Signal processing methods for identification of induction motor bearing fault

Abstract

To diagnose early faults as soon as possible, the feature extraction of vibration signals is very important in real engineering applications. Recently, the advanced signal processing-based weak feature extraction method has been becoming a hot research topic. The dominant mode of failure in rolling element bearings is spalling of the races or the rolling elements. Localized defects generate a series of impact vibrations every time whenever running roller passes over the surface of a defect. Therefore, vibration analysis is a conventional method for bearing fault detection. However, the measured vibration signals of rotating machinery often present nonlinear and non-stationary characteristics. This paper deals with the diagnosis of induction motor bearing based on vibration signal analysis. It provides a comparative study between traditional signal processing methods, such as Power Spectrum, Short Time Fourier Transform, Wavelet Transform, and Hilbert Transform. Performances of these techniques are assessed on real vibration data and compared for healthy and faulty bearing.