Realtime Monitoring Instrument Reliability Of Three Phase Induction Motor Bearing Based On Neural Network (NN) Analysis

Abstract

Three-phase induction motors are one of the most widely used drives in industry and transportation because of their simple construction and reliability. Damage to the induction motor affects the existing production process. Therefore, early detection of induction motor damage is needed to avoid further damage and cause losses to the industry. The method of identifying bearing damage to the induction motor uses a no-load condition. The combination of FFT transformation and artificial neural net is used as a method of identifying the damage. The identification variable used in the method is taken from the stator current signal. To achieve the desired goal, the experimental data used are 10%, SEF 24%. bearing damage to the inside, the outside ball, and the separator. Simulation results show that protoype is able to read 85% of the identified training data for each type of damage