Research on Fault Diagnosis of Rolling Bearing Based on Morlet Wavelet and Scale Space

Abstract

Aiming at the difficulty of extracting the early fault features of rolling bearings, a rolling bearing fault diagnosis method based on Morlet wavelet and scale space was proposed. The frequency band boundary was divided by the scale space to obtain the resonance frequency band, and the frequency band boundary information was substituted into Morlet wavelet to construct a filter bank to filter the signal. Due to the problem of over-segmentation of the scale space division frequency band boundary, the envelope correlation kurtosis was introduced as an index, and a scale space optimization spectrum method was proposed to identify the optimal resonance frequency band of the fault. The validity of the method was verified by simulation signals and bearing fault signals obtained from bench tests, and compared with the fast spectral kurtosis method. The results showed that this method could accurately identify the optimal resonance frequency band and realize bearing fault diagnosis. At the same time, the diagnosis effect was significantly better than fast spectral kurtosis index.

Keywords

• rolling bearing
• Morlet wavelet
• scale space
• fault diagnosis
• correlation kurtosis
• fast spectral kurtosis