Enhanced Empirical Wavelet Transform with Application to Rolling Bearings Fault Diagnosis

Abstract

The Empirical Wavelet Transform (EWT) is a novel method for analyzing the multi-component signals by constructing an adaptive filter bank. Though it is an effective tool to identify the signal components, it has drawback in dealing with some noisy and non-stationary signals due to its coarse spectrum segmentation. To target this problem, an enhanced EWT (EEWT) is proposed to decompose the signal into empirical modes with physical meanings. This method ameliorates the drawback of EWT by taking the spectrum shape of the processed signal into account. It improves the segmentation process by adopting the envelop approach based on the order statistics filter (OSF) and applying three criteria to pick out useful peaks. In this work, the envelope spectrums of extracted empirical modes are applied to the rolling bearing fault diagnosis. Because the EEWT can decompose vibration signal into a set of mono-components, fault features are found clearly in the envelop spectrum. The effectiveness of the proposed method is verified by a simulated signal and a real signal captured from a test rig.