Wordcount: 5910
Premaster’s Thesis
Prevent the current electricity distribution
network from congestion and reduce energy
curtailment
A simulation model on how a centralized battery storage should be sized, to
prevent the current electricity distribution network from congestion and reduce
energy curtailment
22nd June 2020
Pre-Msc Supply Chain Management University of Groningen
Faculty of Economics and Business Nettelbosje 2, P.O. Box 800
2
Summary
The electricity distribution network experiences difficulties, due to the limits of load capacity. Nowadays, these difficulties become more visible because of peaks in supply of renewable energy. These peak-supply moments are caused by the intermittency, variability, and unpredictability of renewable energy resources, and rising number of, for example, photovoltaic solar parks. This research aims to prevent the load capacity from congestion and have zero energy curtailment, by using a
centralized battery storage. To determine how this centralized battery should be sized, the next research question has been formulated: How should a centralized battery storage system be sized, to
prevent the current electricity distribution network from congestion?
The research question has been answered with the use of a simulation model. Results of this simulation model shows that a photovoltaic solar park needs a centralized battery storage, which has the kWh of 2,8% of its total yearly produced renewable energy, to prevent the medium voltage electricity network from congestion and have no energy curtailment at all. Besides, the sensitivity analysis shows that an electricity cable with more load capacity, results in a smaller centralized battery storage considerably. Although, this effect becomes negligible from an electricity cable with a load capacity of 3090 KVA (Kilovolt-Ampere).
Keywords: Photovoltaic Energy, Battery Storage, Congestion, Load Capacity, Curtailment
3
Content
Summary ... 2
Tables and Figures... 4
1 Introduction ... 5 2 Theoretical background ... 7 3 Methodology ... 9 3.1 Problem description ... 9 3.2 Conceptual model ... 9 3.2.1 Objective ... 10 3.2.2 Parameters ... 10 3.2.3 Variables ... 11 3.2.4 Content ... 11 3.2.5 Component list ... 12
3.2.6 Logic flow diagram ... 13
3.3 Experimental setup ... 14
3.3.1 Base case ... 14
3.3.2 Sensitivity analysis ... 15
4 Results ... 17
4.1 Results base case ... 17
4.2 Simulations ... 18
4.2.1 Energy curtailment ... 18
4.2.2 Battery storage requirements ... 19
4.3 Sensitivity analysis ... 20
4.4 Discussion ... 22
5 Conclusion ... 24
References ... 25
Appendices ... 28
Appendix A: N2XSEY 10KV Cable information ... 28
Appendix B: Base case numerical results ... 30
4
Tables and Figures
Table 1 - Parameters ... 11
Table 2 - Variables ... 11
Table 3 - Component list ... 13
Table 4 - Components and values for simulations ... 15
Table 5 - Cable data for sensitivity analysis ... 16
Table 6 - Results sensitivity analysis... 21
Table 7 - Battery storage capacity per N2XSEY 10KV cable ... 22
Table 8 - Results base case ... 30
Table 9 - Results simulations ... 31
Figure 1 - Simplified conceptual model ... 10
Figure 2 - Simplified simulation model ... 12
Figure 3 - Logic flow diagram ... 14
Figure 4 - Battery storage in-use, results of base case... 17
Figure 5 - Monthly energy curtailment in kW, results of the base case ... 18
Figure 6 - Yearly energy curtailment in kW, results of the simulations... 19
Figure 7 - Battery storage requirements, results of simulations ... 20
5
1 Introduction
Today’s energy sources are largely based on fossil fuels, but those will not last forever and are one of the main reasons for current environmental issues (Destouni & Frank, 2010). Reducing the contribution of fossil fuels is one of the greatest challenges for the world’s community. The industry for renewable energy develops new technologies, which has become the focal center of many
researchers across the globe. Renewable energy sources, such as solar or wind, become more common nowadays. All these, and upcoming developments should meet the current needs of society, without compromising the ability of future generations to meet their own needs (Prabakar, et al., 2015; Ray, 2019).
Photovoltaic (PV) solar energy is an example of transforming energy by using a renewable source. It converts solar energy, as a renewable source, directly into energy. This renewable way to produce energy is expected to play a major role in supplying clean and environment-friendly energy, in the 21st century. The PV solar systems can be installed in smaller and larger entities. Smaller entities
are called PV stand-alone-systems installed for a specific household. On a larger scale, there are PV solar parks that produce renewable energy. The produced amount depends on size, location, and availability of sun. These PV solar parks are often located in agricultural areas, while the city center has a higher electricity demand (Kumar, et al., 2018). These locations for solar parks are often cheaper and less covered with shadow from buildings, which makes the PV systems produce more efficiently (Zeman, et al., 2007).
One of the current challenges is that supply and demand of renewable electricity do not occur at the same place, but there is a physical distance in between. Distribution via the current electricity network is required to overcome the distance between supply and demand, but this network has its limits. The maximum load capacity of the current electricity network is not always able to distribute all of the produced renewable energy to its consumers. Especially during peak hours, when the sunshine is at its strongest, which is normally around mid-day (Kosowatz, 2018). Researchers, for example, Bin, et al., (2018) and Goop et. al, (2017) even claim that new initiatives for renewable energy supply cannot be realized, because of the lack of load capacity in the current electricity distribution network. This means that expanding renewable energy supply is not possible before enhancing the electricity network, which is costly and time-consuming (Bin, et al., 2018).
6 manage oversupply and overcome electricity congestion. These strategies are (1) storage, (2) western Energy Imbalance Market (EIM) expansion, (3) demand response, (4) regional coordination, (5) time-of-use rates, (6) electric vehicles, (7) minimum generation and (8) flexible resources (California ISO, 2017). To protect the load capacity of the current electricity distribution network from congestion, the produced energy can also be curtailed. The consequence of curtailment is that overproduction at peak-supply of electricity by renewable sources have to be thrown away (Bird, et al., 2014).
This research focusses on the sizing of a centralized battery storage, in order to prevent the current electricity network from exceeding the load capacity and reduce energy curtailment. The next research question has been formulated: How should a centralized battery storage system be sized, to
prevent the current electricity distribution network from congestion?
This paper is organized as follows. First, the theoretical background of this research, which shows what already has been done in this area and it addresses the differences between existing research and this paper. Secondly, the methodology section will be presented. This includes an
7
2 Theoretical background
This paper aims to size a centralized battery storage and store renewable energy when the electricity distribution network is in use. This helps to prevent the current electricity distribution network from congestion and it reduces energy curtailment as well. This paper is novel in the fact that it combines these two objectives, by using a battery storage, to create flexibility in supply and
distribution. Congestion, curtailment, and battery storages have been researched separated before, but combining them all into one research makes this paper adding value to existing work, done by other researchers. The following two paragraphs will give background information about the subject and thereafter existing research and the differences with this paper will be discussed.
Renewable energy, generated by solar power, has to be integrated into the transmission grid, to overcome the geographical distance between supply and demand (Heide, et al., 2010). The lack of constant supply of energy from renewable sources in solar power, and other renewable sources, leads to intermittency, variability, and unpredictability of supply (Comello & Reichelstein, 2019; Keck, et al., 2019). These three characteristics of renewable energy supply cause a high level of penetration at certain times in the transmission grid, which is called electricity congestion.
Electricity congestion in the transmission grid is currently an important issue in the integration of variable renewable energy (Goop, et al., 2017). Fluctuations in renewable energy supply occur throughout the day and output differ per season, while the load capacity of the electricity network is not able to adapt to these fluctuations. The load capacity cannot be overwritten (Rumpf, 2020). Greater penetration of variable renewable energy on electric grids will result in an increased level of energy curtailment (Biggar & Hesamzadeh, 2014; Bird, et al., 2016). Curtailment is the reduction of output, of a generator, from what it could have produced with the available renewable resources. It occurs on an involuntary basis and became a normal occurrence since the electric power industry (Bird, et al., 2014; Golden & Paulos, 2015).
8 grid in its sensitivity analysis only, to see how the battery storage capacity changes, when expanding the electricity grid. A sensitivity analysis on mutual effects between these factors has not been done before.
Research about a battery storage has been done before, by Elhadidy & Shaahid, (1999), Dai, et al., (2016) and Pan, et al., (2015). These papers focussed on battery storage capacity considering different parameters, but did not taken the load capacity of the current electricity network into account. These three papers had as main aim to gather insights into the supply, instead of supply and
distribution. Therefore, these papers will be used for this paper to gather an insight into the actual supply process. This research’s target is to generate supply and distribute the energy through the current electricity distribution network, to its consumers. It adds value to researches of Elhadidy & Shaahid (1999), Dai, et al., (2016) and Pan, et al., (2015), by taking the distribution into account.
Qian, et al., (2020) addresses the challenge of load capacity in the distribution network. Their paper aimed to calculate the power supply capacity for the distribution network, based on the load distribution constraints, but did not address a centralized battery storage. Qian, et al., (2020) concluded that line and transformer load balancing measures can significantly improve the power supply capacity of a distribution network. The research of Qian, et al., (2020) will be used for the challenges that occur in the distribution process. By adding the centralized battery storage in this paper, the aim is to solve load capacity distribution challenges and create flexibility in supply and distribution.
9
3 Methodology
3.1 Problem description
The main problem addressed in this paper is the congestion in the electricity distribution network, due to increasing peak-supply which occurs because of the increasing number of PV solar parks. This research will analyze how a centralized battery storage should be sized, to store the electricity when the distribution network is in use and minimize the amount or energy curtailment. PV solar electricity is considered, which is connected to the current, in-use, medium voltage electricity network (Bin, et al., 2018). The solar park generated energy will be shown in kWh (Kilowatt-hour), for the time horizon of one year. The electricity network cable that will be used for electricity distribution is the N2XSEY 10KV 3*70/16, which has its load capacity in KV*A (Kilovolt times Ampere). The N2XSEY cable is widely used, also in renewable energy. Besides, this cable is manufactured, following European Standard VDE 0276 Part 620 three-core class 2 stranded bare copper medium voltage cable, which has quality requirements for electricity cables (Eland Cables, 2020; Caledonian Medium Voltage Cables, 2018; Wolse, et al., 2014). The centralized lithium-ion battery will be used to store energy, when it is not possible to distribute the energy directly through the electricity cable, due to the limits of these cables. The size of this centralized lithium-ion battery will be represented in kWh (Kilowatt-hour). This type of centralized battery has been chosen, based on research by Jaiswal (2017). Amounts that can be stored or obtained from storage every hour, depends on the usage of the load capacity of the current electricity distribution network.
3.2 Conceptual model
Figure 1 shows a simplified conceptual model of the simulation model. The PV solar park will generate energy, depending on the clearness of the sun. The generated energy will be distributed to the transformer, which needs to measure the usage of the electricity distribution network (red line/cable) to the village, constantly. When the electricity distribution network is not in use, the transformer will distribute the generated electricity directly through the electricity network to the village. Whenever the transformer measures that the electricity distribution network is already in use, the electricity has to be stored temporarily in the centralized lithium-ion battery storage. When the sun goes down, the
10
Figure 1 - Simplified conceptual model
3.2.1 Objective
The objective is to calculate how a centralized battery storage should be sized, to prevent the electricity distribution network and minimize the amount of energy curtailment. Using a simulation model, with input from a realistic simulation setting. This realistic setting is a 1000 kWp (Kilowatt-peak) PV solar park, connected to a medium-voltage distribution network and using an electricity cable that is commonly used in renewable energy settings that are connected to the grid. Calculating the capacity in kWh of the centralized battery storage will give one of the possible solutions to prevent the electricity network from exceeding its load capacity and it will help to stop energy curtailment.
3.2.2 Parameters
11
Table 1 - Parameters
3.2.3 Variables
Variables are unknown values in advance, but will be the output of the simulation model. Two variables will be seen as the output of this model, the centralized battery storage capacity (in kWh) and the total amount of energy curtailment (in kW). By setting storage capacity as infinite and simulating different values, the energy that has to be curtailed, by each storage capacity, will be found. In the end, there will be a centralized battery storage that has enough capacity to store renewable energy, on a yearly base, when the load capacity of the electricity cable is in-use. The centralized battery storage will be also sized, that helps to stop curtailing renewable energy. An overview of variables can be found in table 2.
Table 2 – Variables
3.2.4 Content
Figure 2 shows the simplified model of the simulation, which is an overview of the parameters that are needed to simulate and find out what the variables are.
Parameter Details
Produced renewable energy kWh, in a horizon of 1 year. The sun clearness data comes from Liander (2009) and the size of the solar park will have 1000 kWp. Cable maximum load capacity KVA, which is a constant factor in the simulations. The solar park
will be connected to the medium voltage distribution network. The cable maximum load capacity will change in the sensitivity analysis.
Demanded electricity kWh, which is an infinite factor. There will always be demand for distributed renewable energy.
Variable Details
Centralized battery storage capacity
kWh, which shows the total amount of capacity needed. This will set as infinite and different storage values will be simulated, to find out how the battery storage should be sized.
Curtailed energy/Thrown away energy
12 This model consists of input, simulation, and output. An additional note to this model is the medium voltage distribution network. The cable used for simulations is fixed, but this will change in the sensitivity analysis. During the simulations, the centralized battery storage capacity will vary to gather the output, which is the battery capacity to achieve 0% energy curtailment and no congestion in the current electricity distribution network. Data will be input in the model and has to be known in advance, as well as assumptions and simplifications. In this case, the following assumptions have been made:
• The PV solar park is connected to the medium voltage distribution network (Bogaert, et al., 2019; Bin, et al., 2018).
• The electricity network cable that is used, in this simulation, in the medium voltage distribution network is the N2XSEY 3*70/16: 10KV * 258A. This is based on the medium voltage distribution network, which should have a KV-value between 1KV and 35 KV (Bogaert & Derksen, 2017). Besides, this specific electricity cable is commonly used in renewable energy settings (Caledonian Medium Voltage Cables, 2018; Eland Cables, 2020). • The outcome of the centralized battery storage is based on 0% of curtailed energy and no
congestion in the electricity cables.
Besides the assumptions, the following simplifications have been made:
• The model will be simplified, by supposing there is always a fixed demand for distributed renewable energy. This creates the possibility to completely focus on the supply-side, which consists of the actual supply-, storage-, and distribution of renewable energy.
• Possible storage- or distribution losses are not taken into account.
3.2.5 Component list
The component list, in table 3, shows all factors for this simulation model. This list also mentions whether the factor is included or excluded, due to the assumptions and simplifications that have been made beforehand.
13
Table 3 - Component list
3.2.6 Logic flow diagram
Figure 3 shows the logic flow diagram of the simulation model, which gives a broad overview of the processes and decisions made. Starting, the PV solar park generates an amount of renewable energy (kWh) depending on the clearness of the sun at that particular hour. The diamond-shaped figure gives a first decision-making moment, whether the load capacity of the electricity network (KV*A) is already used. If not, the electricity will be distributed through this network to its
consumers. The electricity cable will maximally be used, and the remaining energy will be stored in a centralized battery storage. If the solar park generates energy and the load capacity is already in use, the electricity will directly be stored in a centralized battery storage (kWh). Next hour, there will be analyzed whether the electricity cable is in-use or available for electricity distribution. Returning to the start, if the PV solar park does not generate energy and there is energy stored, this energy can be distributed through the electricity cable. The model stops when all the electricity is transported to its consumers and the time horizon is reached. This simulation aims to find a centralized battery storage that is (1) large enough to store peak- supply in summer and is (2) empty at the end of the time horizon, which is one year.
Component Detail Included/Excluded Comment
Supply Solar production (kWh)
Included Effects the electricity distribution and battery storage
Demand Households Excluded Assumption or simplification
Electricity distribution network
Load capacity of the distribution network (KV * A)
Included Based on the medium voltage distribution network and the given electricity cable
Distribution losses Excluded Assumption or simplification Storage Capacity (kWh) Included Based on 0% curtailment and no
congestion
14
Figure 3 - Logic flow diagram
3.3 Experimental setup
This paragraph explains the experiments that are conducted by the simulation model. First, the base case will be described, which is the standard setup of the experiment. Secondly, the variables of the experiment will be changed to gather an answer to the research question.
3.3.1 Base case
The base case consists of different elements. The produced renewable energy, by the PV solar park, has been adapted to 1000 kWp for the realistic setting. This PV solar park is connected to the medium voltage distribution network, where the most common voltage is 10KV (Wolse, et al., 2014). The load capacity of electricity cables in the distribution network depends on:
• Units of the electromotive force (Kilovolt), which differs per cable and electricity network. • Units of electric current (Ampere), which depends on the actual size of the cable explained in
Newtons per millimetre squared (N/mm₂).
15 another calculation, because it has a value in KV * A. This has been converted into kWh by using the electric power formula, widely used in physics, science, and engineering (Kutz, 2013).
Power (Watt) = Force (Volt) * Current (Ampere)
This formula and the simulation model have been inserted in an Excel-file, where data can be adapted, and the simulation runs automatically by formulas. The values of the components can be found in table 4.
Table 4 - Components and values for simulations
3.3.2 Sensitivity analysis
After completing the simulation, there has been found how a centralized battery storage should be sized. Now, a sensitivity analysis will be done. Nowadays, there are projects world-wide, where the current electricity cables are replaced by larger cables with a higher load capacity (Goop, et al., 2017). The sensitivity analysis will analyze the overall influence of the usage of different electricity cables on the battery storage capacity. The electricity cables that will be used for the analysis are also a
N2XSEY, but different cable sizes. These have been chosen because the force in KV stays the same, while the electric current in Ampere increases. Besides, using the same force means the same three-phase voltage, which makes the results comparable. A schedule of the electricity cables can be found in table 5. Set of experiments Number of experiments Size of solar park (kWp) Cable capacity (KVA) Storage capacity (kWh)
Base case 1 1000 kWp 10KV * 258A 1000 kWh
Cable capacity 7 (see chapter 3.3.2)
1000 kWp 10KV * 258A 1000 kWh
16
Table 5 - Cable data for sensitivity analysis
Different cable sizes will be combined with different centralized battery storage capacities, which will result in the amount of energy that has to be curtailed directly. The following centralized battery storage capacities are used: 1000 kWh, 5000 kWh, 10.000 kWh, 20.000 kWh, and 30.000 kWh. These capacities are different from the capacities used for the simulations, where steps of 1000 kWh have been used, because the emphasis in the sensitivity analysis is on the cable size instead of the centralized battery storage capacity.
N2XSEY cable size Kilovolt Load in Ampere (in-ground) Load Capacity (KV*A)
17
4 Results
This results section will consist of different subchapters. First, the results of the base case will be presented. These results are explained on a monthly basis and the battery storage usage will be shown. Secondly, the variables will be changed in the simulation model, corresponding to the data provided in the methodology section, and the results will be presented yearly. After the experiments, the results of the sensitivity analysis will be discussed. Lastly, there will be a discussion paragraph.
The key performance indicators that will be focussed on, are the influence of load capacity of the electricity cable and the changing battery capacity on the amount of energy curtailment. Whenever the medium voltage distribution network and the centralized battery storage is in use, the
overproduction of energy has to be thrown away.
4.1 Results base case
The results of the base case are presented in the following two figures. Figure 4 shows the usage of a centralized battery storage in the base case. The X-axis shows the time in hours and the Y-axis shows the battery usage in kWh. The orange bar shows the maximum amount of battery storage capacity, which is set at 1000 kWh in the base case.
Figure 4 - Battery storage in-use, results of base case
By reviewing figure 4, there can be concluded that this battery does not have enough capacity to store all generated renewable energy. The battery is 50,08% of the time completely filled with renewable energy. When the battery storage is completely full, all additional generated energy has to be thrown away. 0 100 200 300 400 500 600 700 800 900 1000 0 1000 2000 3000 4000 5000 6000 7000 8000 Ba tt ery in k W h Time in hours
18 Figure 5 visualized the amount of energy that has to be curtailed, corresponding to the base case. The X-axis shows the time in months and the Y-axis shows the amount of energy curtailment in kW. The exact numerical results can be found in appendix B.
The base case uses in this simulation model a centralized battery storage of 1000 kWh. Figure 5 shows a curve like the Curve of Gauss, with the amount of monthly energy curtailment. Simulating the base case, results in a total of 467.101,95 kW, over a whole year, of energy that has to be curtailed. There can be concluded that, when using a 1000 kWh battery storage, there has to be thrown away 467.101,95 kW of renewable energy, in order to prevent the electricity distribution network from congestion. In the next chapter, results of the simulations will present how a centralized battery
storage should be sized to achieve electricity network prevention as well as zero electricity congestion.
4.2 Simulations
The simulation model is based on a one-year amount of solar produced energy, using the data explained in the methodology section. The first paragraph will be about yearly energy curtailment and the second about battery storage requirements.
4.2.1 Energy curtailment
Results of the 30 simulations are visualized in figure 6 and show the amount of energy that has to be curtailed every year, depending on the capacity of centralized battery storage. The capacity of the centralized battery storage has been increased by 1000 kWh, every simulation, which can be found on the X-axis. The Y-axis shows the amount of energy that has to be thrown away in kW, yearly. The exact numerical results can be found in appendix C.
10.000 20.000 30.000 40.000 50.000 60.000 70.000 80.000 90.000 En erg y in k W Time in months
Monthly energy curtailment in kW
19
Figure 6 - Yearly energy curtailment in kW, results of the simulations
Figure 6 shows a descending curve, which means that there is a negative relation between centralized storage capacity and energy curtailment. The amount of energy curtailment becomes lower, when a higher battery storage capacity is installed. The curve shows a larger decline of energy curtailment, before 5000 kWh than after. Especially after 5000 kWh, the curve seems almost linear, looking at the dashed line.
Yearly produced energy in this simulation is 1.008.162,66 kW, which is the sum of all twelve months. When installing a 1000 kWh battery storage, 53,7% of the produced energy will be distributed or stored (and distributed later), while 46,3% needs to be curtailed directly because of a full storage. Installing a 15.000 kWh battery, gives 77,3% distribution and 22,7% of energy curtailment. This shows that a larger battery will result in less energy curtailment. All numerical simulation results and curtailment percentages can be found in appendix C.
4.2.2 Battery storage requirements
Another perspective is to analyze the battery storage requirements. Figure 7 shows the battery storage requirements per hour. The X-axis represents the time in hours and the Y-axis shows the battery storage requirements in kWh.
-100.000 0 100.000 200.000 300.000 400.000 500.000 1 0 0 0 2 0 0 0 3 0 0 0 4 0 0 0 5 0 0 0 6 0 0 0 7 0 0 0 8 0 0 0 9 0 0 0 1 0 0 0 0 1 1 0 0 0 1 2 0 0 0 1 3 0 0 0 1 4 0 0 0 1 5 0 0 0 1 6 0 0 0 1 7 0 0 0 1 8 0 0 0 1 9 0 0 0 2 0 0 0 0 2 1 0 0 0 2 2 0 0 0 2 3 0 0 0 2 4 0 0 0 2 5 0 0 0 2 6 0 0 0 2 7 0 0 0 2 8 0 0 0 2 9 0 0 0 3 0 0 0 0 En erg y in k W Battery in kWh
20
Figure 7 - Battery storage requirements, results of simulations
This figure clearly shows one peak, between 5000 and 5500 hours, which is 28.199,01 kWh. This amount of battery storage is needed, when aiming to stop curtailing completely and prevent the electricity distribution network from congestion. It is important to consider whether it is economically worth it to install this battery storage capacity, to gather all renewable energy. This depends highly on the costs of the centralized battery storage and the adding value of storing the peak-supply. It might be more profitable to curtail some energy, by installing a smaller battery storage, but these decisions have to be made by the stakeholders.
There can be concluded that a PV solar park, with a yearly supply of 1.008.162,66 kW should have a centralized battery storage of 28.199,01 kWh, to stop energy curtailment completely and prevent the electricity distribution network from congestion. By dividing the battery storage with the yearly supply, the centralized battery capacity should be 2,8% of the total yearly renewable energy supply.
4.3 Sensitivity analysis
The sensitivity analysis is based on the influence of the physical cable size and the centralized battery storage capacity on the amount of energy that has to be curtailed. Assuming the amount of produced energy will stay the same, while the actual load capacity of the cable will increase, due to its physical size in Ampere. The influence of cable size and battery storage on the amount of energy that has to be curtailed is shown in table 6.
0 5000 10000 15000 20000 25000 30000 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000 5500 6000 6500 7000 7500 8000 8500 Ba tt ery sto ra g e in k W h Time in hours
21 N 2X SE Y 10kV C able s Battery Storage (kWh) 1000 kWh 5000 kWh 10000 kWh 20000 kWh 30000 kWh 3 x 25/16 (65,8%) 663.139,62 kW (59,6%) 600.373,07 kW (57,6%) 580.821,00 kW (54,8%) 552.175,97 kW (52,6%) 530.545,74 kW 3 x 35/16 (61,4%) 618.686,98 kW (54,7%) 550.973,47 kW (52,0%) 524.013,42 kW (49,0%) 493.917,28 kW (46,4%) 467.576,89 kW 3 x 50/16 (54,7%) 551.890,52 kW (45,9%) 463.067,74 kW (43,4%) 437.236,74 kW (40,1%) 404.506,27 kW (37,8%) 381.164,87 kW 3 x 70/16 (46,3%) 467.101,95 kW (33,0%) 332.532,03 kW (30,1%) 303.347,64 kW (12,0%) 120.625,11 kW (0,0%) 0,00 kW 3 x 95/16 (35,4%) 357.047,66 kW (0,0%) 375,70 kW (0,0%) 0,00 kW (0,0%) 0,00 kW (0,0%) 0,00 kW 3 x 120/16 (32,5%) 327.159,34 kW (0,0%) 0,00 kW (0,0%) 0,00 kW (0,0%) 0,00 kW (0,0%) 0,00 kW 3 x 150/16 (29,4%) 300.543,49 kW (0,0%) 0,00 kW (0,0%) 0,00 kW (0,0%) 0,00 kW (0,0%) 0,00 kW
Table 6 - Results sensitivity analysis
Table 6 shows the amount of energy that has to be curtailed, in percentages and kW. For example, when using the N2XSEY 10KV 3 x 35/16 cable and a 5000 kWh centralized battery storage, 550.973,47 kW of energy has to be thrown away directly, which is 54,7% of the total kW produced per year. There can be concluded that a 3 x 120/16 cable and 5000 kWh storage has 0,00 kW curtailment, which is equal to the result of installing a 3 x 70/16 cable and a 30000 kWh battery storage. This can be used by stakeholders, when decisions have to be made whether to expand the load capacity of the electricity distribution network or installing a larger centralized battery storage.
22
Table 7 - Battery storage capacity per N2XSEY 10KV cable
Table 7 shows exactly how much centralized battery storage is needed when using a certain electricity cable and aiming to stop energy curtailment. The difference in storage capacity is a lot higher between smaller electricity cables, than between larger electricity cables. When analyzing the difference between the 3 x 25/16 and the 3 x 50/16, more energy will be distributed directly
throughout the whole year, which requires a smaller centralized battery storage. While the difference, in battery storage, between the 3 x 95/16 and the 3 x 150/16 is smaller, because this will only cover the peak of the (approximately) normally distributed supply in summertime, instead of supply throughout the whole year.
4.4 Discussion
Research has shown a relationship between the usage of an electricity cable, centralized battery storage capacity, and the amount of energy that has to be curtailed. If a PV solar park would like to stop curtailing renewable energy completely and prevent the electricity distribution network from congestion, the centralized battery storage should have a capacity in kWh of 2,8% of the total yearly generated renewable energy. This outcome is a specific outcome of this research, but is generalizable for other similar situations. If a certain situation does not use the same input data, the simulation model can be easily adjusted.
The simulation model is useful for stakeholders in renewable energy supply. They could use the outcomes of this research in their plans, whether to expand the load capacity of the current electricity distribution network or applying a larger centralized battery storage, especially when looking at the results of the sensitivity analysis. This analysis covers a wide range of different cable sizes and different centralized battery storage capacities.
This paper adds valuable information to earlier research, because of the combination of different parameters. The increasing number of PV solar parks causes a higher peak-supply at certain times, which leads to overwriting the maximum load capacity of the electricity distribution network, a
N2XSEY 10kV Cable Load capacity (KV*A) Battery Storage Capacity (kWh)
23 lot faster. This paper can be used for decision-making, whether to enhance the electricity network or the centralized battery storage, to prevent the electricity network from congestion and stop energy curtailment.
24
5 Conclusion
This paper aims to calculate the sizing of a centralized battery storage, in order to prevent the current electricity distribution network from congestion and reduce energy curtailment. The next research question has been formulated: How should a centralized battery storage system be sized, to
prevent the current electricity distribution network from congestion?
A simulation model has been created, where the input is specific data of cable capacity, solar produced energy, and the centralized battery storage capacity. This simulation model shows that a PV solar park needs a centralized battery storage, which has the kWh of 2,8% of its total yearly produced renewable energy, in order to prevent the medium voltage electricity network and have no energy curtailment. Besides, the sensitivity analysis shows that a larger electricity cable results in a smaller centralized battery storage, but this effect is negligible from an electricity cable with a load capacity of 3090 KVA.
When building a PV solar park, this simulation model could be used in the decision-making processes, whether to upgrade the electricity distribution network or installing a centralized battery storage of a certain size. This final decision could be made, based on different considerations, but this simulation model will give an insight into several possibilities.
25
References
Becker, S. et al., 2014. Transmission grid extensions during the build-up of a fully renewable pan-European electricity supply. Energy, Issue 64, pp. 404-418.
Biggar, D. R. & Hesamzadeh, M. R., 2014. The economics of electricity markets. 1st ed. Chichester: John Wiley & Sons Ltd.
Bin, L., Xingchen, L., Xiaoqing, B. & Zhineng, L., 2018. Storage capacity allocation strategy for distribution network with distributed photovoltaic generators. Journal of Modern Power Systems and Clean Energy, 6(6), pp. 1234-1243.
Bird, L., Cochran, J. & Wang, X., 2014. Wind and Solar Energy Curtailment: Experience and Practices in the United States, Denver: National Renewable Energy Laboratory.
Bird, L. et al., 2016. Wind and solar curtailment: A review of international experience. Renewable and Sustainable Energy Reviews, Issue 65, pp. 577-586.
Bogaert, T. & Derksen, L., 2017. Betrouwbaarheid van elektriciteitsnetten in Nederland, s.l.: Netbeheer Nederland.
Bogaert, T., van der Vliet, M., Wolse, H. & Luiten, R., 2019. Betrouwbaarheid van elektriciteitsnetten in Nederland, s.l.: Netbeheer Nederland.
Caledonian Medium Voltage Cables, 2018. Medium Voltage Cables to VDE 0276, Sussex: Addison Cables.
California ISO, 2017. Managing oversupply, Folsom, California: CommPR. Clean Energy Authority, 2010. Clean Energy Authority. [Online]
Available at: https://www.cleanenergyauthority.com/solar-energy-resources/what-is-a-megawatt-and-a-megawatt-hour
[Accessed 16 May 2020].
Comello, S. & Reichelstein, S., 2019. The emergence of cost effective battery storage. Nature Communications, 1(10), pp. 1-9.
Dai, Q. et al., 2016. Capacity Planning of Battery Energy Storage System within Wind Farm. Asia-Pacific Power and Energy Engineering Conference, 1(16), pp. 1827-1831.
Destouni, G. & Frank, H., 2010. Renewable Energy. Ambio, Issue 39, pp. 18-21.
Dutta, G. & Mitra, K., 2017. A literature review on dynamic pricing of electricity. Journal of the Operational Research Society, 68(10), pp. 1131-1145.
Eland Cables, 2020. Eland Cables. [Online]
Available at: https://www.elandcables.com/cables/n2xsey-xlpe-pvc-6-10-12-kv-cable# [Accessed June 2020].
Elhadidy, M. & Shaahid, S., 1999. Optimal sizing of battery storage for hybrid (wind+diesel) power systems. Renewable energy, 1(18), pp. 77-86.
European Renewable Energy Council, 2010. Renewable energy in Europe: market, trends and technologies. 2nd ed. London: Earthscan.
26 Goop, J., Odenberger, M. & Johnsson, F., 2017. The effect of high levels of solar generation on
congestion in the European electricity transmission grid. Applied Energy, 2(205), pp. 1128-1140. Hao, S. & Shirmohammadi, D., 2002. Congestion Management With Ex Ante Pricing for
Decentralized Electricity Markets. Transactions on Power Systems, 4(17), pp. 1030-1036. Heide, D. et al., 2010. Seasonal optimal mix of wind and solar power in a future, highly renewable
Europe. Renewable Energy, 11(35), pp. 2483-2489.
Jaiswal, A., 2017. Lithium-ion battery based renewable energy solution for off-grid electricity: A techno-economic analysis. Renewable and Sustainable Energy Reviews, Issue 72, pp. 922-934. Keck, F., Lenzen, M., Vassallo, A. & Li, M., 2019. The impact of battery storage for renewable
energy power grids in Australia. Energy, 2019(173), pp. 647-657.
Kosowatz, J., 2018. Energy storage smooths the Duck Curve. Mechanical Engineering, 2018(v140 n06), pp. 30-35.
Kumar, J. S., Raja, S. C., Srinivasan, D. & Venkatesh, P., 2018. Hybrid renewable energy-based distribution system for seasonal load variations. International Journal of Energy Research, 3(42), pp. 1066-1087.
Kutz, M., 2013. Handbook of measurement in science and engineering. 2 ed. New Jersey: John Wiley & Sons, Inc.
Liu, S., Bie, Z., Lin, J. & Wang, X., 2018. Curtailment of renewable energy in Northwest China and market-based solutions. Energy Policy, Issue 123, pp. 494-502.
Pan, X., Xu, H., Song, J. & Lu, C., 2015. Capacity Optimization of Battery Energy Storage Systems for Frequency Regulation. International Conference on Automation Science and Engineering, 1(15), pp. 1139-1144.
Peng, T. & Tomsovic, K., 2003. Congestion Influence on Bidding Strategies in an Electricity Market. Transactions on Power Systems, 3(18), pp. 1054-1061.
Poletti, S. & Wright, J., 2020. Real-Time Pricing and Imperfect Competition in Electricity Markets. The Journal of Industrial Economics, 1(68), pp. 93-135.
Prabakar, K., Kale, S. A. & Durai Prabhakaran, R. T., 2015. Renewable energy and sustainable development. 1st ed. New York: Nova Science Publishers, Inc.
Qian, H. et al., 2020. Load Distribution Constraints Based Power Supply Capacity Calculation for a Distribution Network, Nanjing: IOP Conference Series.
Ray, P., 2019. Renewable energy and sustainability. Springer, 2019(21), pp. 1517-1533.
Rumpf, J., 2020. Congestion displacement in European electricity transmission systems - finally getting a grip on it? Revised safeguards in the Clean Energy Package and the European network codes. Journal of Energy & Natural Resources Law.
Schaber, K., Steinke, F., Mühlich, P. & Hamacher, T., 2012. Parametric study of variable renewable energy integration in Europe: Advantages and costs of transmission grid extensions. Energy Policy, Issue 42, pp. 498-508.
27 Tang, N., Zhang, Y., Niu, Y. & Du, X., 2018. Solar energy curtailment in China: Status quo, reasons
and solutions. Renewable and Sustainable Energy Reviews, 1(97), pp. 509-528.
Wolse, H., Derksen, L., Bogaert, T. & Luiten, R., 2014. Betrouwbaarheid van elektriciteitsnetten in Nederland, Nederland: Netbeheer Nederland.
28
Appendices
29
30
Appendix B: Base case numerical results
Month Energy curtailment (in kW)
31
Appendix C: Simulation numerical results
Storage capacity (kWh) Energy curtailment (kW) Energy curtailment in %
1000 kWh (Base case) 467.101,95 kW 46,33% 2000 kWh 395.694,43 kW 39,25 % 3000 kWh 356.391,00 kW 35,35 % 4000 kWh 338.524,77 kW 33,58 % 5000 kWh 332.532,03 kW 32,98 % 6000 kWh 327.894,19 kW 32,52 % 7000 kWh 322.721,22 kW 32,01 % 8000 kWh 317.843,12 kW 31,53 % 9000 kWh 313.307,24 kW 31,08 % 10000 kWh 303.347,64 kW 30,09 % 11000 kWh 288.475,23 kW 28,61 % 12000 kWh 271.533,80 kW 26,93 % 13000 kWh 258.596,92 kW 25,65 % 14000 kWh 244.599,77 kW 24,26 % 15000 kWh 228.349,79 kW 22,65 % 16000 kWh 212.454,08 kW 21,07 % 17000 kWh 190.241,92 kW 18,87 % 18000 kWh 163.005,49 kW 16,17 % 19000 kWh 137.117,79 kW 13,60 % 20000 kWh 120.625,11 kW 11,96 % 21000 kWh 106.654,40 kW 10,58 % 22000 kWh 88.135,58 kW 8,74 % 23000 kWh 68.238,82 kW 6,77 % 24000 kWh 59.094,44 kW 5,86 % 25000 kWh 49.575,16 kW 4,92 % 26000 kWh 30.078,21 kW 2,98 % 27000 kWh 10.458,60 kW 1,04 % 28000 kWh 441,64 kW 0,04 % 29000 kWh 0 kW 0,00 % 30000 kWh 0 kW 0,00 %
