Neural approach for bearing fault detection in three phase induction motors

The induction motor has been widely used in various industrial applications. Thus, several studies have presented strategies for the diagnosis and prediction of failures in these motor. One strategy used recently is based on intelligent systems, in particular, artificial neural networks. The purpose of this paper is to present an alternative tool to traditional methods for detection of bearing failures using on a perceptron network with signal analysis in time domain. Experimental results are presented to validate the proposal.