关键词: 地铁供电系统, 自愈控制, 多智能体系统, IEC 61850, 故障诊断, 智能决策

Abstract: As metro systems have become a critical component of urban public transportation, the stability and reliability of their power supply systems are paramount. Unlike traditional high-voltage transmission grids, metro power supply systems operate within enclosed AC-DC hybrid networks, facing unique challenges such as high dynamic load variations and short but dense power supply links. Additionally, faults in metro power systems directly impact train operation safety, necessitating self-healing controls with millisecond-level response times and high robustness. Traditional fault handling methods are increasingly inadequate to address the demands of modern complex operating environments and expanding power supply networks. Consequently, self-healing control methods for power systems have garnered widespread attention. By leveraging real-time monitoring, intelligent diagnostics, and automated recovery, self-healing control ensures the safety and continuity of power supply systems during faults. This paper begins with a comprehensive review of the research progress on self-healing control methods for power systems, focusing on their key technologies and application status. It further explores the specific applications of Multi-Agent Systems (MAS) and the IEC 61850 communication standard in self-healing control within the unique context of metro power systems. Additionally, this paper analyzes the challenges of self-healing technologies in com-plex scenarios, such as real-time response capabilities, the reliability of artificial intelligence algorithms, and their role in system safety. Finally, future research directions are proposed, emphasizing the integration of big data, intelligent algorithms, and distributed control. This review study provides valuable theoretical support for self-healing control in metro power systems and contributes significantly to enhancing the safety, stability, and intelligence of these systems.

Key words: Metro power supply system, self-healing control, multi-agent System, IEC 61850, fault diagnosis, intelligent decision-making