Mitigating cascading failure in power grids with deep reinforcement learning-based remedial actions

• Published in: Reliability engineering & system safety Vol. 250; p. 110242
• Main Authors: Zhang, Xi, Wang, Qin, Bi, Xiaowen, Li, Donghong, Liu, Dong, Yu, Yuanjin, Tse, Chi Kong
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2024
• Subjects: Cascading failure; Deep reinforcement learning; Mitigation; Power grid; Proximal policy optimization; Remedial action; Mitigation; Deep reinforcement learning; Proximal policy optimization; Remedial action; Cascading failure; Power grid
• ISSN: 0951-8320; 1879-0836
• DOI: 10.1016/j.ress.2024.110242

Power grids are susceptible to cascading failure, which can have detrimental consequences for modern society. Remedial actions, such as proactive islanding, generator tripping, and load shedding, offer viable solutions to mitigate cascading failure in power grids. The success of applying these solutions lies in the timeliness and the appropriate choice of actions during the rapid propagation process of cascading failure. In this paper, we introduce an intelligent method that leverages deep reinforcement learning to generate adequate remedial actions in real time. A simulation model of cascading failure is first presented, which combines power flow distribution and the probabilistic failure mechanisms of components to accurately describe the dynamic cascading failure process. Based on this model, a Markov decision process is formulated to address the problem of deciding on the remedial actions as the failure propagates. Proximal Policy Optimization algorithm is then adapted for the training of underlying policies. Experiments are conducted on representative power test cases. Results demonstrate the out-performance of trained policy over benchmarks in both power preservation and decision times, thereby verifying its advantages in mitigating cascading failure in power grids. •A Markov decision process is formulated to determine real-time remedial action scheme (RAS) for mitigating cascading failure in power systems.•A proximal policy optimization (PPO) adaptation is developed for the training of RAS policies offline.•Numerical studies are performed based on cascading failure simulations in IEEE 39-bus and 57-bus systems.•The PPO-based RAS can effectively slow down or inhibit the propagation of cascading failure.