Bearing fault detection in induction motor using pattern recognition techniques

In this paper a procedure based on pattern recognition technique is presented for fault diagnosis of rolling element bearings through artificial neural networks (ANN). The artificial neural networks are trained with a subset of the experimental data for known machine conditions. The networks are tested using the remaining set of data. In this method the characteristic features of time and frequency domain vibration signals of the rotating machinery with normal and defective bearings have been used as inputs to the ANN. The features are obtained from direct processing of the signal segments using very simple preprocessing. Three different cases; healthy, inner race defect, and outer race defect is classified using the proposed algorithm. The obtained results indicate that using time-domain features can be effective in the diagnosis of various motor bearing faults quickly and with high precision.