Deep Reinforcement Learning Based Approach for Optimal Power Flow of Distribution Networks Embedded with Renewable Energy and Storage Devices
This study proposes a deep reinforcement learning (DRL) based approach to analyze the optimal power flow (OPF) of distribution networks (DNs) embedded with renewable energy and storage devices. First, the OPF of the DN is formulated as a stochastic nonlinear programming problem. Then, the multi-peri...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2021-01-01
|
| Series: | Journal of Modern Power Systems and Clean Energy |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/9465777/ |
| _version_ | 1828922385934319616 |
|---|---|
| author | Di Cao Weihao Hu Xiao Xu Qiuwei Wu Qi Huang Zhe Chen Frede Blaabjerg |
| author_facet | Di Cao Weihao Hu Xiao Xu Qiuwei Wu Qi Huang Zhe Chen Frede Blaabjerg |
| author_sort | Di Cao |
| collection | DOAJ |
| description | This study proposes a deep reinforcement learning (DRL) based approach to analyze the optimal power flow (OPF) of distribution networks (DNs) embedded with renewable energy and storage devices. First, the OPF of the DN is formulated as a stochastic nonlinear programming problem. Then, the multi-period nonlinear programming decision problem is formulated as a Markov decision process (MDP), which is composed of multiple single-time-step sub-problems. Subsequently, the state-of-the-art DRL algorithm, i.e., proximal policy optimization (PPO), is used to solve the MDP sequentially considering the impact on the future. Neural networks are used to extract operation knowledge from historical data offline and provide online decisions according to the real-time state of the DN. The proposed approach fully exploits the historical data and reduces the influence of the prediction error on the optimization results. The proposed real-time control strategy can provide more flexible decisions and achieve better performance than the pre-determined ones. Comparative results demonstrate the effectiveness of the proposed approach. |
| first_indexed | 2024-12-13T22:20:22Z |
| format | Article |
| id | doaj.art-784f4b2a2faa40179fce7179c46063b8 |
| institution | Directory Open Access Journal |
| issn | 2196-5420 |
| language | English |
| last_indexed | 2024-12-13T22:20:22Z |
| publishDate | 2021-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | Journal of Modern Power Systems and Clean Energy |
| spelling | doaj.art-784f4b2a2faa40179fce7179c46063b82022-12-21T23:29:24ZengIEEEJournal of Modern Power Systems and Clean Energy2196-54202021-01-01951101111010.35833/MPCE.2020.0005579465777Deep Reinforcement Learning Based Approach for Optimal Power Flow of Distribution Networks Embedded with Renewable Energy and Storage DevicesDi Cao0Weihao Hu1Xiao Xu2Qiuwei Wu3Qi Huang4Zhe Chen5Frede Blaabjerg6School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China,Chengdu,ChinaSchool of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China,Chengdu,ChinaSchool of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China,Chengdu,ChinaTechnical University of Denmark,Department of Electrical Engineering,Kgs. Lyngby,Denmark,2800School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China,Chengdu,ChinaAalborg University,Department of Energy Technology,Aalborg,DenmarkAalborg University,Department of Energy Technology,Aalborg,DenmarkThis study proposes a deep reinforcement learning (DRL) based approach to analyze the optimal power flow (OPF) of distribution networks (DNs) embedded with renewable energy and storage devices. First, the OPF of the DN is formulated as a stochastic nonlinear programming problem. Then, the multi-period nonlinear programming decision problem is formulated as a Markov decision process (MDP), which is composed of multiple single-time-step sub-problems. Subsequently, the state-of-the-art DRL algorithm, i.e., proximal policy optimization (PPO), is used to solve the MDP sequentially considering the impact on the future. Neural networks are used to extract operation knowledge from historical data offline and provide online decisions according to the real-time state of the DN. The proposed approach fully exploits the historical data and reduces the influence of the prediction error on the optimization results. The proposed real-time control strategy can provide more flexible decisions and achieve better performance than the pre-determined ones. Comparative results demonstrate the effectiveness of the proposed approach.https://ieeexplore.ieee.org/document/9465777/Deep reinforcement learning (DRL)optimal power flow (OPF)wind turbinedistribution network |
| spellingShingle | Di Cao Weihao Hu Xiao Xu Qiuwei Wu Qi Huang Zhe Chen Frede Blaabjerg Deep Reinforcement Learning Based Approach for Optimal Power Flow of Distribution Networks Embedded with Renewable Energy and Storage Devices Journal of Modern Power Systems and Clean Energy Deep reinforcement learning (DRL) optimal power flow (OPF) wind turbine distribution network |
| title | Deep Reinforcement Learning Based Approach for Optimal Power Flow of Distribution Networks Embedded with Renewable Energy and Storage Devices |
| title_full | Deep Reinforcement Learning Based Approach for Optimal Power Flow of Distribution Networks Embedded with Renewable Energy and Storage Devices |
| title_fullStr | Deep Reinforcement Learning Based Approach for Optimal Power Flow of Distribution Networks Embedded with Renewable Energy and Storage Devices |
| title_full_unstemmed | Deep Reinforcement Learning Based Approach for Optimal Power Flow of Distribution Networks Embedded with Renewable Energy and Storage Devices |
| title_short | Deep Reinforcement Learning Based Approach for Optimal Power Flow of Distribution Networks Embedded with Renewable Energy and Storage Devices |
| title_sort | deep reinforcement learning based approach for optimal power flow of distribution networks embedded with renewable energy and storage devices |
| topic | Deep reinforcement learning (DRL) optimal power flow (OPF) wind turbine distribution network |
| url | https://ieeexplore.ieee.org/document/9465777/ |
| work_keys_str_mv | AT dicao deepreinforcementlearningbasedapproachforoptimalpowerflowofdistributionnetworksembeddedwithrenewableenergyandstoragedevices AT weihaohu deepreinforcementlearningbasedapproachforoptimalpowerflowofdistributionnetworksembeddedwithrenewableenergyandstoragedevices AT xiaoxu deepreinforcementlearningbasedapproachforoptimalpowerflowofdistributionnetworksembeddedwithrenewableenergyandstoragedevices AT qiuweiwu deepreinforcementlearningbasedapproachforoptimalpowerflowofdistributionnetworksembeddedwithrenewableenergyandstoragedevices AT qihuang deepreinforcementlearningbasedapproachforoptimalpowerflowofdistributionnetworksembeddedwithrenewableenergyandstoragedevices AT zhechen deepreinforcementlearningbasedapproachforoptimalpowerflowofdistributionnetworksembeddedwithrenewableenergyandstoragedevices AT fredeblaabjerg deepreinforcementlearningbasedapproachforoptimalpowerflowofdistributionnetworksembeddedwithrenewableenergyandstoragedevices |