Research on Transformer Fault Prediction Based on Graph Neural Network

Abstract

Power transformer is an important part of the power generation system, as the bottom of the power network, directly connected with the user, only the normal work state of the transformer can ensure the safety and smooth operation of the power system. If the transformer breaks down, it will directly not only affect social production, can timely find power transformer latent fault, accurate judgment fault types and characteristics, to maintenance personnel to formulate appropriate repair plan provides guidance significance, at the same time to improve the power grid reliability and reduce the economic consequences of the transformer faults has important practical significance