1
Relationship between the size of a solar
park and the amount of curtailment
Written by Robin Paauw University of Groningen Faculty of Economics and Business Pre-MSc Supply Chain Management
2
Relationship between the size of a solar park and the
amount of curtailment
Abstract
With climate change and the urge for renewable resources a new ‘problem’ appears; curtailment of renewable energy methods. The ongoing mismatch between demand and supply of these resources can become a problem. Energy has to be thrown away after a certain period in the case when there is no use for it due to low local demand. This research focuses on the relationship between the size of a solar park and the amount of curtailment. The purpose of this research is to find out what this relationship is and what factors influence the amount of curtailment. It performs a simulation model that gives insight into the different decisions that are made before curtailment happens. The results show that mostly the mismatch between supply and demand result in curtailment. However, looking at the size of a solar park mostly the storage capacity can offer a solution to bringing down the amount of curtailment.
Keywords: renewable energy, renewable solar energy heating, self-sufficiency, sustainability, curtailment
3 Table of Contents 1. INTRODUCTION ... 4 2. LITERATURE REVIEW ... 6 3. RESEARCH METHOD ... 9 3.1 Problem description ... 9
3.2 Conceptual (simulation) model ... 10
3.2.1 Component list ... 10
3.2.2 Process flow diagram ... 10
3.3 Simulation model ... 11 3.4 Experimental design ... 13 3.4.1 Data explanation ... 13 3.4.2 Set of experiments ... 13 3.4.3 Base case ... 14 3.4.4 Other experiments ... 14
4. RESULTS & DISCUSSION ... 16
4.1. Results ... 16
4.2 Base case ... 16
4.3 Solar park peak capacity ... 19
4.4 Storage capacity ... 19
4.5 Hydrogen storage inventory beginning ... 21
5. CONCLUSION ... 22
4
1. Introduction
Production of solar energy has been increasing over the years and also demand has been increasing intensively (CBS, 2019). The energy that is being produced by solar parks is first used to meet the demand of the city or town that is in charge of the solar panels. If there’s more energy than demand, the energy goes into the electricity grid or gets stored. Yet the world is increasing the number of renewable energy resources it is still only a very tiny part of the total energy production. (CBS, 2019)
The problem that occurs with renewable energy is that the production is quite difficult to match with the actual demand on a local, national and international scale. At times there is more energy than demand and vice versa (Liu, Bie, Lin & Wang, 2018). In cases of more energy than demand, it means that there is leftover energy that can be used to meet the demand in other places in the country.
While this sounds like something very useful and efficient, there are some downsides to it. One of them is the fact that there is a maximum capacity on the electricity grid, which means there is a maximum to the amount of energy that can be used to satisfy the demand of other cities or towns to meet the demand. After reaching the maximum capacity it can be stored for a certain period but after that curtailment takes place; the energy is thrown away. Curtailment only happens for solar parks if the production exceeds both grid and storage capacity.
The size of a solar park affects the level of curtailment in various ways. Though, curtailment is also seen as something that sometimes is taken for granted if you look at curtailment from an economic perspective. Curtailment is often a cheaper method rather than storing it or selling it to the electricity grid.
5
There has not been much research about the relationship between the size of solar parks and the amount of curtailment. Although what has been found out is that a variety of factors, such as generation mix, market structure, operating rules and transmission grid affect the operation of renewable energy generators, and therefore affect curtailment. (Bird, Cochran, & Wang, 2016). The size of solar parks was not taken into account within previously mentioned research.
Taking into account that there is a maximum capacity on the electricity grid and storage, this means that a solar park ideally is designed to meet the demand of a city including the maximum amount of energy that can be used on the electricity grid or can be stored. Without having to use the electricity grid when the demand is lower than production a question arises. How does
the size of a solar park affect the amount of curtailment? (taking into account that there’s a
maximum capacity on the electricity grid).
This research question is relevant as curtailment is literally a waste of energy which is a waste of money as some solar panels are useless when this happens. This paper takes a look at curtailment from a sustainability perspective, rather than an economic perspective. The sustainability perspective looks at curtailment in a way that it is unnecessary and can be avoided if the solar park is just big enough to meet the demand of the city including the amount that can be used by transporting it on the electricity grid. This method is the most environmentally friendly. This paper deals with the energy curtailment of solar parks and answers to the research question on how the size of the solar park affects the amount of curtailment.
6
2. Literature Review
Previous research proves that curtailment mostly happens during periods of low local demand, which is often during warmer periods in the year (Liu, Bie, Lin, & Wang, 2018). It is also proven that curtailment mostly occurs during 09:00 and 15:00 (Dumlao & Ishihara, 2019). Demand mostly influences the amount of curtailment (Bird, Cochran, & Wang, 2016) and is hard to influence. Storing energy is mostly not an option due to financial reasons as from a sustainability perspective storing is more interesting (Parra, et al., 2018). The question does arise whether curtailment is something that should be taken for granted or that it is something that can be easily influenced. However, there is no research about the sizes of solar parks (capacity) and the relationship it has with curtailment. There has been some research about the size of wind parks and curtailment, however this is different. This research fills the information gap on the relationship between solar park sizes (which includes the storage size, inventory size & peak capacity) with the amount of curtailment.
Solar energy curtailment can be defined as ‘’the wasted energy which the photovoltaic power plants can generate electricity based on the solar resources but could not do it due to various reasons from the grid sides’’ (Tang, Zhang, Niu, & Du, 2018). Where in this definition photovoltaics is the conversion of light into electricity (Knier, 2009). A broader definition for renewable sources in general is ‘’the use of less wind or solar power than is potentially available at a given time’’ (Bird & Milligan, 2016). Previously mentioned definition by Bird & Miligan will be used within this research.
Bird et al. (2015) found that curtailment of wind and solar resources typically occurs because of transmission congestion of lack of transmission access, but it can also occur for reasons such as the demand being lower in certain periods leading to excess energy. Other research found that curtailment is mostly influenced due to the demand for energy and the amount of solar energy, solar approximation can help foresee the amount of curtailment (Dumlao & Ishihara, 2019). While both articles stress the fact that curtailment often occurs during low demand periods it does not provide information on sizes of solar parks and how that influences curtailment.
7
curtailment that has been mentioned is oversupply which means that excess renewable energy cannot be integrated into the grid, renewable power supplies must be curtailed. Dumlao & Ishihara (2019) collected data which is similar to the research of Liu et al. They based their data on hourly information about the demand and the supply of solar energy but also about the transmission (total load) and the amount of curtailment that is taking place. Their analysis showed that after a certain point there is a gap between production and consumption, which is the amount of curtailment that is taken place. They found that there was a peak in the amount of curtailment during winter and that there was less curtailment during the summer (Dumlao & Ishihara, 2019). This same research shows that between 09:00 and 15:00 the most curtailment occurs. While previously mentioned research focuses on when curtailment occurs, there is no information gathered about the internal factors of the solar park, all factors within these researches are external, hence this paper is relevant.
Research by Sevilla, Parra, Wyrsch, Patel, Kienzle & Korba (2018) provides information on the dilemma that comes with storing energy and if it can prevent curtailment from happening. However, the research concludes that curtailment is a financially more attractive option than storing it. Storing energy has mostly been researched for wind energy. In research by Waite & Modi (2016) on storing wind energy, they found that curtailment often occurs during October – March and due to less demand but on a storage perspective they found that during other periods the storage capacity exceeds the needs which means there is more storage capacity than is needed during most of the year.
8
9
3. Research method
3.1 Problem description
The problem within the research is to decide the size of the solar park. The size of the solar park (and therefore the number of solar panels) decides how much solar energy can be generated. If the solar park is too big, there will be too much solar energy generated. Too much solar energy results in curtailment, as not everything can be sold to the electricity grid or can be stored. While a too small solar park results in not being able to meet the demand of the local community or not making optimal use of the electricity grid capacity and storage methods. This leads to losses in money and sustainable ways of generating energy. This is a challenging problem because it can be really difficult to find the optimal balance between supply and demand as there are factors that have to be taken into account that can vary.
10
3.2 Conceptual (simulation) model
3.2.1 Component list
Component Detail Includes/exclude Comment
Size of a solar park
Solar park peak capacity
Number of solar panels
Include
Exclude
The maximum amount of MWp that a solar park can produce.
As there are different sizes and different kind of solar panels on the market, this research does not take solar panels into consideration. Therefore, it considers the park peak capacity as mentioned above.
Curtailment Amount of
curtailment taking place
Include Amount of solar energy that goes to curtailment measured per hours in MWh.
Local demand Local community demand of energy
Include Local electricity demand (MWh) per hour per day
Electricity grid Solar electricity grid peak capacity
Include The maximum amount of MWp that the electricity grid can distribute to other places. Storage capacity Solar storage
capacity
Include The amount of electricity in MWh that can be generated Hydrogen inventory beginning Amount of hydrogen energy that is already stored at the start.
Include The amount of hydrogen energy in MWh at the start of the period
3.2.2 Process flow diagram
11
between demand and supply. Supply is hard to influence as solar energy supply depends on the amount of solar light. If one follows the process described in figure 1 one would notice that each step is a possibility and that curtailment is the last possible decision that can be taken with the energy.
FIGURE 1 – Curtailment process of energy
3.3 Simulation model
This section explains step by step how the simulation model can be used, this model can be found in figure 2.
1. The simulation model starts with the generation of solar energy by the use of a solar park with solar panels.
2. This solar energy is being used to supply the local community.
3a. In case there is enough supply it can be used to meet the local energy demand.
3b. In case there is not enough supply it can be bought from the electricity grid (out of scope). Buying from the electricity grid is not considered within this research, because it offers additional energy rather than energy that is being curtailed and is therefore not relevant. 4a. If there is leftover energy (meaning that the local community has enough according to the model) it can then be sold to the electricity grid. Though, the electricity grid has a maximum capacity.
4b. If the grid capacity is not yet reached, then it can be sold to the electricity grid. If the capacity is reached, then the energy can be stored.
5a. If there is enough storage capacity the electricity can be stored to use at a later moment. 5b. If there is not enough storage capacity, then the model has no other option than to curtail the energy that is produced. Energy can only be stored for a short amount of time and if there is no other purpose with it, it is being curtailed.
6. The model only works for a specific amount of time. All the final decisions in this model end up going to the decision if the time horizon continues or stops. If the time horizon is reached, then the model ends. If not, then the model can be used for a new hour.
Size of solar park
Amount of energy generated
Use for local demand or sell
to electricity grid
12
FIGURE 2 – Logical process flow diagram curtailment process
The amount of curtailment is being calculated as following:
13
3.4 Experimental design
3.4.1 Data explanation
The data being used within this experiment is data that was gathered by Liander (2008) which is based on 10.000 households. The data gives a clear insight into the production of solar energy per hour and the solar energy demand in that specific hour.
The data is used to be able to find gaps between supply (production) and demand of solar energy and therefore the data can be used to find out when curtailment happens and how the size of a solar park influences this. Within the experiments, the size of the solar park (the solar park peak capacity but also the storage capacity) is being looked into by the use of different experiments that influence this.
Within the experiment, there is a fixed electricity grid capacity, which is 2 MWh. This means that every hour there is 2 MWh that can be transported to the grid in case there is more supply. This maximum grid capacity is an assumption as real numbers on maximum grid capacities were not found in the literature. Also, the hydrogen inventory is set with a minimum of 500 MWh. This parameter can vary to a higher number but can never become lower than 500 MWh as a safety so that there is always enough energy and the model doesn’t run out of energy.
3.4.2 Set of experiments
Set of experiments
Number of experiments
Solar park peak capacity (MWp) Storage capacity (MWh) Electricity grid capacity per hr (MWh) Beginning hydrogen inventory Base case 1 50 500 2 500
Solar park peak capacity
21 0-100 (with steps of 5)
500 2 500
Solar park peak capacity
7 0-6000 (with steps of 1000)
500 2 500
Storage capacity 16 50 500-2000 (with
steps of 100)
2 500
14 Hydrogen inventory beginning 6 50 1000 2 500-1000 (steps of 100) 3.4.3 Base case
The base case is the normal system, which is the system of how it functions in a normal situation without any additions. The model assumes that there is always 500 MWh needed to store the electricity, this is the minimum capacity that prevents the model from running out of energy. This storage capacity is being used completely in the base case. In this situation, there will be one experiment. There is a maximum capacity on the electricity grid per hour of 2 MWh, with a solar peak capacity of 50 MWp.
3.4.4 Other experiments
The first group of experiments is based on the solar park peak capacity. The solar park peak capacity is in the base case set on 50 MWp. Within the experiment, it might be interesting to see what the effect of change the peak capacity has on the amount of curtailment. With a start of 0 MWp peak capacity, it means that there is no extra capacity during periods. With steps of 5 MWp, a possible relationship might be noticeable. To be sure the experiment provides the right answers needed another group of experiments will be done to see if the relationship found in the first experiment is true by doing 7 individual experiments with increasing the solar park peak capacity with steps of 1000 MWp up to 6000 MWp. If this gives the same relationship, then it is safer to say that this is truly the relationship.
The third group of experiment bases on the storage capacity. Solar energy can be stored, and the storage capacity is something that might influence the amount of curtailment. The expectation is that with more storage capacity the amount of curtailment should be lower as more energy can be stored. In the normal 500 MWh can be stored. Within this experiment steps of 100 MWh are taken to see what influence it has on the total amount of curtailment. A fourth experiment will be done and is based on the storage capacity that is needed to get zero curtailment in one year. With 20922 MWh there is zero curtailment.
15
experiments, the storage size is set on 500 MWh and the inventory will be increased with steps of 100 up till 1000 MWh to see the relationship between the amount of curtailment and the inventory level.
16
4. Results & Discussion
4.1. Results
This chapter includes all results that were gathered for this research that focuses on the research question on how the size of a solar park affects the curtailment. The size of a solar park includes several aspects (as mentioned in the previous chapter). The first insight is given in the base case, this is a case that could be the normal situation when curtailment is not taken into account. This is a situation as to where curtailment is taken for granted. After that, a few experiments are done to see how it influences the amount of curtailment. A first experiment looks into the solar park peak capacity and tries to find a relationship with curtailment. After that, another experiment looks into the storage capacity (also known as hydrogen storage size) to see if there is a potential relationship.
All findings in this chapter help answer the question of how the size of a solar park influences the amount of curtailment. It gives an answer on the relationship between several previously mentioned variables and the solar park and gives an answer on how these variables can be used to influence the amount of curtailment and could help make better decisions on how to organize a solar park. The goal is to become as sustainable as possible, so to throw as less solar energy away as possible. 4.2 Base case Set of experiments Number of experiments
Solar park peak capacity (MWp) Storage capacity (MWh) Electricity grid capacity per hr (MWh) Beginning hydrogen inventory Base case 1 50 500 2 500
17
FIGURE 3 – Curtailment per month based on base case
Figure 3 is based on the numbers in table 1. This table shows the monthly curtailment based on the base case experiment. What can be seen is that there is almost no curtailment in January, and in February, November and December there is no curtailment.
TABLE 1 – Monthly amount of curtailment (MWh)
The reverse pattern is found in the daily electricity demand (figure 4). The daily electricity demand shows that there is less demand during the periods where there is more curtailment taking place. This is in line with what Liu et al. (2018) already described that in case of oversupply, curtailment takes place which is also the case within this research. This can also be seen in the same figure; solar energy production goes up while demand becomes lower. The effect that it has is that there is oversupply and therefore curtailment takes place as there is maximum storage capacity.
Month January February March April May June
Curtailment
(in MWh) 32,1124
0,0000 1743,5305 3778,9831 5188,4600
3674,6202
Month July August September October November December
Curtailment
(in MWh) 1743,5305 4168,3416
18
FIGURE 4 – Daily local electricity production & demand (MWh)
In total 201 days out of the 365 days, there was curtailment. During the other days, there was no curtailment which can be seen in figure 5. This figure also shows that curtailment approximately starts at the 57th day and ends at the around the 281st day of the year, except for 1 day in January. The pattern that was shown in figure 4 is also shown in figure 5 that from March till September there is more curtailment than during the other months of the year. Previously mentioned is due to less demand and more supply while using the maximum grid- and storage capacity.
FIGURE 5 – Curtailment per day based on the base case
19
4.3 Solar park peak capacity
The experiments below show the settings of the simulation model:
Set of
experiments
Number of experiments
Solar park peak capacity (MWp) Storage capacity (MWh) Electricity grid capacity per hr (MWh) Beginning hydrogen inventory Solar park peak capacity 21 0-100 (with steps of 5) 500 2 500 Solar park peak capacity 7 0-6000 (with steps of 1000) 500 2 500
The second experiment and third experiments to increase the solar park peak capacity did not give any surprising outcome. Increasing the solar park peak capacity led to more curtailment. The more the production capacity gets increased the more the amount of curtailment will be. The linear relationship can be found in figure 6. Producing more electricity than is needed to fulfil demand eventually leads to curtailment due to oversupply. Especially since producing more solar energy does not seem to be the solution to the problem, it seems to be a mismatch between supply and demand.
FIGURE 6 – Solar park peak capacity relationship with curtailment
4.4 Storage capacity
The experiments below show the settings of the simulation model focussing on storage capacity.
Set of
experiments
Number of experiments
20 Storage capacity 1 50 20922 2 500 Storage capacity 16 50 500-2000 (with steps of 100) 2 500
Storage capacity influences the amount of curtailment. A few experiments were set up by changing the storage capacity. The optimal situation is a situation without any curtailment. The optimal situation with zero curtailment is when the storage capacity is 20922 MWh. From that point (with the other settings that were mentioned above) there is no curtailment, more capacity would still lead to zero curtailment however a lower capacity leads to curtailment.
What is also noticeable is that the amount of curtailment decreases when the storage capacity increases. While this sounds logic there is no linear line that goes downwards. There is even an increase between a storage capacity of 1100 – 1200 MWh which could indicate that it a higher storage capacity does not always lead to less curtailment. This all depends on when the capacity is needed and for how long it is needed, as energy cannot be stored forever. This can be seen in figure 7.
FIGURE 7 – Relationship between storage capacity and curtailment
21
4.5 Hydrogen storage inventory beginning
Set of
experiments
Number of experiments
Solar park peak capacity (MWp) Storage capacity (MWh) Electricity grid capacity per hr (MWh) Beginning hydrogen inventory Hydrogen inventory beginning 6 50 1000 2 500-1000 (steps of 100)
The hydrogen storage seems to slightly affect the amount of curtailment, however the amount of curtailment that differs within the different situation is quite minimal. In figure 8 it is shown that at it’s highest there is approximately 25805 MWh curtailment and at it’s lowest there is approximately 25735 MWh curtailment. There seems to be no linear effect nor was found out why this sudden increase in curtailment happens while increasing the hydrogen storage inventory at the beginning of the period.
FIGURE 8 – Curtailment in relationship to hydrogen storage inventory
The hydrogen storage levels do not seem to give any different view; all they add is that there is more energy being stored during the warmer periods of the year (March – August). This can be seen in figure 9. As already mentioned earlier the capacity has to be a lot higher to decrease the amount of curtailment.
22
5. Conclusion
This research should give an answer to the question: ‘’How does the size of a solar park affect
the amount of curtailment?’’. This research took several aspects of a solar park into account:
solar park capacity, peak capacity, storage capacity and hydrogen inventory.
Curtailment occurs during periods of low demand and high supply; it is mostly due to seasonal and diurnal reasons. During warmer periods in the year between March and October, there is more curtailment taking place – due to a mismatch between supply and demand – and the storage and grid possibilities are extensively used. While during colder periods in the winter less curtailment is taking place as demand is increasing and supply is decreasing.
The first and most important factor that is needed to answer the question is the peak capacity. This factor alone could already answer the research question: an increase in peak capacity results in an increase of the amount of curtailment as producing more electricity than is needed to fulfil demand eventually leads again to oversupply. This means that the size of a solar park does not always have to be larger to be efficient which is something that should be taken into account while designing or building a solar park
However, more factors were taken into account in this research. The second factor is the storage capacity and has a huge influence on the amount of curtailment. In the experiments that were done, it was possible to reach zero curtailment, but that would take storage of more than forty times the storage in a normal situation. From a sustainability perspective, this would be perfect, however this is financially close to impossible. An increase in storage capacity affects the amount of curtailment positively; the amount of curtailment decreases. A third factor is the that was taken into account is the hydrogen storage inventory. This last factor does not seem to have lots of impact on the amount of curtailment as energy can only be stored for a short time and cannot exceed the maximum storage capacity.
23
A recommendation for future research would be to focus on the mismatch between supply and demand. This current research only gives insights in when curtailment occurs, however it does not provide methods to prevent curtailment from happening during low demand periods with oversupply occurring. This future research could possibly fill that gap.
24
Bibliography
CBS. (2019, June 3). CBS. Retrieved from Share of renewable energy up to 7.4 percent: https://www.cbs.nl/en-gb/news/2019/22/share-of-renewable-energy-up-to-7-4-percent NS, L. (2015). Introduction: Solar Energy Conversion. Chemical Reviews, 2.
Rathi, A. (2017, December 11). Batteries can’t solve the world’s biggest energy-storage
problem. One startup has a solution. . Retrieved from Quarts:
https://qz.com/1133123/batteries-cant-solve-the-worlds-biggest-energy-storage-problem-one-startup-has-a-solution/
Steffens, F. (1991). Solar Energy: battery energy storage control. Journal of power sources, 20.
Ogunniyi, Emmanuel, & Pienaar. (2017). Overview of Battery Energy Storage Sysetm
Advancement for Renewable (Photovoltaic) Energy Applications. Cape Town, South
Africa: Researchgate.
Dimitriev, O., Sun, H., & Tsukaka, Y. (2020, October 4). Principles of solar energy storage.
Energy Storage, p. 13.
Destouni, G., & Frank, H. (2010). Renewable Energy. Ambio, 18-21.
Tang, N., Zhang, Y., Niu, Y., & Du, X. (2018). Solar energy curtailment in China: Status quo, reasons and solutions. Renewable and sustainable Energy, 509-528.
Knier, G. (2009, August 6). How do photovoltaics work? Retrieved from NASA Science: https://science.nasa.gov/science-news/science-at-nasa/2002/solarcells
Hollick, J. (1998). Solar cogeneration panels. Renewable Energy v15 n1, 15(1), 195-200. Ashutosh, T., & Sergiy, V. (2014). Advanced energy materials. New Jersey: Hoboken; John
Wiley & Sons.
Visa, I., Duta, A., Moldovan, M., Burduhos, B., & Neagoe, M. (2020). Solar Energy
Conversion Systems in the Built . Brasov, Romania: Springer .
Heydenreich, G., Kruger, C., & Santos, R. (2015, November 18). A smart grid electricity hub.
IEEEE AFRICON International Conference.
Bird, L., & Milligan, M. (2016). Wind and solar energy curtailment: A review of international experience. Renewable & Sustainable energy reviews, 65.
Dumlao, S., & Ishihara, K. (2019, September 20). Reproducing solar curtailment with Fourier analysis using Japan dataset. Science Direct, Energy Reports, 6, 199-205.
25
Caiso. (2017). 2017. Retrieved from Impacts of renewable energy: https://www.caiso.com/Documents/CurtailmentFastFacts.pdf
Liu, S., Bie, Z., Lin, J., & Wang, X. (2018). Curtailment of renewable energy in Northwest China solutions. Energy Policy(123), 494-502.
Frew, B., Cole, W., Denholm, P., Frazier, W., Vincent, N., & Margolis, R. (2019, November 22). Sunny with a chance of Curtailment: Operating US Grid with Very High Levels of Solar Photovoltaics. iScience(21), 436-447.
Liander. (2020, April 20). Beschikbare Data Liander Electriciteitsnet. Retrieved from Liander: https://www.liander.nl/partners/datadiensten/open-data/data
Golden, R., & Paulos, B. (2015). Curtailment of Renewable Energy in California and Beyond.
The elctricity journal, 1040(6190), 36-50.
Parra, D., Wyrsch, N., Patel, M. K., Wyrsch, N., Kienzle, F., & Korba, P. (2018). Techno-economic analysis of battery storage and curtailment in a distribution grid with high PV penetration. Journal of Energy Storage , 17, 73-83.
Bird, L., Cochran, J., & Wang, X. (2016). Wind and SOlar Energy Curtailment: Experience and Practices in the United States. Renewable and Sustainable Energy Reviews, 577-586.
Waite, M., & Modi, V. (2016). Modelling wind power curtailment with increased capacity in a regional electricity grid supplying a dense urban demand. Applied Energy(183), 299-317.
