The Application of ICA-SVM Method for Identifying Multiple Faults in Asynchronous Motors

Abstract

Vibration signal of asynchronous motors are detected with multiple sensors and used to identifying multiple faults of motors in this paper. Independent components analysis method (ICA) is applied to compress measurements from several channels into a smaller amount of channel combinations and separate related vibration signal from interfering vibration sources, and support vector machine (SVM) based multi-class classifiers are used to identify multiple faults of asynchronous motors. Adventures of the proposed methods are that they are data-based and are not necessary to build an analytical model, and SVM-based identification can provide better performance with few datasets. Experiment results are provided to illustrate the performance of the detection and identification methods.