Pre-MSc paper
Renewable energy logistics
An analysis of the relationship between curtailment and the size of a wind park in supplying the heat demand of a group of self-sufficient households
June 20, 2020
University of Groningen
Faculty of Economics and Business Pre-MSc Supply Chain Management
Author: Jasper Bosma Student number: S4185641
Email: j.m.bosma.3@student.rug.nl
Abstract
Heating households without renewable energy is carbon dioxide intensive, the urge to reduce carbon dioxide emissions is high. One way to reduce carbon dioxide emissions is to heat households with renewable energy, such as heating households with hydrogen generated from wind parks. But curtailment is one of the reasons not to engage in using renewable energy sources. Therefore, this research focusses on investigating the relationship between curtailment and the size of a wind park in supplying the heat demand of a group of self-sufficient households. A simulation study has been performed to get insights into this relationship. Results show that regardless of the number of households, there is a positive relationship between the size of a wind park and the amount of curtailment. To make hydrogen heating households from renewable energies viable, the otherwise curtailed energy needs to be used in, for example, supplying the electricity demand of these households.
Key words: renewable energy, hydrogen heating, self-sufficient city, curtailment
1. Introduction
In the past centuries, fossil energy has been the mainstream energy that has been used and produced. This is because the production of fossil energy is demand-driven, and the production can be regulated very easily. But the use of fossil energy can cause environmental pollution. Therefore, the search for clean renewable energies, such as wind and solar energy, is very important to supply energy in the long-term (Ni et al., 2006).
In contrast to fossil energy, the use of renewable energy sources imposes daily and seasonal fluctuations in power supply, caused by weather conditions or by the diurnal cycle (Böttcher, Görke, Kolditz, & Nagel, 2017).Because of the unstable production of this renewable energy, energy may be thrown away due to overproduction. This is called curtailment, which in the ideal world is as small as possible (Burke & O’Malley, 2011). In the ideal world, supply and demand are equal, so that no curtailment takes place
Over the past years, wind energy has been considered as a promising solution for clean and sustainable energy development. Hydrogen heating, from wind energy, can potentially be a feasible zero-carbon alternative to natural gas (Dodds et al., 2015), but wind curtailment remains a challenge to wind power development (Jicheng, Qiushuang, Junjie, Weidong & Jing, 2019). Because the production of wind energy is weather-dependent, the production is highly variable and differs from the energy demand. A way to prevent curtailment is by using the correct wind data to better predict the demand and production of wind energy (Burke & O’Malley, 2011).
Another way to prevent energy curtailment is to store the generated energy in a hydrogen storage system (Zhang & Wan, 2014) (Amrouche, S. O., Rekioua, D., & Rekioua, T., 2015) ( Carton & Olabi, 2010). A hydrogen storage system can reduce fluctuations and therefore reduce curtailment. To enable wind power to be stored in a hydrogen storage, the generated energy first needs to go through an electrolyzer to convert the energy into hydrogen (P2G). When the hydrogen needs to be converted back into energy, the hydrogen needs to go through a fuel-cell, which converts the hydrogen into energy (G2P) (Amrouche, S. O., Rekioua, D., & Rekioua, T., 2015). For this research, a fuel-cell is disregarded, because this paper focusses on the use of hydrogen.
Energy-neutral or carbon-neutral housing (or self-sufficiency) becomes more and more relevant in today’s society of building strategies (Chang, N.-B., Rivera, B. J., & Wanielista, M. P. 2011) Energy curtailment in a self-sufficient city can occur due to various reasons, for example, due to the hydrogen storage that is full and the heat demand is already met or due to fulfilled heat demand and there is no storage where the energy can be stored. In the case of a self-sufficient city with a wind park as their hydrogen energy source, a wind park that is too large generates large amounts of energy in peak times. This causes energy curtailment because the over-produced energy can’t be used or stored when there is no storage or if the storage is full. But when the wind park is too small for the self-sufficient city and the heat demand can’t be met, the inhabitants don’t have any heat. And because a self-sufficient city isn’t dependent on the national or regional electricity grid (Van den Dobbelsteen et al., 2014), the demand has to be fulfilled by the wind park. The size of a wind park is due to this an important aspect to look at when building and/or thinking of a self-sufficient city. Therefore, it is relevant to look at what the relationship is between curtailment and the size of a wind park (in the means of the wind park peak capacity (MW)) in supplying the heat demand of a group of self-sufficient households. In this study, the relationship between these elements is analyzed by addressing the following question:
“What is the relationship between curtailment and the size of a wind park in supplying the heat demand of a group of self-sufficient households?”
better understanding of this relationship so that the future heat demands of wind energy can be more fulfilled, and the curtailment of this energy can be reduced. This analysis is executed by the method of simulation. Simulation is: “experimentation with a simplified imitation (on a computer) of an operations system as it progresses through time, for the purpose of better understanding and/or improving that system” (Robinson, 2004).
This paper is further organized as follows. First, a literature review is presented on wind energy and wind energy curtailment. Then in the Methodology section, a simulation study is being done on the production of wind energy in a wind park. The outcomes of this simulation study and the possible relationship between the elements will be presented in the results and discussion section. The paper will conclude with remarks in the conclusion.
2. Literature Review
Previous research about autonomous wind/hydrogen systems (Ø. Ulleberg, T. Nakken, A Ete, 2010) (T. Nakken, L.R. Strand, E. Frantzen, R. Rohden, P.O. Eide, 2006) focused on supplying energy to self-sufficient households. What they concluded was that the surplus (curtailment) from this generated energy should be used to heat the households. Heating households, without renewable energy, is very carbon dioxide intensive, but it’s hardly getting attention in the literature (Dodds, P. E., Staffell, I., Hawkes, A. D., Li, F., Grünewald, P., McDowall, W., & Ekins, P., 2015). The urge to heat households with renewable energies, such as hydrogen from wind energy, is therefore very high. Where previous research focused on using the wind/hydrogen system for electricity use only, this research focuses on using the hydrogen only for heating self-sufficient households. Because the relationship between curtailment and the size of a wind park in supplying the heat demand of self-sufficient households has not yet been investigated, this research contributes to the existing literature.
Autonomous wind/hydrogen systems are able to supply self-sufficient households with wind power using hydrogen as the energy storage, this is what Øystein Ulleberg et al. (2010) & T. Nakken et al. (2006) found. They contributed research about an autonomous wind/hydrogen system in Utsira in Norway, supplying 10 households. They also found that the remaining left-over energy can and must be used to meet the local heat demand, so they researched only the supply of energy to the households. But from this article we can conclude that it’s relevant to look at the use of hydrogen, what comes from the generated wind power, to heat the households.
Hydrogen heating has a large share in the energy consumption and carbon dioxide emissions globally, but it’s hardly getting attention in the literature compared to electricity and transport as energy consumers and air pollutants. (Dodds, P. E., Staffell, I., Hawkes, A. D., Li, F., Grünewald, P., McDowall, W., & Ekins, P., 2015). And because of this large share of energy consumption and carbon dioxide emissions, the urge to find other and better solutions to heat households is very important. Dodds et al. (2015) also state that hydrogen is a very potential climate-neutral alternative to the use of natural gas in heating the households.Therefore, this study contributes to the literature in the way that it investigates the relationship between curtailment and the size of a wind park in supplying the heat demand in a self-sufficient city.
3. Methodology
3.1 Problem description
The main problem addressed in this paper is that because wind energy production is so fluctuating but has the same wind park peak capacity throughout the year and the heat demand in the colder months is higher and in the warmer months is lower (see figure 3), there can be a lot of curtailed energy. Which makes hydrogen heating not viable for self-sufficient households. To make hydrogen household heating viable for a group of self-sufficient households, the wind energy curtailment needs to be reduced. Energy curtailment in a self-sufficient city is probably very likely to occur, due to the fact that the self-self-sufficient households are fully depended on the wind park, the wind park needs to be large enough to satisfy the needs of the households. But in peak times the wind park generates too much electricity to store, which is then curtailed. Therefore, to support the literature on this topic, this paper is concerned with what the relationship is between curtailment and the size of a wind park in supplying the heat demand of a group of self-sufficient households.
The parameters, which are known in advance, are the heat demand, the size of the wind park, hydrogen storage size, and the beginning hydrogen inventory. The parameters are further explained in 3.2.2.
The variables, which are not yet known in advance and which are the outcomes of the model, are the amount of energy produced, wind to curtailment, hydrogen storage at beginning of period, wind to hydrogen storage, wind to demand, storage to demand, curtailment growth, percentage demand fulfilled and the percentage of curtailment in relation to the demand. The variables are further explained in 3.2.3.
The other variables, hydrogen storage at beginning of period, wind to hydrogen storage, wind to demand, storage to demand are mostly there to calculate the amount of curtailment and are therefore important to have in the simulation model.
3.2 Conceptual simulation model
The conceptual simulation model consists of the following aspects, which describes the simulation model that is developed; the objectives, input, output, content, assumptions, and simplifications. These aspects will be first described and finally presented as the conceptual simulation model.
3.2.1 Objectives
The purpose of the simulation model is to research what the relationship is between the curtailment and the size of a wind park in supplying the heat demand (in hydrogen (MWh)) of a group of self-sufficient households. The specific objective is to determine what the relationship is between the curtailment and the size of a wind park supplying the hydrogen for heating a group of self-sufficient households and ultimately learn from this research to help reduce the amount of energy that is being curtailed. Because if there is less energy curtailed, the more viable renewable energies become in heating households with hydrogen.
3.2.2 Input
The input of the simulation model, which can be adjusted throughout the research, is mainly the ‘wind park peak capacity’, the wind park peak capacity (WPPC) is measured in MW. The simulation model also consists of three fixed parameters, which are the:
1. ‘Hydrogen storage size’. The hydrogen storage size is needed to store the left-over hydrogen when the demand is met. Without storage, all left-over hydrogen has to be curtailed. And in times when there is no/not enough wind power to fulfill the demand, the heat demand has to be fulfilled by using the hydrogen storage.
2. ‘Beginning hydrogen inventory’. Beginning hydrogen storage is leftover storage from the year before.
3.2.3 Output
The simulation model has multiple variable outputs: Note: all the variables are measured in MWh.
1. ‘Wind energy produced’. The wind energy produced (WEP) indicates the amount of energy that is produced in that hour.
2. ‘Wind to hydrogen storage’. The wind to hydrogen storage (WTHS) indicates the amount of produced wind energy goes to the hydrogen storage when the demand is met. This data can be used for calculating the number of times that the produced energy can fulfill the heat demand and thus have a surplus.
3. ‘Wind to demand’. The wind to demand (WTD) indicates the amount of produced wind energy goes to the LHD. This data can be used for calculating the number of times that the produced energy can fulfill the heat demand.
4. ‘Storage to demand’. The storage to demand (STD) indicates the amount of energy that is in the HSBP goes to the LHD when there is not enough wind energy produced that hour. This data can be used for calculating the number of times the produced energy is not enough for fulfilling the heat demand and therefore the hydrogen has to come from the storage to fulfill the heat demand.
5. ‘Wind to curtailment’. The wind to curtailment (WTC) indicates the amount of wind energy that is being curtailed due to overproduction. This data can be used to calculate the number of times there is wind energy curtailed and how much.
6. ‘Percentage of demand fulfilled’, the percentage of the demand that is fulfilled indicates the extent to which the demand is fulfilled.
7. ‘Percentage of curtailment in relation to demand’, the percentage of the curtailment in relation to the demand indicates how much percentage of the demand is being curtailed.
8. ‘Curtailment growth’, the curtailment growth indicates with what factor the curtailment has grown relative to the previous (other) experiments.
3.2.4 Content
Figure 1: overview of the simulation study
The content, scope, and level of detail are shown in the logical flow diagram of the simulation in figure 2. The simulations as presented in figure 2 works as follows:
- When at the start a specific WPPC is filled in, an amount of wind energy is produced. With this wind energy, the model looks at whether there is enough energy produced to meet the heat demand.
o If there is not enough energy produced, the model looks if there is enough hydrogen in the hydrogen storage to meet the heat demand.
§ If that is the case, the amount of hydrogen needed is grabbed from the storage.
§ If there is not enough in the storage, the heat demand is not met.
o If there is enough energy produced to meet the heat demand, that energy goes to the heat demand.
- If there is more energy produced than needed, the model looks if there is enough space in the hydrogen storage.
o If there is enough space in the hydrogen storage, the left-over energy goes to the hydrogen storage.
o If there is not enough space in the hydrogen storage to store the energy, the energy is curtailed.
Figure 2: logic flow diagram of the simulation study
3.2.5 Assumptions Conversion efficiencies:
- Electricity to hydrogen ≈ 30% loss (Ananthachar, V., & Duffy, J. J., 2005) Wind energy produced
Enough produced to meet demand? Wind to demand Yes Storage to demand No Next hour Yes No End of time horizon? Start End Energy left? No Yes
General:
- Equipment doesn’t fail (no down- or repair time)
- The heat demand data is sufficiently accurate for predicting future demand
- Every household has a hydrogen boiler, for which they can use hydrogen as their heating source
3.2.6 Simplifications
The infrastructure is ‘closed’, which means that there is no connection with other countries (or cities) modelled. This is done to keep the study achievable within the timeframe.
3.3 Experimental setup
The base settings for the simulation are shown intable 1.
Table 1: base case settings Hydrogen storage size Beginning hydrogen inventory Wind park peak capacity Electricity to hydrogen conversion efficiency Number of households Base settings 500 MWh 500 MWh 25 MW 0,7 1000
(Ananthachar, V., & Duffy, J. J., 2005). In figure 4 the local heat demand of the base case, with a thousand households, is shown.
Figure 3: Percentage demand fulfilled Figure 4: Local heat demand (MWh)base case Table 2: experimental factors Number of experiments Wind park peak capacity Hydrogen storage size Number of households Base case 1 25 MW 500 MWh 1000
Wind park peak capacity case 31 5 to 35 MW (with steps of 1) 500 MWh 1000 Sensitivity analysis households 20 25 MW 500 MWh 100 to 2000 (with steps of 100)
With the sensitivity analysis, uncertainty is reduced. The sensitivity analysis is performed by changing the number of households with steps of 100 from 100 to 2000, so 20 experiments are performed. These number of 100 to 2000 are chosen because it gives a good representation of the reality in multiple sizes of self-sufficient cities with respect to the WPPC of 25 MW of the base case.
4. Results and discussion
The results and discussion are structured as follows. In section 4.1 the results of the base case are presented. In section 4.2 the results of the experiments on the WPPC are presented and the research question is answered. In section 4.3 a sensitivity analysis is performed by changing the number of households in the base case, these results are then reflected on the research question.
4.1 base case
In table 3 the values of the variables of the base case are presented, the total heat demand, the total amount of wind to curtailment, total storage to demand, total wind to demand, total wind to hydrogen storage, the curtailment growth, the percentage of demand fulfilled and the percentage of the curtailment in relation to the demand.
Table 3: base case experiment
Wind park peak capacity 25 MW
Total heat demand 14578 MWh
Total wind to curtailment 11174 MWh Total storage to demand 6371 MWh
Total wind to demand 8128 MWh
Total wind to hydrogen storage 9003 MWh Curtailment growth (relative to the
previous experiment)
1,08
Percentage of demand fulfilled 100% Percentage of curtailment in relation to
demand
77%
capacity, the wind park produces energy at the same capacity (see figure 5). This energy is then curtailed (see figure 6) because the storage is too small to store all this hydrogen.
Figure 5: Wind energy produced (MWh) base case
Figure 6: Wind to curtailment (MWh) base case
So, in the base case, it can be seen that the wind park produces too much energy, 177% of the total demand. This overproduction of 77% (11174MWh) is then curtailed. This curtailment mostly occurs during the summer period, since the heat demand is very low, but the production remains the same.
4.2 Wind park peak capacity case
In all of the 31 experiments, the cumulative curtailment, cumulative storage to demand, cumulative wind to demand, cumulative wind to hydrogen storage, curtailment growth, percentage of the demand fulfilled and the percentage of the curtailment in relation to the demand are calculated and the results are shown in table 4 in appendix 1. The experiments are done by increasing the WPPC with 1MW every experiment from 5 to 35MW.
Figure 7: Curtailment growth WPPC case
Figure 8: Cumulative curtailment WPPC case
Figure 9: Percentage demand fulfilment WPPC case
So, the relationship between curtailment and the size of a wind park in supplying the heat demand of a group of self-sufficient households is that the higher the WPPC is, the higher the amount of curtailed hydrogen/energy is.However, this doesn’t grow exponentially. To reduce the amount of curtailed energy, and thus make a self-sufficient city more viable, the curtailed energy can be used to supply the households with electricity. however, if the otherwise curtailed energy is first stored in the hydrogen storage, an electrolyzer and a fuel-cell must be used with the conversion of hydrogen to electricity and vice versa. Conversion losses of approximately 30% are also involved in this process. Otherwise, the generated electricity can be directly used to provide electricity to the households.
4.3 Sensitivity analysis households
The sensitivity analysis is performed by changing the number of households in the base case
from 100 to 2000 with 20 steps of 100. The results of the sensitivity analysis can be found in table 5 in appendix 2, where the results of the total heat demand, cumulative wind to curtailment, cumulative storage to demand, cumulative wind to demand, cumulative wind to hydrogen storage, curtailment growth, percentage of the demand fulfillment and the percentage of curtailment in relation to the demand are calculated for the 20 experiments.
Figure 11: Cumulative curtailment sensitivity analysis
Figure 12: Curtailment growth sensitivity analysis
Figure 13: Percentage demand fulfilment sensitivity analysis
Conclusion
In this research, the research question “What is the relationship between curtailment and the size of a wind park in supplying the heat demand of a group of self-sufficient households?” is answered by using a simulation model. The simulation model represented real-life scenarios where the heat demand of a group of self-sufficient households is supplied by a wind park, converting the generated wind energy into hydrogen.
The answer to the research question is, that there is a positive relationship between the size of a wind park and the amount of curtailment. An increase in the size of a wind park results in an increase of curtailment, this is regardless of the number of households. This is because the heat demand of the households is very unstable and varies a lot throughout the year. To remain the self-sufficiency of the city, the wind park needs to produce a certain high amount of energy to supply the peaks (mostly around the winter) in the heat demand. This then creates a lot of curtailment when the demand is low (mostly around the summer) because the wind park has the same wind park peak capacity throughout the whole year.
Therefore, it is not viable to have a wind park solely to supply the heat demand of a group of self-sufficient households. It is possible to look at the possibilities of using this otherwise curtailed energy as a source of, for example, the electricity. This is somewhat the same as what Øystein Ulleberg et al. (2010) & T. Nakken et al. (2006) found. They concluded that the left-over energy must be used as a source to heat households. This means that stand-alone hydrogen and electricity systems generated by wind parks for supplying only the heat or electricity demand of self-sufficient households are not viable. For further research, it is advised to research the combination of the heat and electricity demand for self-sufficient households supplied by wind energy.
However, there are some limitations to this research. Firstly, for the sensitivity analysis, there is only 1 aspect that is being looked at (the households). For future research, to do a sensitivity analysis in more detail, a full factorial design could be executed. Secondly, the heat demand data which is used in this research comes from 2008. To be more precise, more recent data could have been used to make the simulation model more reliable. Thirdly, the simulation model only simulated the data for one year. To make the model more reliable, and thus this research, multiple years could’ve been simulated. Lastly, this research didn’t include the economic aspect/impact on using hydrogen as a heating source and if it’s economically feasible to use the left-over (otherwise curtailed) energy as a source of for example the electricity.
References
A. van den Dobbelsteen, N. Tillie, S. Broersma, M. Fremouw. 2014. The Energy Master Plan: Transition to self-sufficient city regions by means of an approach to local energy potentials. 30th International Plea Conference, 16-18 December 2014, CEPT University, Ahmedabad
Ananthachar, V., & Duffy, J. J. 2005. Efficiencies of hydrogen storage systems onboard fuel cell vehicles. Solar Energy, 78(5): 687–694.
Böttcher, N., Görke, U. J., Kolditz, O., & Nagel, T. 2017. Thermo-mechanical investigation of salt caverns for short-term hydrogen storage. Environmental Earth Sciences, 76(3): 98.
Burke, D., & O'Malley, M. 2011. Factors influencing wind energy curtailment. Transactions
on Sustainable Energy, 2(2): 185-193.
Carton, J., & Olabi, A. 2010. Wind/hydrogen hybrid systems: Opportunity for ireland's wind resource to provide consistent sustainable energy supply. Energy, 35(12): 4536-4544.
Chang, N.-B., Rivera, B. J., & Wanielista, M. P. 2011. Optimal design for water conservation and energy savings using green roofs in a green building under mixed uncertainties. Journal
of Cleaner Production, 19(11): 1180–1188.
Dodds, P. E., Staffell, I., Hawkes, A. D., Li, F., Grünewald, P., McDowall, W., & Ekins, P. 2015. Hydrogen and fuel cell technologies for heating: A review. International journal of
hydrogen energy, 40(5): 2065-2083.
Essent kennisbank – Gemiddelde gasverbruik. 2020
https://www.essent.nl/kennisbank/energie-besparen/inzicht-in-verbruik/gemiddelde-gasverbruik. May 12.
Jicheng, L., Qiushuang, W., Junjie, H., Weidong, Z., & Jing, Y. 2019. Collaboration strategy and optimization model of wind farm-hybrid energy storage system for mitigating wind curtailment. Energy Science & Engineering, 7(6): 3255-3273.
Lucy Y. Pao and Kathryn E. Johnson – 2009. ‘A Tutorial on the Dynamics and Control of
Wind Turbines and Wind Farms’ American Control Conference, Hyatt Regency Riverfront,
St. Louis, MO, USA
Nakken, T., Strand, L. R., Frantzen, E., Rohden, R., & Eide, P. O. 2006. The Utsira wind-hydrogen system–operational experience. European wind energy conference (pp. 1-9).
Ni, M., Leung, M., Sumathy, K., & Leung, D. 2006. Potential of renewable hydrogen production for energy supply in hong kong. International Journal of Hydrogen Energy, 31(10): 1401-1412.
Robinson, S. 2004. Simulation: the practice of model development and use (Vol. 50). Chichester: Wiley.
Suganthi, L., & Samuel, A. 2012. Energy models for demand forecasting—a review.
Renewable and Sustainable Energy Reviews, 16(2): 1223-1240.
Ulleberg, Ø., Nakken, T., & Ete, A. 2010. The wind/hydrogen demonstration system at Utsira in Norway: Evaluation of system performance using operational data and updated hydrogen energy system modeling tools. International Journal of Hydrogen Energy, 35(5): 1841-1852.
Vargas, L., Bustos-Turu, G., & Larrain, F. 2015. Wind power curtailment and energy storage in transmission congestion management considering power plants ramp rates. Ieee
Transactions on Power Systems, 30(5): 2498-2506.
