Neural detector of rolling bearing faults in induction motors supplied from the frequency converter

In the paper possibility of neural networks application for the detection of bearing faults (such as: balls, outer and inner races damages) in the induction motor supplied from the frequency converter is presented. The training and testing data of neural networks were obtained from the experiments carried out with a set of specially prepared bearings (with different damages). Magnitudes of the characteristic harmonics obtained from the spectral analysis of the mechanical vibrations and the envelope analysis were used for training and testing of the developed neural detectors. Neural detector of rolling bearing faults in induction motors supplied from the frequency converter).