Rolling-Element Bearing Fault Detection in Three-Phase Induction Motors

Due to the widespread usage of three-phase induction motors in various industries, including power plants, refineries, petrochemicals, mining, transportation, shipping, etc., requirements for monitoring their conditions to detect various incipient defects are increasingly growing. It is necessary to detect the induction motor faults, which may cause the motor to stop unintentionally, in the shortest time with the highest accuracy in order to eliminate the problem with prior planning before spreading it. Some effective indices for detecting different types of the rolling-element bearing localized faults are derived from the motor vibration signal. In this paper, the mentioned indices are measured and analyzed in different scenarios to evaluate their reliability using laboratory test results. The results show that the influence of the motor temperature in the range from 30 to 75oC on the magnitudes of the indices is between 1 to 3 dbs; considering the magnitudes of the indices, such influence is rather negligible. Similarly, increasing the motor load from no-load up to full-load can influence the magnitude of the best selected indices by up to 2 dbs. In addition, increasing the localized defect severity (the hole diameter) from 2 to 5 mm usually affects the indices magnitudes by an oscillating manner with a maximum oscillation amplitude of 20 dbs.