Intelligent Diagnosis of Subway Traction Motor Bearing Fault Based on Improved Stacked Denoising Autoencoder

Aiming at the problem that complex working conditions affect effect of manual feature extraction in bearing fault diagnosis metro traction motor, a method motor based on improved stacked denoising autoencoder (SDAE) is proposed. This extracts features directly from original vibration signal through deep learning, reduces dependence processing technology and experience, solves unsatisfactory extracting values under conditions. The SDAE network structure accuracy studied experiments, best parameters are selected. test results show proposed can well extract condition variable speed load; when using data sets with conditions, classification better than many traditional methods.