• CSCD核心库收录期刊
  • 中文核心期刊
  • 中国科技核心期刊

ELECTRIC POWER CONSTRUCTION ›› 2023, Vol. 44 ›› Issue (12): 85-94.doi: 10.12204/j.issn.1000-7229.2023.12.007

• Fundamental Theory and Key Technology of New Power System Resilience ?Hosted by Professor XU Yin, Senior Engineer SHI Shanshan and Associate Professor WEI Wei? • Previous Articles     Next Articles

Rollout Algorithm-Based Post-Disaster Distribution System Emergency Repair Optimization

LI Mingxuan1(), WEI Wei1(), XU Yin2(), ZHANG Qiqi3()   

  1. 1. Department of Electrical Engineering, Tsinghua University, Beijing 100084, China
    2. School of Electrical Engineering, Beijing Jiaotong University, Beijing 100044, China
    3. State Grid Shanghai Municipal Electric Power Company, Shanghai 200122, China
  • Received:2023-06-05 Published:2023-12-01 Online:2023-11-29
  • Supported by:
    National Key R&D Program of China under Grant(2022YFB2405500);Science and Technology Project of State Grid Corporation of China(52094023001H)


In order to cope with the loss from power outages caused by extreme disasters, it is necessary to construct a reasonable post-disaster recovery model and explore effective post-disaster emergency decision-making methods. In this study, a Markovian decision process model for post-disaster emergency recovery considering critical load failure events is constructed. It also considers the impact of outage duration on the functionality of different types of critical loads when calculating the losses. For post-disaster emergency decision-making, an online decision-making method based on a preview algorithm is proposed to quickly evaluate the action-state value function (i.e., Q function) by means of multi-scenario simulation and generate an online optimization strategy. It is demonstrated that the decision scheme based on the Q function generated by the rollout algorithm is better than that provided by the underlying strategy in the simulation. The proposed method does not rely on precise mathematical models and parameters and can update the policy in real time according to the system state, which is a more practical approach than that of the traditional offline optimization model.

Key words: resilience, distribution system, post-disaster restoration, online decision, rollout algorithm

CLC Number: