Online condition monitoring and fault detection in induction motor bearings

Abstract

Induction motors (IMs) are commonly used in industry. Online IM health condition monitoring aims to recognize motor defect at its early stage to prevent motor performance degradation and reduce maintenance costs. The most common fault in IMs is related to bearing defects. Although many signal processing techniques have been proposed in literature for bearing fault detection using vibration and stator current signals, reliable bearing fault diagnosis still remains a challenging task. One of the reasons is that a rolling element bearing is not a simple component, but a system; its related features could be time-varying and nonlinear in nature. The objective of this study is to investigate an online condition monitoring system for IM bearing fault detection. The monitoring system consists of two main modules: smart data acquisition (DAQ) and bearing fault detection. In this work, a smart current sensor system is developed for data acquisition wirelessly. The DAQ system is tested for wireless data transmission, consistent data sampling, and low power consumption. The data acquisition operation is controlled by using an adaptive interface. In bearing fault detection, a generalized Teager-Kaiser energy (GTKE) technique is proposed for nonlinear bearing feature extraction and fault detection using both vibration and current signals. The proposed GTKE technique will demodulate the signal by tracking the instantaneous signal energy. An optimization method is proposed to enhance the fault-related features and improve signal-to-noise ratio. The effectiveness of the proposed technique is verified experimentally using a series of IM tests. The robustness is examined under different operating conditions.