Fault self-healing multi-strategy collaborative optimization method and system for power distribution network

Through the multi-agent reinforcement learning and strategy collaboration mechanism, the state space and reward function of the agent are designed, which solves the problem that existing technology is difficult to cope with complex fault scenarios, and improves the flexibility and adaptability of the distribution network fault self-healing system, ensuring the reliability and stability of the power grid.

CN119940582APending Publication Date: 2025-05-06GUIZHOU POWER GRID CO LTD
View PDF 4 Cites 3 Cited by

Patent Information

Application Number
CN202411702916.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-26
Publication Date
2025-05-06

Smart Images

  • Figure CN119940582A_ABST
    Figure CN119940582A_ABST
Patent Text Reader

Abstract

The invention discloses a power distribution network fault self-healing multi-strategy collaborative optimization method and system. The method comprises the following steps: acquiring a self-healing strategy; designing an agent according to the self-healing strategy, and designing a state space, an action space and a reward function of the agent; training the agents by using a multi-agent reinforcement learning algorithm, a strategy cooperation mechanism and a transfer learning method to obtain a self-healing strategy combination; and establishing a multi-dimensional evaluation index system, and evaluating the self-healing strategy combination. According to the power distribution network fault self-healing multi-strategy collaborative optimization method provided by the invention, based on a multi-agent reinforcement learning theory and a game theory, different self-healing strategies are regarded as independent agents, and an optimal strategy combination is realized through collaborative learning and optimization. And meanwhile, a transfer learning technology is introduced, so that the adaptability of the model in a new scene is improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Microgrid group interactive game strategy learning evolution method based on Nash Q learning

    CN116050632A

  • Active power distribution network elastic self-healing method considering dynamic micro-grid and controller

    CN117791560A

  • Dynamic management method for security resources of Internet of Vehicles based on transfer learning and multi-agent game

    CN118540693A

  • Multi-agent system fault self-recovery method and device, computer equipment and medium

    CN118689685A

Cited By

  • Power-cut plan arrangement method and system based on multi-agent interpretable reinforcement learning

    CN120197915A

  • Power distribution network fault self-recovery method based on neural network and multi-agent cooperation

    CN122051987A

  • Power distribution network fault self-healing method based on neural network and multi-agent cooperation

    CN122051987B