Research on Bearing Fault Diagnosis Based on Slice Spectrum and Cepstrum

Abstract

In this paper, high-order spectral analysis and cepstrum analysis are combined to diagnosis of bearing failure. The high-order spectrum can effectively suppress the Gaussian noise and extract the weak fault signal, but it has the limitation when dealing with the added signal. The cepstrum, which less affected by the transmission path, can distinguish the periodic components in the sideband and simplify the clustered sideband into a single spectral line. In this paper, the two methods are combined to conduct the bispectrum analysis of the vibration signal, and then do the Inverse Fourier transform to the logarithm of the bispectrum model, that is, the cepstrum analysis of the third-order cumulant. The method proposed in this paper is verified by the vibration signal data of a bearing, and the results show that the method proposed in this paper can overcome the shortcomings of the traditional single method and make effective diagnosis of the bearing failure.