Bearing Fault Diagnosis Based on Empirical Wavelet Transform and Singular Value Decomposition

Abstract

Aiming at the problem of noise in the signal in bearing fault diagnosis, a diagnosis method based on empirical wavelet transform and singular value decomposition was proposed. Firstly, the vibration signal of the fault bearing outer ring was decomposed by empirical wavelet transform, and then the signal was reconstructed by singular value decomposition. The combination of the two not only decomposed the frequencies contained in the respective components, but also eliminated noise components and had high reliability in analyzing signals. The spectrum of the signal was analyzed to identify the fault signal in the Hilbert spectrum and compared to the theoretically calculated frequency. The experimental results show that the method can accurately determine bearing faults.