Rolling Bearing Fault Diagnosis Research Based on Wavelet Spectrum Analysis

The rolling bearing fault diagnosis method is proposed.Wavelet default threshold method is adopted to make data denoising processing,and 5 layer wavelet decomposition is made to denoised vibration data.Wavelet reconstruction is done for the layer of fault characteristic frequency,power spectral density is computed finally.The simulation result for vibration signal of rolling bearing fault denotes that this method can effectively identify the inner ring,outer ring and rolling body faults of rolling bearing.