Physics-informed Deep Reinforcement Learning-based Adaptive Generator Out-of-step Protection for Power Systems

被引:0
|
作者
Hossain, Ramij R. [1 ,2 ]
Mahapatra, Kaveri [1 ]
Huang, Qiuhua [1 ]
Huang, Renke [1 ]
机构
[1] Pacific Northwest Natl Lab, Richland, WA 99354 USA
[2] Iowa State Univ, Iowa City, IA 50011 USA
来源
2023 IEEE POWER & ENERGY SOCIETY GENERAL MEETING, PESGM | 2023年
关键词
Out-of-step; Generator protection; Deep reinforcement learning; Augmented random search; Action Mask;
D O I
10.1109/PESGM52003.2023.10252299
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
This article presents a deep reinforcement learning-based control framework for adaptive generator protection in wide-area power systems. Out-of-step (OOS) generator tripping is an effective emergency control measure for mitigating system-wide black-out risks following any severe disturbance. Traditional protection schemes utilize rule-based mechanisms that fail to adapt to changing operating conditions. With the recent advances in deep reinforcement learning (DRL), the primary objective of our proposed methodology is to learn a DRL agent that: (a) can timely identify and isolate the affected generators after any potential disturbance and thereby maintain the system stability and (b) can adapt in unseen scenarios. But learning to identify an optimal set of generators for bulk power systems under various operating conditions is prohibitive due to: (a) the combinatorial nature of the problem, (b) the exponential increase of action space, and (c) the ultra-selectivity of the generator trip-action. To address these key challenges, we utilized the concept of action masks integrating system physics in the learning process, thereby blocking unnecessary actions in the exploration phase of the policy training, where the action masks are learned in conjunction with the DRL policy. In the policy part, we utilized a derivative-free parallel augmented random search (PARS)-based DRL algorithm, which is fast and highly scalable. Finally, we validated the proposed methodology with IEEE 300-bus systems.
引用
收藏
页数:5
相关论文
共 50 条
  • [41] Motion control of autonomous underwater vehicle based on physics-informed offline reinforcement learning
    Li, Xinmao
    Geng, Lingbo
    Liu, Kaizhou
    Zhao, Yifeng
    Du, Weifeng
    OCEAN ENGINEERING, 2024, 313
  • [42] Simultaneous mapping of nearshore bathymetry and waves based on physics-informed deep learning
    Chen, Qin
    Wang, Nan
    Chen, Zhao
    COASTAL ENGINEERING, 2023, 183
  • [43] Heat source field inversion and detection based on physics-informed deep learning
    Chi, Yimeng
    Li, Mingliang
    Long, Rui
    Liu, Zhichun
    Liu, Wei
    INTERNATIONAL COMMUNICATIONS IN HEAT AND MASS TRANSFER, 2025, 164
  • [44] Ice-flow model emulator based on physics-informed deep learning
    Jouvet, Guillaume
    Cordonnier, Guillaume
    JOURNAL OF GLACIOLOGY, 2023,
  • [45] Physics-Informed Neural Networks for Learning the Parameters of Commercial Adaptive Cruise Control Systems
    Apostolakis, Theocharis
    Ampountolas, Konstantinos
    2023 62ND IEEE CONFERENCE ON DECISION AND CONTROL, CDC, 2023, : 1523 - 1528
  • [46] Deep Reinforcement Learning-Based Power Management for Chiplet-Based Multicore Systems
    Li, Xiao
    Chen, Lin
    Chen, Shixi
    Jiang, Fan
    Li, Chengeng
    Zhang, Wei
    Xu, Jiang
    IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, 2024, 32 (09) : 1726 - 1739
  • [47] Explore missing flow dynamics by physics-informed deep learning: The parameterized governing systems
    Xu, Hui
    Zhang, Wei
    Wang, Yong
    PHYSICS OF FLUIDS, 2021, 33 (09)
  • [48] Machine Learning-Based MPC of Batch Crystallization Process Using Physics-Informed RNNs
    Wu, Guoquan
    Wu, Zhe
    IFAC PAPERSONLINE, 2023, 56 (02): : 2846 - 2851
  • [49] A data-driven tracking control framework using physics-informed neural networks and deep reinforcement learning for dynamical systems
    Faria, R. R.
    Capron, B. D. O.
    Secchi, A. R.
    De Souza, M. B.
    ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2024, 127
  • [50] An adaptive physics-informed deep learning method for pore pressure prediction using seismic data
    Zhang, Xin
    Lu, Yun-Hu
    Jin, Yan
    Chen, Mian
    Zhou, Bo
    PETROLEUM SCIENCE, 2024, 21 (02) : 885 - 902