An Efficient Hilbert–Huang Transform-Based Bearing Faults Detection in Induction Machines

Abstract

This paper focuses on rolling elements bearing fault detection in induction machines based on stator currents analysis. Specifically, it proposes to process the stator currents using the Hilbert-Huang transform. This approach relies on two steps: empirical mode decomposition and Hilbert transform. The empirical mode decomposition is used in order to estimate the intrinsic mode functions (IMFs). These IMFs are assumed to be mono-component signals and can be processed using demodulation technique. Afterward, the Hilbert transform is used to compute the instantaneous amplitude (IA) and instantaneous frequency (IF) of these IMFs. The analysis of the IA and IF allows identifying fault signature that can be used for more accurate diagnosis. The proposed approach is used for bearing fault detection in induction machines at several fault degrees. The effectiveness of the proposed approach is verified by a series of simulation and experimental tests corresponding to different bearing fault conditions. The fault severity is assessed based on the IMFs energy and the variance of the IA and IF of each IMF.