An effective approach to rolling bearing diagnosis based on Adaptive Redundant Second-Generation Wavelet

De-noising and extraction of weak signals are crucial to fault prognostics, and the wavelet transform has been widely used in signal de-noising. In this paper, a new method, which combines the Adaptive Redundant Second-Generation Wavelet (ARSGW) and the Hilbert transform, is proposed. The ARSGW is applied to reveal the transient components of the signal in time domain clearly. Then the Hilbert transform is used to extract fault features of rolling bearing from the wavelet packets. The analysis results of the vibration signals from the experiment and the machine tool spindle show that the proposed method can detect the faults of the rolling bearing effectively.