Application of core vector machines for induction motor drive fault diagnosis

Abstract

This paper presents a core vector machine (CVM)-based framework for induction motor drive fault diagnosis. The proposed method can be used for diagnosis of different electrical drive faults such as switches short circuit, switches open circuit, etc. To classify motor drive faults, a CVM has been trained for each one. The proposed CVM-based fault detection algorithm has a very small training time and space in comparison with support vector machines (SVMs) and artificial neural networks (ANNs)-based in addition to the other kinds of fault detection algorithms. The proposed algorithm produces few support vectors (SVs), and therefore is faster than existing algorithms. One of the main points in application of a machine learning method is feature selection. In this study, a new decision tree (DT)-based feature selection algorithm has been presented. The proposed CVM based framework has been applied to a test case. The simulation results show the effectiveness and the accuracy of the proposed method for fault diagnosis of induction motor drive. The effectiveness of the proposed feature selection algorithm has also been investigated. The simulation results demonstrate the effectiveness of the proposed feature selection algorithm.