Envelope demodulation method based on SET for fault diagnosis of rolling bearings under variable speed

Abstract

Time-frequency analysis can effectively reveal the fault characteristics of rolling bearings under variable speed conditions. However, the traditional time-frequency analysis method cannot meet the requirements of anti-noise ability and time-frequency concentration. The synchroextracting transform (SET) as a novel time-frequency analysis method makes up for these defects, so the envelope demodulation time-frequency analysis method based on SET for fault diagnosis is proposed. Firstly, the best resonant band is selected by the adaptive band-pass filter based on fast spectral kurtosis. Then, the amplitude envelope of filtered signal is calculated by the Hilbert transform. Finally, SET is applied to analyze the amplitude envelope, and the fault information can be extracted from the envelope demodulation spectrum. The simulation signal and experimental signal are analyzed respectively. The results prove the effectiveness of the method. Compared with the short-time Fourier transform (STFT) and synchrosqueezing transform (SST), the proposed method is more prominent in the anti-noise property and time-frequency concentration. In addition, the performance in case of the band-pass filter failure is tested. The results show that the proposed method is still valid, and has excellent robustness and self-adaptability. Therefore, the proposed method shows significant application prospects and promotion value for fault diagnosis of variable-speed rolling bearing.