanish wind farm where they discovered that the farm had to be disconnected frequently from the power grid to maintain balance on the electricity network. Of the 48.8 MW that is delivered by the wind farm, 18.4 % could not be delivered and was an electricity surplus. Which, according to their calculations, consisted of 13 GWH of hydrogen per year.

There is no ideal energy storage system due to its various variations, specifications, and applications. Nevertheless, Cavallo (2001); Chen et al. (2009); Korpaas, Holen, and Hildrum (2003); McDowall (2006) acknowledge that electrical Energy Storage Systems (EESS) are crucial to electricity networks that include VRE sources. To provide standby power, store electricity peak loads, and scale down curtailment. Additionally, Cavallo (2001) argues that EESS should not only be added to the network but rather identifying the whole network as a unified system.

Currently, hydrogen storage systems are intensively investigated as a consequence of its great cooperation with renewable energy sources (Chen et al., 2009). Hydrogen storage is capable to cover large loads of electricity but is mainly used as seasonal storage (Bocklisch, 2015). A Hybrid Energy Storage System (HESS) could be implemented to store electricity for both short- and long-term. Batteries, especially lithium-ion, play an important role in HESS thanks to its rapid response to load changes, high energy efficiency (60-98), and suitableness with variable power. Such a system that combines hydrogen and batteries into one system is called a Hybrid Hydrogen Battery Energy Storage System (HHBESS) (Bocklisch, 2015; Chen et al., 2009).

Be that as it may, HHBESS has multiple issues. The main problem with HHBESS is the costs (due to high investment costs and operation costs), the size (the system requires a large area), and the actual benefits of the whole system (thanks to the low round-trip efficiency of the hydrogen storage) (Chen et al., 2009; Thien et al., 2015). The studies of Scamman, Newborough, and Bustamante (2015); Ubertini, Facci, and Andreassi (2017) showed that HHBESS is already applied on a very small scale (i.e. in the Telecom industry, concerning only several kW) and medium scale (i.e. hospitals and industrial plants concerning multiple MW), because the costs per kW of the HHBESS are in both cases low. It is not clear, however, if this conclusion also applies to the usage of HHBESS on a local scale, i.e. a village.

does a hybrid hydrogen-battery electricity storage system affect the curtailment and feed-in of electricity when it is implemented on a local scale?”.

This is done by comparing the functionality of a small electricity network that does not have an HHBESS to one including an HHBESS. Although the electricity network is mentioned as a united entity, this paper only focuses on the generation of electricity, the storage of VRE and meeting the demand. The VRE is generated by wind energy and the consumer of electricity is a single village. The main goal is to provide a better understanding of the suitability of an HHBESS in a village and how much this will reduce the curtailment of electricity.

This report starts with a theoretical background about HHBESS and the market it has to cooperate with. Followed by, the methodology chapter including the problem description, the model design, and the assumptions and limitations of the model. The next chapter presents the results of the experiments that have been done with the designed model. At last, the discussion on and conclusion of the findings are presented. 2 Related Work

This section is divided into two parts, the first part is about the market aspects and the second part is about the technical aspects. 2.1 Addressing the demand and supply

for variable renewable electricity Taking a glance at the past, the demand for electricity increased incrementally and so will the share of renewable energy do in the future Sawin et al. (2014). The study of Kondziella and Bruckner (2016) reports that the current storage is approximately 33 GW but claims that in the best case scenario 40-100 GW is required in Western Europe 2050. Meaning that there is a gap of 7-67 GW storage capacity. Looking at the future of our energy network, Marb´an and Vald´es-Sol´ıs

(2007) emphasizes that there will be two energy distribution networks operating the electric and the hydrogen network, requiring far more progressive distribution network structures. The essence of Marb´an and Vald´es-Sol´ıs (2007) statement is that it is likely that there will be a shift towards a more decentralized electricity network. This study contributes to this matter by analyzing the applicability of HHBESS on a local scale and therefore making it possible to determine whether HHBESS is a suitable option to such a decentralized electricity network.

Marb´an and Vald´es-Sol´ıs (2007); Mazloomi and Gomes (2012) mention that centralized electricity grids will be replaced with multiple sub-networks that are all connected. Considering hydrogen production Mazloomi and Gomes (2012) said that it will possibly be generated in gigantic plants and transmitted to distribution outlets, such as fuel stations, due to the high costs that come along with hydrogen storage. Regardless of what kind of shape the future energy distribution network will be both claim that the future distribution network is moving towards a decentralized and localized distribution network.

the most critical source of renewable back-up energy, with a limited potential of circa 10% of the average consumption. Hence the importance of investigation the potential of an HHBESS.

Urs´ua, San Mart´ın, Barrios, and Sanchis (2013) argues that wind and solar energies are the best options for integration with electrolyzers because of the focus and research that it received. One notable aspect to consider before evaluating renewable energy systems is the size and timescale of wind variability. Beaudin, Zareipour, Schellenberglabe, and Rosehart (2010) claims that small wind farms more often have hourly fluctuation compared to an entire area. As the share of variable renewable resources increases over time, so are the peak fluctuations that are caused by renewable energy resources Sawin et al. (2014). The study of Papaefthymiou and Dragoon (2016) argues that not only traditional concerns of supply adequacy are necessary to take into account but that overabundance is going to be fairly important to consider. One of the possible solutions is to use the renewable electricity surplus in other sectors (e.g. heat and transport). Contributing to the overall reduction of curtailment (Papaefthymiou & Dragoon, 2016).

2.2 The production, application, and storage of hydrogen

Hydrogen is a promising energy carrier or fuel and is largely investigated in (Balat, 2008; Ball & Wietschel, 2009; Guti´errez-Mart´ın et al., 2010; Korpaas et al., 2003; Luo, Wang, Dooner, & Clarke, 2015; Mazloomi & Gomes, 2012). Most quantities of renewable electricity surplus are stored by producing hydrogen with electrolysis, also known as power-to-gas (P2G) (Boudellal, 2018). Despite that this currently is the most common way to produce hydrogen, it consists of two core disadvantages that cause issues to implement hydrogen widely according to

G¨otz et al. (2016). The disadvantages are the relatively low round-trip efficiency and the high costs of producing and storing hydrogen. G¨otz et al. (2016) surely is right about these two disadvantages and it sheds insight on the difficulty of using hydrogen for the electrical storage method. Be that as it may, recent studies have shown that using a combination of multiple storage methods has a positive effect on these two disadvantages (Hemmati, Mehrjerdi, & Bornapour, 2020; Schreider & Bucher, 2018). This paper contributes to the existing knowledge by providing a better understanding of the potential of hydrogen and batteries when implementing on a local scale.

The studies of Ahluwalia, Papadias, Peng, and Roh (2019); Ball and Wietschel (2009); Boudellal (2018); G¨otz et al. (2016); Lund, Lindgren, Mikkola, and Salpakari (2015); Urs´ua et al. (2013) focus on hydrogen specific aspects. Claiming that the Alkaline water electrolyzer is the most developed and cheapest electrolyzer with efficiencies between 65 and 75% to produce hydrogen. Furthermore, they claim that hydrogen can be stored in various ways (i.e cavern storage, gaseous pressure tanks, liquefied tanks, metal hybrids, pipelines (by increasing the pressure), lined rock cavern, cryogenic storage).

pipeline network, at least not in its current state, but that is is possible to use a mixture that consists 90:10% of natural gas and hydrogen respectively.

2.3 Hybrid Hydrogen Battery Storage Systems

To demonstrate the potential of HHBESS Hemmati et al. (2020) analyzed the financial effect of adding batteries to hydrogen storage systems. According to their research, implementing a battery in hydrogen storage systems saves approximately 117,000 dollars per year. Similarly, Schreider and Bucher (2018) investigated the implementation of a 12.5 MW battery in a hydro energy storage system at Pfreimd hydro plant in Bavaria, Germany. Although the cost reduction was not included in their research, one of their conclusions was that the usage of a hybrid hydro plant in combination with a 12.5 MW battery increased the flexibility of the plant. Despite that Schreider and Bucher (2018) their research was not investigating an HHBESS it does show the suitability for and attention of HESS. Likewise, this study investigates the potential of a HESS but differentiates regarding the type of HESS, namely an HHBESS. Also, the size of the HHBESS that is required to provide the demand is different from these two types of research.

The difference between a hydrogen-only system and an HHBESS is that an HHBESS has a better round-trip efficiency which therefore requires lower total capacity. Also, an HHBESS has a higher response rate than hydrogen only (Scamman et al., 2015). The reason why an HHBESS is different compared to hydrogen-only is because of the high reliability, high flexibility, and high efficiencies (varying between 60 to 98%) of batteries (Chauhan & Saini, 2014; Scamman et al., 2015; Shabani & Andrews, 2015). Another benefit of batteries is that the costs have decreased by a fourth over the last 14 years due to the electrical vehicle market

(Schreider & Bucher, 2018). A drawback of batteries is that it requires around five times more often maintenance than hydrogen-only storage systems and it is not capable of storing a large quantity of electricity for a long period. Henceforth, maintenance costs are a large part of the total lifetime costs (Shabani & Andrews, 2015). Another drawback of batteries is that their durability is influenced by the State of Charge (SOC). Functioning incrementally longer when a high SOC is applied instead of a low SOC. However, combining these two storage methods results in less maintenance, higher efficiencies, and higher reliability (Scamman et al., 2015).

The previously mentioned studies show the wide range of attention on HHBESS and the variety of its application. Henceforth, I propose a scientific contribution by investigating the suitability of an HHBESS in combination with VRE on a local scale to identify the effect on the curtailment and the feed-in of the electricity network.

3 Methods

This section is divided into four parts, the first part elaborates the main problem that is addressed by this paper after that is the model design addressed The next part is about the assumptions and limitations of the model. The last part discusses the experimental setup. 3.1 Problem description

supplying the demand. One way to reconcile the supply and demand and utilize generated electricity is by storing the electricity surplus in an HBBESS.

How much influence this has on curtailment and shortage is analyzed by considering a wind farm that supplies the electricity demand of a village and compare it with a system that does not include an HHBESS. The design of such an HHBESS model is shown in figure 1 and will be discussed more in-depth in the next paragraph.

3.2 HHBESS design

The HHBESS starts with the generation of electricity by using wind turbines. The wind turbines supply to the direct current bus (DC bus) via a wild rectifier to stabilize the electricity output. The generated electricity primarily supplies the load of the village, which is connected to the DC bus via a converter.

However, when there is no demand for electricity the wind turbine supplies the battery and electrolyzer. The residual electricity is first supplied to the battery. Once the battery has reached the maximum state of charge (SOC) then the electricity is stored in the hydrogen tank. The reason for implementing a SOC improves the lifetime of the battery. The battery is supplied first because this is more suited for short-term storage and hydrogen storage is suited for seasonal storage. Also, the hydrogen tank has a lower round trip efficiency than the battery. However, if the hydrogen tank is also full and there is still an electricity surplus then the remaining electricity surplus is stored in the battery until it is fully charged. Because this study aims to reduce curtailment. If both battery and hydrogen tanks are full and there is still a surplus then this is curtailed. Meaning that this model is a stand-alone system that does not cooperate with an external grid.

Figure 1: HHBESS layout (Scamman et al., 2015)

On the other hand, when there is a power shortage the electricity is supplied first from the battery because the battery has a faster response rate, capable to store electricity for a shorter period, and has a higher round-trip efficiency (Blanco & Faaij, 2018). Once the battery has reached minimum SOC and there is still an electricity demand then the remainder is supplied from the hydrogen tank. A simplification of such an HHBESS design is shown in figure 1. For a detailed description of how the HHBESS design that is shown in figure 1 works please refer to chapter 3.3.

3.2.1 Village load Wind energy

Figure 2: Average electricity generation of 20 years (KNMI, 2020)

the lifetime of a hydrogen storage (Zakeri & Syri, 2015). To calculate the available electricity per wind speed formula (1) is used (Calculations, 2012). The electricity price is set at 0.24$ per kW (Rijksoverheid, 2020).

The village load is based on a village of 2960 households that requires averagely 2,832 kW per year per household (Nibud, 2020). Figure 3 shows the demand for 2960 households in kWh (Liander, April 20, 2020). Initially, this data sheet contained 10,000 households but this paper investigates a smaller village so the data-sheet is divided by 10,000 households and then multiplied with 2960 households. The location of the village is in Eelde, Netherlands (2960 households). The reason for choosing Eelde is to increase the reliability of the results because the wind station that provides the data of wind speed is located in Eelde. Another reason for analyzing the province Groningen is that wind energy is largely available both on-shore and off-shore (CBS, 2018). Groningen also functions as the gas network hub for whole Europe connecting pipeline networks of both nationally and internationally (Osborne, 2018). Making it ideally for hydrogen transportation via the gas pipeline network. Furthermore, (Ball Weeda 2015 example thesis) argue that from a geology perspective the Northern parts are considered to be suitable for underground hydrogen storage. Therefore, making Groningen a suitable location to implement an HHBESS.

3.2.2 Battery

This HHBESS design makes use of Lithium-ion batteries (Li-Lithium-ion) due to their high round-trip efficiency (circa 98%), fast response time, and low costs. The batteries contain a minimum and a maximum SOC of 20% and 80% respectively. To increase the lifetime performance of the battery (Hesse, Schimpe, Kucevic, & Jossen, 2017; Mongird et al., 2019).

3.3 Model and assumptions

An illustration of how the model operates is shown in figure 4. For the sake of clarity, the model will be discussed based on two aspects the supply side and the demand side. Note that this section does not go in detail about how each aspect is calculated only the number of the formula that has been used is mentioned. For more information regarding the calculations please refer to chapter 8 appendix A.

Figure 4: Decision model

3.3.1 The supply side of the model

Starting with the supply side of the model, the generated electricity primarily supplies the village its demand, after that the battery and finally the hydrogen tank are supplied. The battery and hydrogen tank are only supplied if there is an electricity surplus. The battery is only supplied until the maximum SOC is reached because this improves the lifetime performance of the battery. After that the battery is fully charged then the electricity surplus is supplied to hydrogen tank. If the hydrogen tank is also full and there is still an electricity surplus then the remainder is stored in the battery on top of the max SOC. The reason for neglecting the maximum

SOC in this situation is because having less curtailment is a higher priority in this study than improving lifetime performance. Also, as mentioned in the previous section, the electricity surplus has to go through the electrolyzer when storing it in a hydrogen tank. Finally, if there is still an electricity surplus after both the battery and hydrogen tank are fully supplied then this has to be dumped and results into curtailment.

3.3.2 The demand side of the model The second part, the demand side of the model, is based on a data sheet that contains average electricity demand per hour of a village. The wind turbines primarily deliver electricity to the households, but if the generated electricity is too low, then the ”power shortage” is supplied from the storage. First from the battery until the minimum SOC is reached and then from the hydrogen tank via a fuel cell until the minimum SOC is reached. If both the battery and the hydrogen tank are empty and there is still electricity needed then this is indicated by the system as ”shortage”. Although this is a continuous process, the model stops when demand is supplied or if a shortage has been reached.

3.3.3 Assumptions and limitations of the model

The model, as described above, is developed in Python. The time horizon of the simulations is a total of 20 years. The supply data sheet contains 20 different years but the demand data sheet only contains one year used 20 times. The reason for this is that it was not possible to obtain more demand

Parameter Setting Reference

Wind energy Standard wind turbine 2MW, costs of $2,2 mil each, amount of wind turbines set at 4, annually generating 49,000 MW (Islam et al., 2013; Kumar et al., 2016)

Electrolyzer Alkaline electrolyzer capacity set at 1,500 kW, with efficiency of 70 % (Chen et al., 2009; Klumpp, 2016; Mazloomi & Gomes, 2012) Hydrogen compressed

storage tank Capacity varying at range between 0-218,000 kW, costs 3055$/kWh (Zakeri & Syri, 2015)

Fuel cell Unlimitted with efficiency of 50 % (Chen et al., 2009; Mazloomi & Gomes, 2012; Scamman et al., 2015) Electricity price 0.24 $/kWh when selling to customer (Rijksoverheid, 2020)

Battery Lithium-ion battery, efficiency of 98 %, costs 469 $/kWh_{Life cycle: 3500 cycles, capacity ranging between 0 - 32,000 kW} (Hesse et al., 2017; Mongird et al., 2019)

Minimum SOC Min.SOC battery = 20 %, min.SOC hydrogen = 10 % (Chen et al., 2009; Mazloomi & Gomes, 2012; Mongird et al., 2019; Warner, 2015) Maximum SOC Max.SOC battery = 80 %, max.SOC hydrogen = 100 % (Chen et al., 2009; Mazloomi & Gomes, 2012; Scamman et al., 2015; Warner, 2015) Size village 2960 households, each household = average of 2,832 kW/per year (Liander, April 20, 2020)

data sheet of different years.

The amount of wind turbines is set at 4 because then it is capable of generating just as much as the total electricity demand. Each wind turbine generates circa 49,000,000 kW over twenty years and the total demand of a village of 2960 is 176,183,654 kW. The capacity of the electrolyzer is set at 1,500 kW to make sure that it is possible to store the average electricity surplus of 269 kWh for each of the four wind turbines. The battery capacity cannot be larger than 32 MW. This is based on (Zakeri & Syri, 2015) which mentioned that currently, the largest Li-ion storage is 32MW. For an overview of all the parameters used in this model please refer to table 1 in chapter 3.4

This model also contains several limitations. The first limitation is that transportation is neglected for the sake of simplicity.Next limitation is that hydrogen is only used for electricity purpose and not for other applications (i.e methanation, heating, or fuels). Furthermore, the self-discharge rate is not included because this is very low per hour (2.5 % per month for the battery). Next, the costs of the HHBESS are only for indication purpose implemented because including all cost aspects and investigating them is a whole research by itself. It does, however, indicate the suitability of an HHBESS on a local scale. Also, the model is not capable to address issues such as environmental friendliness, safety, space, etc. Solely financial and technical aspects are included such as the costs ($), revenue ($), capacity (kWh), etc. The sixth limitation is the model does not take the wild rectifier and the converter into account because of the sake for simplicity. The last limitation is that there the designed model is not capable of recognizing short term storage (batteries) and seasonal storage (hydrogen).

3.4 Experimental setup

In this study, the base case includes the same model as described in chapter 3.2 and works the same as described in chapter 3.3 only without any storage applications. Leaving out the storage applications makes it possible to compare it to when the model includes an HHBESS and determine the differences.

The model is used to calculate how much the curtailment, shortages, costs, revenue, and electricity delivered from storage are when increasing the total storage capacity. Before being able to find these variables and make a substantiated conclusion, it is first necessary to determine the ”base case”. The results of the base case are shown in table 2 in chapter 4.1, including both several parameters and variables for clarity purpose. The variables include curtailment, shortage, watt from storage, costs, revenue, and costs -revenue. In which curtailment and shortage are both mentioned twice. One including the capacity in kW and one as a percentage of the total generated electricity for the curtailment and a percentage of the total demand for the shortage. Consequently, the parameters of the model that are shown are the number of wind turbines, electrolyzer capacity and efficiency, power demand, electricity generated, and electricity delivered from wind turbines.

4 Findings

This chapter section provides the main results that are related to answer the research question of this paper. So, identifying the relation between the HHBESS and the curtailment and shortage of electricity on a local scale is mainly discussed. All simulations are done over a twenty year time period. Starting with the explanation of the base case, followed by the elaboration on the curtailment, shortage, watt delivered from storage and cost, and finally a sensitivity analysis applied to determine the reliability of the findings.

4.1 The base case

The base case is a representation of the electricity network without an HHBESS implemented, meaning that the electricity surplus cannot be stored and has to be curtailed. Looking at the results that are shown in figure 2 it becomes clear that more than half of the generated electricity is tossed away and almost half of the demand is not delivered. Showing that there is a lot room left for improvement which us the essence of the study of Guti´errez-Mart´ın et al. (2010).

Furthermore, a large difference between the percentage hours that have a shortage and the percentage of watt shortage can be noted. Indicating that there are more days having a relatively small shortage.

An interesting result of the base case is that even though the curtailment and shortages are high it is still profitable to supply the village with electricity. This can be traced back to relatively low costs of a standard 2MW wind turbine. The profitability of using a wind park signifies the financial attractiveness of using a wind park to supply the wind park.

So, it safe to say that the base case provides a good picture of the potential and at the same time the problem of having a variable resource supplying the village demand.

Variable/parameter Result Curtailment (%) 52.9 Cost-revenue ($) -10,545,774 Costs ($) 11,804,955 Revenue ($) 22,350,729 Wind turbines 4 Battery capacity (kW) 0 Tank capacity (kW) 0 Electrolizer capacity (kW) 1,500 Electrolizer efficiency (%) 0.7 Power demand (kW) 176,183,654 Electricity generated (kW) 197,915,853 Watt from storage (kW) 0 Watt directly from wind turbines (kW) 93,117,126 Curtailment (kW) 104,779,984 Total watt shortage (kW) 83,055,617 Hours with shortages (%) 67.6 Watt shortage (%) 47.1

Table 2: Results base case

4.2 The curtailment and shortage

Implementing an HHBESS has a positive affect on the reduction of the curtailment and shortage of electricity which is shown in figure 5. Whereas the total capacity of the HHBESS increases, both the curtailment and shortage decrease. Especially, the first 30,000 kW storage capacity provides a significant difference compared to the base case of almost 20 %. The reason for causing this effect is that the Li-ion battery has a maximum of 32,000 kWh (corresponding with the largest size Li-ion battery in 2015). Dropping the curtailment almost with 20% by implementing 30,000 kW storage capacity.

This is possible because the model does not differentiate between the storage time a certain storage method has. Nevertheless, it can be assumed that both curtailment and shortage do not go below the 20% for a village of 2,960 households. Another reason for not being able to get below the 20% is because the electrolyzer capacity is set at 1,500 kW. Meaning that it is not possible to store more than 1,500 kW no matter how large the generated electricity is. A sensitivity analysis should be done on the electrolyzer capacity whether it is possible to have 0% curtailment and shortage if the electrolyzer capacity is increased.

Figure 5: Curtailment and shortage as percentage of total generated electricity

Furthermore, increasing the storage capacity also has a positive influence on making the feed-in smoother thanks to the increasing electricity that is supplied from the storage. In the base case it became clear that there was a electricity shortage for almost half the time, which represents a poor electricity feed-in. Implementing an HHBESS with a larger storage capacity is capable to utilize the generated electricity better and therefore provide a more constant feed-in to the grid. Figure 6 shows the total electricity supplied from the HHBESS to supply the demand of the village. However, as the curve of the graph shows it becomes clear that just increasing storage capacity does not equally increase electricity delivered from storage.

Finally, the cost of an HHBESS increase

almost linear and is dependable on the size of the total capacity of the HHBESS, as shown in figure 7. Once the total storage capacity of the HHBESS reaches 70,000 kW each additional 10,000 kW reduces both curtailment and shortage less than 1%, as shown by figure 5. Therefore indicating that the most suitable capacity is not the largest storage capacity. When looking at figure 7 the costs of an HHBESS that has a capacity of 70,000 kW is 158,666,679 $. In fact when looking to costs it becomes clear that the largest capacity is definitely not the most suitable option. For instance, when the total capacity is doubled (140,000 kW) the costs increase 2.34 times. All things considered, making an HHBESS with a capacity of 70,000 kW the most suitable capacity for this village.

Figure 6: watt delivered from storage to supply demand

beneficial aspects. Including the reduction of the curtailment and shortage for at least 20%, depending on the storage capacity, and making the electricity feed-in smoother. However, a nuance has to be made on this implication namely that it is not possible to stabilize the electricity network solely by implementing an HHBESS, keeping in mind that electricity network is supplied by VRE. Possible reason for the limitation on the potential of an HHBESS is the low round-trip efficiency.

4.3 Sensitivity analysis

First, the electrolyzer capacity is assumed to have an influence curtailment, shortage and electricity delivered from storage is the electrolyzer capacity. To determine whether this hold true and if so how much, the electrolyzer capacity is increased to 10,000kW instead of 1,500kW. The result of having a higher electrolyzer capacity indicates that it becomes possible to decrease the variables. However, still not making it possible to decrease the curtailment and shortage to 0% within the 0 - 250,000 kW storage capacity range, as shown in figure 8.

Figure 8: Capacity electrolyzer set at 10,000 kW

The difference between figure 8 and 5 is that at figure the beginning decreases faster, while figure 8 decreases more gradually. Meaning that it takes more storage capacity to reduce the curtailment. However, in this case it is possible to get below 20%, indicating that a higher electrolyzer capacity in combination with HHBESS would lead

to a lower curtailment and shortage. Also, increasing the capacity of the electrolyzer has less effect on shortage than it has on curtailment, which makes sense because the electrolyzer is the factor between the hydrogen tank and the supply side.

The second factor that is assumed to play a role in determining the height of curtailment, shortage and electricity delivered from storage is the relatively low round-trip efficiency. To find out if the limited extend of the potential of an HHBESS is caused by the low round-trip efficiency a sensitivity analysis is applied in which the round-trip efficiency of hydrogen is 70% instead of 35%. This is achieved in the model by increasing the efficiency of the fuel cell to 100% instead of 50%, hence increasing the total round-trip efficiency to 70%. The result of having a higher round-trip efficiency results into a lower curtailment and shortage and a higher amount of electricity that is delivered from storage. Especially, the shortage reduces considerably as shown in figure 9.

The difference between figure 9 and 5 is that figure 9 shows that the curtailment drops less but that the shortage drops more. Meaning that it has less effect on the curtailment but a larger effect on the shortage. Which makes sense, because the the fuel cell is the factor between the hydrogen tank and the demand side.

Figure 9: Efficiency fuel cell set at 100% 5 Discussion

influence on the reduction of both curtailment and shortage compared to the base case. Also, that an HHBESS is only capable to reduce the curtailment to 21%. Indicating that HHBESS can not solely be implemented when striving for a curtailment of zero percent or a fully sustainable electricity network. Reflecting this on the reality then this is already the case. Although fossil fuel is the largest resource it is not the only one (i.e nuclear energy, wind energy, solar energy, hydro energy, etc.)(Moka et al., 2014; Turconi et al., 2013). As Kondziella and Bruckner (2016) already claimed it is necessary to have a cooperation of multiple methods for generating energy, storing energy, transporting energy and using energy to maintain a stable and manageable electricity distribution network. Which is supported by the findings of this report.

The second finding that is worth mentioning is that HHBESS is most beneficial when it does not exceed a total capacity of 70,000 kW. After that the curtailment and shortage decrease less than 1% per 10,000 kW addition capacity while the costs increase per additional 10,000 kW. Despite the fact that the total curtailment and shortage are reduced 50% at a capacity of 70,000 kW there is still circa 20% of the electricity considered as a shortage and curtailed. Altogether indicating that an HHBESS is applicable on a local scale but is not a ”perfect fit” for a village of 2960 households.

All in all, the findings of this study confirm existing theory because the round-trip efficiency and the costs are also the most challenging factors of an HHBESS. The contribution of this study to existing theory is the fact that the most suitable storage capacity to supply a village of 2960 households is 70,000 kW.

Future research

A suggestion for future research would be to investigate if the limit of 20% curtailment and shortage can be removed by implementing an

additional resource. One of the characteristic that the extra energy resource should have is that it does not have the same generating time as wind energy. Then it becomes possible to determine if these two energy sources are capable of supporting each other. Another suggestions would be to analyse what the influence would be when a resources is added that provides a more stable electricity output, for instance biomass energy.

6 Conclusion

Interestingly is the discovery that no matter how large the total capacity of the HHBESS is the curtailment and shortage will never get below the 20%. This conclusion only holds true for the HHBESS that is designed with the same parameter settings as discussed in this paper. The sensitivity analysis indicated that both the round-trip efficiency and the capacity of the electrolyzer made it possible for to get below the 20% curtailment and shortage. Despite the fact that the curtailment and shortage are both reduced with more than 50% compared to the base case, there is still a considerable percentage remaining after this reduction. This indicates that an HHBESS is not suitable for for a network that does not include other storage methods or multiple resources operating in it.

Furthermore, the most suitable capacity of the HHBESS is determined at 70,000 kW because after this point each additional step of 10,000 kW decreases the curtailment and shortage with less than 1%. The costs of an HHBESS with a capacity of 70,000 kW is $158,666,679.

References

Ahluwalia, R., Papadias, D., Peng, J., & Roh, H. (2019). System level analysis of hydrogen storage options.

Balat, M. (2008). Potential importance of hydrogen as a future solution to environmental and transportation problems. International journal of hydrogen energy, 33(15), 4013–4029. Ball, M., & Wietschel, M. (2009). The

hydrogen economy: opportunities and challenges. Cambridge University Press.

Beaudin, M., Zareipour, H., Schellenberglabe, A., & Rosehart, W. (2010). Energy storage for mitigating the variability of renewable electricity sources: An updated review. Energy for sustainable development, 14(4), 302–314.

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.

Bocklisch, T. (2015). Hybrid energy storage systems for renewable energy applications. Energy Procedia, 73(2015), 103–111.

Boudellal, M. (2018). Power-to-gas: Renewable hydrogen economy for the energy transition. Walter de Gruyter GmbH & Co KG.

Calculations, W. T. P. (2012). Rwe npower renewables. Mechanical and Electrical Engineering Power Industry, The Royal Academy of Engineering.

Cavallo, A. J. (2001). Energy storage technologies for utility scale intermittent renewable energy systems. J. Sol. Energy Eng., 123(4), 387–389. CBS. (2018). Hernieuwbare energie in

nederland 2018. Retrieved from https://longreads.cbs.nl/ hernieuwbare-energie-in -nederland-2018/

Chauhan, A., & Saini, R. (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.

Chen, H., Cong, T. N., Yang, W., Tan, C., Li, Y., & Ding, Y. (2009). Progress in electrical energy storage system: A critical review. Progress in natural science, 19(3), 291–312.

G¨otz, M., Lefebvre, J., M¨ors, F., Koch, A. M., Graf, F., Bajohr, S., . . . Kolb, T. (2016). Renewable power-to-gas: A technological and economic review. Renewable energy, 85, 1371–1390. Guti´errez-Mart´ın, F., Confente, D., & Guerra,

I. (2010). Management of variable electricity loads in wind–hydrogen systems: the case of a spanish wind farm. International Journal of Hydrogen Energy, 35(14), 7329–7336. Hemmati, R., Mehrjerdi, H., & Bornapour,

M. (2020). Hybrid hydrogen-battery storage to smooth solar energy volatility and energy arbitrage considering uncertain electrical-thermal loads. Renewable Energy.

Hesse, H. C., Schimpe, M., Kucevic, D., & Jossen, A. (2017). Lithium-ion battery storage for the grid—a review of stationary battery storage system design tailored for applications in modern power grids. Energies, 10(12), 2107. Islam, M. R., Mekhilef, S., & Saidur, R.

(2013). Progress and recent trends of wind energy technology. Renewable and Sustainable Energy Reviews, 21, 456–468.

Klumpp, F. (2016). Comparison of pumped hydro, hydrogen storage and compressed air energy storage for integrating high shares of renewable energies—potential, cost-comparison and ranking. Journal of Energy storage, 8, 119–128.

https://projects.knmi.nl/ klimatologie/uurgegevens/ selectie.cgi

Kondziella, H., & Bruckner, T. (2016). Flexibility requirements of renewable energy based electricity systems– a review of research results and methodologies. Renewable and Sustainable Energy Reviews, 53, 10–22.

Korpaas, M., Holen, A. T., & Hildrum, R. (2003). Operation and sizing of energy storage for wind power plants in a market system. International Journal of Electrical Power & Energy Systems, 25(8), 599–606.

Kumar, Y., Ringenberg, J., Depuru, S. S., Devabhaktuni, V. K., Lee, J. W., Nikolaidis, E., . . . Afjeh, A. (2016). Wind energy: Trends and enabling technologies. Renewable and Sustainable Energy Reviews, 53, 209–224.

Liander. (April 20, 2020). Electriciteit 2009. Retrieved from https:// www.liander.nl/partners/ datadiensten/open-data/ data

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.

Luo, X., Wang, J., Dooner, M., & Clarke, J. (2015). Overview of current development in electrical energy storage technologies and the application potential in power system operation. Applied energy, 137, 511–536.

Marb´an, G., & Vald´es-Sol´ıs, T. (2007). Towards the hydrogen economy? International journal of hydrogen energy, 32(12), 1625–1637.

Mazloomi, K., & Gomes, C. (2012). Hydrogen as an energy carrier: Prospects and challenges. Renewable

and Sustainable Energy Reviews, 16(5), 3024–3033.

McDowall, J. (2006). Integrating energy storage with wind power in weak electricity grids. Journal of Power sources, 162(2), 959–964.

Moka, S., Pande, M., Rani, M., Gakhar, R., Sharma, M., Rani, J., . . . others (2014). Alternative fuels: an overview of current trends and scope for future. Renewable and Sustainable Energy Reviews, 32, 697–712.

Mongird, K., Viswanathan, V. V., Balducci, P. J., Alam, M. J. E., Fotedar, V., Koritarov, V. S., & Hadjerioua, B. (2019). Energy storage technology and cost characterization report(Tech. Rep.). Pacific Northwest National Lab.(PNNL), Richland, WA (United States).

Nibud. (2020). Gemiddeld energie en water verbruik. Retrieved from https:// www.nibud.nl/consumenten/ energie-en-water/

Osborne, A. (2018). Buisleidingen in nederland een marktverkenning. Retrieved from https:// www.rijksoverheid.nl/ documenten/rapporten/ 2018/09/03/buisleidingen -in-nederland-een -marktverkenning

Papaefthymiou, G., & Dragoon, K. (2016). Towards 100% renewable energy systems: Uncapping power system flexibility. Energy Policy, 92, 69–82. Petten, J. (2014). ”putting science

into standards: Power-to-hydrogen and hcng”.

Rijksoverheid. (2020). Electriciteit prijs. Retrieved from https://energiecijfers .databank.nl/dashboard/ energieprijzen

Scamman, D., Newborough, M., & Bustamante, H. (2015). Hybrid hydrogen-battery systems for renewable off-grid telecom power. International Journal of Hydrogen Energy, 40(40), 13876–13887.

Schreider, A., & Bucher, R. (2018). An auspicious combination: Fast-ramping battery energy storage and high-capacity pumped hydro. Energy Procedia, 155, 156–164.

Shabani, B., & Andrews, J. (2015). Standalone solar-hydrogen systems powering fire contingency networks. International Journal of Hydrogen Energy, 40(15), 5509–5517.

Singh, S., Jain, S., Venkateswaran, P., Tiwari, A. K., Nouni, M. R., Pandey, J. K., & Goel, S. (2015). Hydrogen: A sustainable fuel for future of the transport sector. Renewable and Sustainable Energy Reviews, 51, 623– 633.

Steinke, F., Wolfrum, P., & Hoffmann, C. (2013). Grid vs. storage in a 100% renewable europe. Renewable Energy, 50, 826–832.

Thien, T., Axelsen, H., Merten, M., Zurm¨uhlen, S., M¨underlein, J., Leuthold, M., & Sauer, D. U. (2015). Planning of grid-scale battery energy storage systems: Lessons learned from a 5 mw hybrid battery storage project in germany. In Proc. battcon int. battery conf (pp. 18–1).

Turconi, R., Boldrin, A., & Astrup, T. (2013). Life cycle assessment (lca) of electricity generation technologies: Overview, comparability and limitations. Renewable and sustainable energy reviews, 28, 555–565.

Ubertini, S., Facci, A. L., & Andreassi, L. (2017). Hybrid hydrogen and mechanical distributed energy storage. Energies, 10(12), 2035.

Union, E. (2018). Climate and energy framework policy (2030). Retrieved from https://ec.europa.eu/

clima/policies/strategies/ 2030 en

Urs´ua, A., San Mart´ın, I., Barrios, E. L., & Sanchis, P. (2013). Stand-alone operation of an alkaline water electrolyser fed by wind and photovoltaic systems. International journal of hydrogen energy, 38(35), 14952–14967.

Warner, J. T. (2015). The handbook of lithium-ion battery pack design: chemistry, components, types and terminology. Elsevier.

7 Acknowledgements

8 Appendix A: Used formulas

To calculate available power at different wind speeds the following formula is used:

P ower = 1/2pA(v3)C (1) In which, Power = watt in kW, p = air density in kg/m3, A = swept area in m2, v = wind speed in m/s,

C = power coefficient in decimals.

To calculate total power of wind the following formula is used:

T otal power wind turbine = power per wind turbine ∗ amount of wind turbines (2) In which,

Power per wind turbine = in kWh

To calculate power delivered to households the following formula is used:

P ower delivered directly f rom wind turbine =X(T otal power wind turbine−demand households) (3)

In which, Demand households is in kWh and is never below zero. Also, to have total power delivered directly from wind turbine it is necessary to take the sum of all hours. If demand is higher then electricity is delivered from storage.

To calculate electricity surplus the following formula is used:

Electricity surplus = generated electricity − demand households (4) In which,

electricity surplus is equal or higher than zero.

To calculate amount of generated electricity delivered to storage the following formula is used:

Electricity delivered to battery = electricity surplus (5) And,

electricity delivered to hydrogen tank = Electrolyzer ef f iciency ∗ electricity surplus (6) And,

T otal electricity stored =X(electricity delivered to battery+electricity delivered to tank) (7) In which,

Electricity surplus = is kWh.

The conditions are that there has to be a electricity surplus, is space available in storage, and cannot be higher than tank available tank capacity.

To calculate curtailment the following formula is used:

Curtailment = electricity surplus − total electricitydelivered to storage (8) To calculate electricity delivered from storage the following formula is used:

electricity delivered f rom storage = total actual storage value − power shortage (9) In which both are in kWh. The conditions are that actual storage value cannot be below SOC and electricity delivered from storage cannot be below zero. If it would be below zero then there is a real shortage indicated as ”shortage” instead of ”power shortage”.

To calculate shortage the following formula is used:

Shortage = generated electricity + actual electricity in storage − demand households (10) In which all three aspects are in kWh and the condition is that the outcome is shown as an absolute number.

To calculate revenue the following formula is used:

Revenue =X(demand households) −X(shortages) ∗ electricity price (11) In which, demand households and shortages are in kWh and electricity price in $.

To calculate the total costs the following formula is used:

T otal costs = Investment battery + replacement cost+

investment hydrogen tank + investment electrolyzer+ costwindturbines − revenue

(12)

In which,

All prices per watt are in $/kWh, all capacities are in kWh.

Investment battery = battery capacity ∗ battery capacity price (13)

Replacement cost = battery price per watt∗battery capacity total hours/total hours year) (14) Hydrogen tank investment = tank capacity ∗ tank price per W att (15)

Investment electrolyzer = Electrolyzercapacity ∗ electrolyzerpriceperW att (16)