Broken rotor bar fault classification for induction motor based on support vector machine-SVM

The detection and classification of faults for induc- tion motors (I. M) are significant important for researchers. Early detection of the faults prevents the stopping for industrial systems and reduces the cost of maintenance. In this article, broken bars fault classification in 3 phi induction motor are presented and successfully evaluated. The proposed scheme is used the Support Vector Machine (SVM) technique. The envelope of stator current has been estimated in order to extract five features. Principal Component Analysis are presented in order to fed the classifier with suitable classes. The experimental results showed the success of the proposed method for the diagnosis of broken rotor bar for induction motor.