A Study on Machine Learning and Artificial Intelligence Methods in Detecting the Minor Outer-Raceway Bearing Fault

To increase the reliability of induction motor (IM), several techniques have been proposed in condition monitoring and fault diagnosis. Bearings are the most sensitive part of IM, and fault occurring must be considered. In industry, scratch seems the most frequently occurring fault in bearing, and only few researches have encountered this issue. This paper is motivated by considering hole and scratch as faulty factor. A fast Fourier transform analysis is carried out, the features are extracted and used for training the diagnostic algorithm. Two types of diagnosis methods are proposed; machine learning algorithm and artificial intelligence method (AI). Among the various machine learning algorithms, Support Vector Machine (SVM) is selected because of its superiority over the data preprocessing. Deep Learning Algorithm (DL) is selected in case of AI because of its intrinsic property over feature learning. A Convolutional Neural Network (CNN) architecture is originally used for fault characterization. The advantage of both the diagnosis methods and the possibility in detecting the minor bearing faults are discussed. Finally, effectiveness of the proposed methods is validated based on the experimental and diagnosis results.