Journal of Electrical Engineering and Automation

Rapid fault location and efficient self-healing of distribution networks are crucial for ensuring the stable operation of power systems. Traditional methods have problems such as time-consuming location and delayed self-healing responses, and the introduction of artificial intelligence technology provides a new path to solve these problems. By constructing a fault feature recognition model based on deep learning, key information in fault signals can be accurately extracted to achieve rapid determination of fault types and locations; combining reinforcement learning algorithms to optimize self-healing control strategies can dynamically adjust recovery plans according to the grid topology and load status, shortening the fault recovery time. This research provides technical support for improving the reliability and resilience of distribution networks and is of great significance for ensuring continuous power supply.

Artificial intelligence; Distribution network; Fault location; Self-healing control; Power system