The Research Of The Rolling Bearing Fault Diagnosis Based On Methods Of Modern Signal Analysis

Rolling bearing is the widely used but easily to be damaged part in rotating machine.The non-steady vibration signal produced by the rolling bearing which is in work is very complicated.It not only contains the information of rolling bearing itself, but also contains the information of a large number of other moving parts.This paper summarizes the vibration characteristics and fault mechanism of the rolling bearing.Bases on the rolling bearing faults' experiments,the article adopts the Eigenvalue Analysis,the Power Spectrum Analysis,the Cepstrum Analysis,the Short-Time Fourier Transform(STFT),the Wavelet analysis,the Quadratic Time-Frequency Distribution,the Stochastic Resonance and other methods to analyse the experiments' data of the rolling bearing,especially introduces the Quadratic Time-Frequency Distribution and the Stochastic Resonance,contrasts the applicability of different data analyzing methods and enriches the modern signal analysis methods in the rolling bearing fault analysis.