ANN-based for detection, diagnosis the bearing fault for three phase induction motors using current signal

Abstract

This paper develop a novel, non-intrusive approach for fault-detection and diagnosis scheme of bearing faults for three-phase induction motor using stator current signals with particular interest in identifying the outer-race defect at an early stage. The most common bearing problem is the outer race defect in the load zone. The empirical mode decomposition (EMD) technique is proposed for analysis of non-stationary stator current signals. The stator current signal is decomposed in intrinsic mode function (IMF) using empirical mode decomposition. The extracted IMFs apply on the wigner-ville distribution (WVD) to have the contour pattern of WVD. Then, artificial neural network is used for pattern recognition that can effectively detect outer-race defects of bearing. The experimental results show that stator current-based monitoring with winger-ville distribution based on EMD yields a high degree of accuracy in fault detection and diagnosis of outer-race defects at different load conditions, also, a more significant and reliable indicator for detection and diagnosis of outer-race defects using artificial neural network. Experimental investigation is done and reported in the paper.