Abstract: The order multi-scale morphology demodulation approach is proposed to extract the fault characteristic frequency from the vibration signal of a fault gearbox with rotating speed fluctuation. The chirplet path pursuit algorithm is used to obtain the rotation speed signal of the gearbox. According to the rotation speed signal, the time domain vibration signal of the gearbox is resampled at constant angle increments. An algorithm based on the information of local peaks of the signal is used to figure out the structuring elements for multiscale morphology analysis. The structuring elements are used to carry out morphology operations on the resampled signal. Spectral analysis on the average value of the operation results is carried out, so as to accomplish the order multi-scale morphology demodulation. Since the noise components do not correlate with the chirplet functions closely, the chirplet path pursuit algorithm can extract the rotation speed effectively even if the signal-to-noise ratio of the vibration signal is very low. By averaging the results of each scale morphology analysis, the multi-scale morphology demodulation can restrain the noise effectively. The proposed approach has good anti-noise ability, and is suitable for analyzing the actual vibration signal of a gearbox with rotating speed fluctuation. Simulation and practical application examples prove the validity and superiority of the proposed method.

Key words: Chirplet, Fault diagnosis, Multi-scale morphology, Order tracking, Signal-to-noise ratio