Fault Diagnosis Bearings in Induction Motor Using Improved Extended Classifier System

Equipment and machines that use electric motors are widely used in the industry. Bearings as one of the most used engine parts that are constantly under load/rotation and may crash much sooner than other parts. Extensive research has therefore been conducted on the healthmonitoring of induction motor bearings. Bearing faults are often local and occur in the outer/inner race, cage and balls. The amplitude/period of the frequent pulses due to the defect repeating with the rotational speed can be analyzed for fault detection. Vibration signals have thus valuable informationthat can be use in bearing health monitoring. The fault location and sometimes its severity is then determined by an appropriate fault identification algorithm. In this paper, an intelligent system is designed to monitor the health of induction motor bearings. A comparison is made between the resultsof different methods of classification. Both computational cost and the percentage of correct predictions are compared with the tests. The collection of data on health status of faults in the outer/inner ring resulted from the proposed method shows an improvement over other methods discussed in the paper.

Keywords:

Intelligent Systems , Induction Motor , Health monitoring , signals