An enhanced morphology gradient product filter for bearing fault detection

Abstract

Abstract This paper presents a signal processing scheme, namely enhanced morphology gradient product filter (EMGPF), for rolling element bearing fault detection. In this scheme, a morphology gradient product operation (MGPO) is firstly proposed to extract impulsive features of a raw signal according to a comprehensive investigation of the working mechanism of the reported morphological operations. Then, a higher-order spectrum analysis method, the third-order cumulant slice spectrum, is used to improve the performance of the MGPO based morphology filter for the purpose of highlighting fault features further. Experimental vibration signals were employed to evaluate the effectiveness of the proposed EMGPF. Results show that the proposed method has a superior performance in extracting fault features of defective rolling element bearing over four reported morphology filters.