The processing of the rolling bearing´s fault signal based on wavelet analysis

A key problem of a rolling bearing´s fault diagnosis is the feature extraction of the fault signal. Wavelet analysis can accomplish local analysis in the time-domain and frequency-domain, which is widely used in many different fields, such as signal processing, image processing and fault diagnosis. In this paper, we analyze the fault signal of a rolling bearing under different rotation rates, and analyze the vibration signal with damage in the inner ring, with damage in the outer ring and with damage in the ball. Basing on the wavelet analysis, signals are separated and reconstructed and a new method to figure out the characteristic parameter of the AE signal of rolling bearings is studied.