Battery energy storage design and operation in an HVDC-system with WPP clusters
(2019) In CODEN:LUTEDX/TEIE EIEM01 20191Industrial Electrical Engineering and Automation
- Abstract
- The electricity sector is in the beginning of a large shift from centralized dispatchable production to distributed variable production. With this necessary change, new challenges arise. One crucial challenge is related to upholding the system balance in the power grid. In this master thesis, a Battery Energy Storage System (BESS) has been implemented in an HVDC-system with large amounts of Wind Power Plant (WPP) clusters, for performing the primary control, to see if this could contribute to the development of renewables. The purpose of this work is to find out what control strategy could be used to balance the DC-grid both short-term and long-term as well as in what capacity range the BESS would have to be.
Power flow simulations... (More) - The electricity sector is in the beginning of a large shift from centralized dispatchable production to distributed variable production. With this necessary change, new challenges arise. One crucial challenge is related to upholding the system balance in the power grid. In this master thesis, a Battery Energy Storage System (BESS) has been implemented in an HVDC-system with large amounts of Wind Power Plant (WPP) clusters, for performing the primary control, to see if this could contribute to the development of renewables. The purpose of this work is to find out what control strategy could be used to balance the DC-grid both short-term and long-term as well as in what capacity range the BESS would have to be.
Power flow simulations have been performed based on the prevailing conditions of a project (Baltic InteGrid) striving to commission roughly 11 GW of offshore wind power joined between Sweden, Poland and Lithuania. Wind data from the area has been obtained in order to create estimated power outputs from the wind power plants as well as appropriate electricity market interactions for the connection points onshore. Due to the electricity market being based on estimated power output from the WPPs, simulations when no forecast error existed as well as when a large forecast error existed were performed. Additionally, due to the electricity intraday market is planned to change from operating on a one-hour-basis to a 15-min-basis, the impact of this has been evaluated. Lastly, due to the DC-system being connected to the AC utility grid, the primary control of the AC-grid is available for use. Both simulations with no AC-grid dependency as well as simulations allowing some AC-grid dependency were performed.
To an existing simulation model of Baltic InteGrid created in OpenModelica, a simplified BESS was implemented. The BESS is performing both short-term energy management (periods of seconds) and long-term energy management (periods of hours). The short term management is utilizing a DC-droop voltage control. To decrease the BESS capacity, the electricity intraday market is utilized in order to sell or buy energy in order to regulate the State of Charge (SOC) of the BESS. For the intraday market, three control strategies were implemented to decide if and to what extent a market interaction was needed. Method 1 utilizes only the current SOC of the BESS. Method 2 utilizes the predicted SOC of the BESS at one time-unit forward and Method 3 utilizes the predicted SOC of the BESS for both one time-unit forward and two time-units forward.
Through extensive simulations, it was found that all methods solved the problem of regulating the DC-grid with little or no AC-grid dependency. It was also found that the intraday market time-basis was of very large importance. Furthermore, both the forecast error and allowing the AC-grid to perform the primary control to some extent led to a much smaller BESS capacity generally. Method 1 was the least complex to implement but was largely outperformed by the more complex Method 2 and Method 3 in all aspects considered. Method 3 was better than Method 2 in terms of lower magnitudes of market interaction as well as a smaller amount of energy being transferred for altering the SOC of the BESS. However, Method 2 gave a slightly better result in terms of AC-grid dependency. The BESS capacity needed was somewhere in the range of 130-4000 MWh for Method 1 depending on the intraday market time-basis, forecast error as well as AC-grid dependency. The figures for Method 2 and Method 3 were 60-1900 MWh. Ruling in the number of SOC-cycles and thereby evaluating the lifetime of the BESS, initial results suggest a larger BESS capacity would be needed. For Method 2 and 3, allowing AC-grid dependency and a 15 min Elbas time-basis a capacity in the range of 120 - 180 MWh depending on wind power forecast error would be the bare minimum for reaching an acceptable theoretical lifetime.
Through very brief economic calculations it was concluded that implementing a BESS could with high possibility be a superior choice compared to the alternative of upgrading the AC utility grid to handle the large fluctuations. However, regarding if Method 2 or Method 3 should be utilized in order to regulate the BESS, more research with higher resolution is needed. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8974958
- author
- Åkesson, Harald LU
- supervisor
- organization
- course
- EIEM01 20191
- year
- 2019
- type
- H3 - Professional qualifications (4 Years - )
- subject
- keywords
- BESS, Energy storage, State of charge, Windpower, HVDC
- publication/series
- CODEN:LUTEDX/TEIE
- report number
- 5421
- language
- English
- id
- 8974958
- date added to LUP
- 2022-01-27 11:50:28
- date last changed
- 2022-01-27 11:50:28
@misc{8974958,
abstract = {{The electricity sector is in the beginning of a large shift from centralized dispatchable production to distributed variable production. With this necessary change, new challenges arise. One crucial challenge is related to upholding the system balance in the power grid. In this master thesis, a Battery Energy Storage System (BESS) has been implemented in an HVDC-system with large amounts of Wind Power Plant (WPP) clusters, for performing the primary control, to see if this could contribute to the development of renewables. The purpose of this work is to find out what control strategy could be used to balance the DC-grid both short-term and long-term as well as in what capacity range the BESS would have to be.
Power flow simulations have been performed based on the prevailing conditions of a project (Baltic InteGrid) striving to commission roughly 11 GW of offshore wind power joined between Sweden, Poland and Lithuania. Wind data from the area has been obtained in order to create estimated power outputs from the wind power plants as well as appropriate electricity market interactions for the connection points onshore. Due to the electricity market being based on estimated power output from the WPPs, simulations when no forecast error existed as well as when a large forecast error existed were performed. Additionally, due to the electricity intraday market is planned to change from operating on a one-hour-basis to a 15-min-basis, the impact of this has been evaluated. Lastly, due to the DC-system being connected to the AC utility grid, the primary control of the AC-grid is available for use. Both simulations with no AC-grid dependency as well as simulations allowing some AC-grid dependency were performed.
To an existing simulation model of Baltic InteGrid created in OpenModelica, a simplified BESS was implemented. The BESS is performing both short-term energy management (periods of seconds) and long-term energy management (periods of hours). The short term management is utilizing a DC-droop voltage control. To decrease the BESS capacity, the electricity intraday market is utilized in order to sell or buy energy in order to regulate the State of Charge (SOC) of the BESS. For the intraday market, three control strategies were implemented to decide if and to what extent a market interaction was needed. Method 1 utilizes only the current SOC of the BESS. Method 2 utilizes the predicted SOC of the BESS at one time-unit forward and Method 3 utilizes the predicted SOC of the BESS for both one time-unit forward and two time-units forward.
Through extensive simulations, it was found that all methods solved the problem of regulating the DC-grid with little or no AC-grid dependency. It was also found that the intraday market time-basis was of very large importance. Furthermore, both the forecast error and allowing the AC-grid to perform the primary control to some extent led to a much smaller BESS capacity generally. Method 1 was the least complex to implement but was largely outperformed by the more complex Method 2 and Method 3 in all aspects considered. Method 3 was better than Method 2 in terms of lower magnitudes of market interaction as well as a smaller amount of energy being transferred for altering the SOC of the BESS. However, Method 2 gave a slightly better result in terms of AC-grid dependency. The BESS capacity needed was somewhere in the range of 130-4000 MWh for Method 1 depending on the intraday market time-basis, forecast error as well as AC-grid dependency. The figures for Method 2 and Method 3 were 60-1900 MWh. Ruling in the number of SOC-cycles and thereby evaluating the lifetime of the BESS, initial results suggest a larger BESS capacity would be needed. For Method 2 and 3, allowing AC-grid dependency and a 15 min Elbas time-basis a capacity in the range of 120 - 180 MWh depending on wind power forecast error would be the bare minimum for reaching an acceptable theoretical lifetime.
Through very brief economic calculations it was concluded that implementing a BESS could with high possibility be a superior choice compared to the alternative of upgrading the AC utility grid to handle the large fluctuations. However, regarding if Method 2 or Method 3 should be utilized in order to regulate the BESS, more research with higher resolution is needed.}},
author = {{Åkesson, Harald}},
language = {{eng}},
note = {{Student Paper}},
series = {{CODEN:LUTEDX/TEIE}},
title = {{Battery energy storage design and operation in an HVDC-system with WPP clusters}},
year = {{2019}},
}