Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search
This letter proposes a data-driven, model-free method for load frequency control (LFC) against renewable energy uncertainties based on deep reinforcement learning (DRL) in continuous action domain. The proposed method can nonlinearly derive control strategies to minimize frequency deviation with fas...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Journal Article |
| Language: | English |
| Published: |
2020
|
| Subjects: | |
| Online Access: | https://hdl.handle.net/10356/141500 |
| _version_ | 1826123294442520576 |
|---|---|
| author | Yan, Ziming Xu, Yan |
| author2 | School of Electrical and Electronic Engineering |
| author_facet | School of Electrical and Electronic Engineering Yan, Ziming Xu, Yan |
| author_sort | Yan, Ziming |
| collection | NTU |
| description | This letter proposes a data-driven, model-free method for load frequency control (LFC) against renewable energy uncertainties based on deep reinforcement learning (DRL) in continuous action domain. The proposed method can nonlinearly derive control strategies to minimize frequency deviation with faster response speed and stronger adaptability for unmolded system dynamics. It consists of offline optimization of LFC strategies with DRL and continuous action search, and online control with policy network where features are extracted by stacked denoising auto-encoders. Numerical simulations verify the effectiveness and advantages of proposed method over existing approaches. |
| first_indexed | 2024-10-01T06:02:18Z |
| format | Journal Article |
| id | ntu-10356/141500 |
| institution | Nanyang Technological University |
| language | English |
| last_indexed | 2024-10-01T06:02:18Z |
| publishDate | 2020 |
| record_format | dspace |
| spelling | ntu-10356/1415002020-06-09T01:47:22Z Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search Yan, Ziming Xu, Yan School of Electrical and Electronic Engineering Engineering::Electrical and electronic engineering Continuous Action Search Deep Reinforcement Learning This letter proposes a data-driven, model-free method for load frequency control (LFC) against renewable energy uncertainties based on deep reinforcement learning (DRL) in continuous action domain. The proposed method can nonlinearly derive control strategies to minimize frequency deviation with faster response speed and stronger adaptability for unmolded system dynamics. It consists of offline optimization of LFC strategies with DRL and continuous action search, and online control with policy network where features are extracted by stacked denoising auto-encoders. Numerical simulations verify the effectiveness and advantages of proposed method over existing approaches. MOE (Min. of Education, S’pore) 2020-06-09T01:47:22Z 2020-06-09T01:47:22Z 2018 Journal Article Yan, Z., & Xu, Y. (2019). Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search. IEEE Transactions on Power Systems, 34(2), 1653-1656. doi:10.1109/TPWRS.2018.2881359 0885-8950 https://hdl.handle.net/10356/141500 10.1109/TPWRS.2018.2881359 2-s2.0-85056584019 2 34 1653 1656 en IEEE Transactions on Power Systems © 2018 IEEE. All rights reserved. |
| spellingShingle | Engineering::Electrical and electronic engineering Continuous Action Search Deep Reinforcement Learning Yan, Ziming Xu, Yan Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search |
| title | Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search |
| title_full | Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search |
| title_fullStr | Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search |
| title_full_unstemmed | Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search |
| title_short | Data-driven load frequency control for stochastic power systems : a deep reinforcement learning method with continuous action search |
| title_sort | data driven load frequency control for stochastic power systems a deep reinforcement learning method with continuous action search |
| topic | Engineering::Electrical and electronic engineering Continuous Action Search Deep Reinforcement Learning |
| url | https://hdl.handle.net/10356/141500 |
| work_keys_str_mv | AT yanziming datadrivenloadfrequencycontrolforstochasticpowersystemsadeepreinforcementlearningmethodwithcontinuousactionsearch AT xuyan datadrivenloadfrequencycontrolforstochasticpowersystemsadeepreinforcementlearningmethodwithcontinuousactionsearch |