Wavelet Decomposition for the Detection and Diagnosis of Faults in Rolling Element Bearings

Condition monitoring and fault diagnosis of equipment and processes are of great concern in industries. Early fault detection in machineries can save millions of dollars in emergency maintenance costs. This paper presents a wavelet-based analysis technique for the diagnosis of faults in rotating machinery from its mechanical vibrations. The choice between the discrete wavelet transform and the discrete wavelet packet transform is discussed, along with the choice of the mother wavelet and some of the common extracted features. It was found that the peak locations in spectrum of the vibration signal could also be efficiently used in the detection of a fault in ball bearings. For the identification of fault location and its size, best results were obtained with the root mean square extracted from the terminal nodes of a wavelet tree of Symlet basis fed to Bayesian classier.