Deep learning based transformer fault diagnosis method

Abstract

The detection of dissolved gas in oil is an important method for transformer fault diagnosis. By analyzing the type and content of gas in oil, the latent fault in equipment can be found as soon as possible, and the fault can be monitored. Aiming at the problems of low accuracy and scalability of traditional David triangle method and three ratio method, this paper proposes a fault diagnosis algorithm based on deep learning, which relies on the historical data of known fault gas types and concentrations, reserves the fault experience pool of transformer equipment, constructs a fault diagnosis model based on BP neural network, and uses the fault sample data that has occurred to complete the training and testing of the model. The experimental results show that the fault diagnosis model based on BP neural network has high accuracy.