SR-GNN Based Fault Classification and Location in Power Distribution Network

Accurately evaluating the fault type and location is important for ensuring reliability of power distribution network. A mushrooming number distributed generations (DGs) connected to system brings challenges traditional classification methods. Novel AI-based methods are mostly based on wide area measurement with assistance intelligent devices, whose economic cost somewhat high. This paper develops a super-resolution (SR) graph neural network (GNN) method in It can accurately evaluate only by obtaining measurements some key buses network, which reduces construction system. The IEEE 37 Bus used testing proposed verifying its effectiveness. In addition, further experiments show that has certain anti-noise capability robust resistance change, reconfiguration, access.