Diagnosis Of Oc And Sc Faults In Bldc Drive Using Naive Bayes Classifier

Abstract

- In this paper a Brushless Direct Current (BLDC) drive fault detection method is proposed based on the Naive Bayes Classifier. Naive Bayes classifier is one of the best machine learning methods available. Two major flaws in BLDC drives are open circuit and short circuit flaws that are addressed in this article. The effects of faults on BLDC drive are studied in the first stage. Then finding the BLDC drive’s feature values, i.e. current, voltage, speed, and torque in open circuit, short circuit faults and without any fault, these are called training data. After that, the extracted functions are used to train the classifiers to classify the type of fault that occurs in the system and the corresponding step response parameters. Compared to current methods of fault diagnosis (FD), such as the Hall effect sensor, Interturn faults are implemented in disparate device areas, our approach does not need complicated vector tracking observers, and it is capable of managing up to two different faults while significantly improving the diagnostic speed. Matlab simulation and experimental results verify the efficacy of the proposed method and demonstrate its advantages over FD method.