METHOD OF FAULT FEATURE EXTRATION BASED ON CEEMD AND FASTICA

Abstract

In order to solve the problem that the fault feature information of rolling bearing is difficult to be separated,a new method of fault feature extraction is presented,which is based on the complementary ensemble empirical mode decomposition（ CEEMD） and fast independent component analysis（ Fast ICA）. First,analyze the CEEMD vibration signals,decompose them into intrinsic mode function（ IMF） components signal of different scales; then through the sensitivity evaluation algorithm,decompose and recombine the signals,and use Fast ICA to reduce their noise; in the end,conduct Hilbert envelope spectrum analysis to the signals separated by the Fast ICA,to obtain the fault feature information. This method is applied to the fault analysis of rolling bearing vibration signal,and was proved to be valid.

Keywords

• CEEMD;Sensitive factor;Fast independent component analysis;Rolling bearing;Fault diagnosis