Research and Application of Intelligent Diagnosis Method of Mechanical Fault Based on Transformer Vibration and Noise and BP Neural Network

In this paper, a fault diagnosis method based on the transformer vibration noise and BP neural network is proposed. The transformer vibration noise signal is obtained by the vibration noise detection system, and the eigenvalue is calculated by FFT. The eigenvalue is used as the input of the trained BP neural network and the type of transformer fault is predicted. The validity of the method is verified by the diagnosis of typical state test of 6 kinds of transformers. This method makes full use of the vibration and noise signal of transformer, and realizes the diagnosis of transformer live fault by BP neural network algorithm. It greatly improves the fault diagnosis rate of the transformer. It greatly improves the fault diagnosis rate of the transformer and provides an effective way for the transformer live inspection.