A Fault Diagnosis Method for High-Speed Train Wheelset Bearings Based on Deep Learning

Abstract

Aiming at the problem that the traditional fault diagnosis method is difficult to effectively extract the fault features of the high-speed train wheelset bearing signal, this paper proposes a fault diagnosis method based on the two-dimensional image method. First, the one-dimensional vibration signal is converted into a Two-dimensional grayscale image, eliminating the influence of expert experience on the feature extraction process. Then an improved network model is proposed, which can automate the process of feature extraction and fault diagnosis. Finally, this paper simulates the complex driving environment of high-speed trains by adding noise with different SNRs to the vibration signal and analyzes the influence of noise on the diagnostic ability of the method. The results show that the method is effective.