The purpose of this paper is to design a distributed intelligent distribution grid management system for rapid self-healing,and to construct a complete system architecture by using real-time monitoring,machine learning fault detection and location,deep reinforcement learning self-healing control and other technologies.A 10-day empirical study on a 10 kV distribution network in a suburb shows that the system can detect faults within 16.8 ms and complete self-healing within 178.3 ms,and the accuracy of fault location reaches 96.5%,which proves that the system can improve the power supply reliability and self-healing ability of the distribution grid.
关键词
分布式智能配电网/快速自愈/机器学习
Key words
distributed intelligent distribution grid/rapid self-healing/machine learning
维普
万方数据