A Simple Continuous Wavelet Transform Method for Detection of Rolling Element Bearing Faults and its Comparison with Envelope Detection

Sunil Tyagi, SK Panigrahi

Abstract: Envelope Detection (ED) is traditionally used to identify the rolling element bearing faults. ED involves carrying considerable signal processing. In past few years Wavelet Analysis (WA) has been widely used an alternative tool for detection of machinery fault diagnosis as WA is able to provide multi-resolution analysis and give both frequency and time information suitable to analyse non-stationery signals such as vibration signal from bearing. This paper presents a very simple Wavelet analysis technique using Continuous Wavelet Transform (CWT) that do not require complex signal processing. A comparative study of effectiveness of ED and CWT method for bearing fault diagnosis is presented in this paper and it is shown with help of experimental results that the proposed simple CWT based method is better fault diagnostic tool for bearing fault identification.

Keywords: Ball bearing, Continuous Wavelet Transform CWT, Envelope Detection ED, Fault Diagnosis, High Frequency Resonant Technique HFRT, Hilbert Transform, Rolling element bearing, Machinery Vibration