Diagnosis of ball bearing with various degree rolling element defects based on local mean decomposition

This paper aims to explore fault characteristics extraction method to deal with the light and severe rolling element faults based on local mean decomposition (LMD) and Fourier transform (FT). The characteristics of LMD are studied by processing multi-component frequency modulation and amplitude modulation (FM-AM) signal, which are usually used to describe the bearing fault signals. Based on the simulation analysis, the envelope spectrum method called LMD-FT is used to deal with the vibration signals of rolling balling bearing with various element faults. Moreover, the fault features extracted by LMD are compared with those obtained from conventional Hilbert transform (HT). The results demonstrate that LMD method is an effective way to identify the characteristics of rolling element defect generated at the initial stage.