A Comparative Study between Machine Learning Algorithm and Artificial Intelligence Neural Network in Detecting Minor Bearing Fault of Induction Motors

Shrinathan Esakimuthu Pandarakone, Yukio Mizuno, Hisahide Nakamura

Most of the mechanical systems in industries are made to run through induction motors (IM). To maintain the performance of the IM, earlier detection of minor fault and continuous monitoring (CM) are required. Among IM faults, bearing faults are considered as indispensable because of its high probability incidence nature. CM mainly depends upon signal processing and fault detection techniques. In recent decades, various methods have been involved in detecting the bearing fault using machine learning (ML) algorithms. Additionally, the role of artificial intelligence (AI), a growing technology, has also been used in fault diagnosis of IM. Taking the necessity of minor fault detection and the detailed study about the role of ML and AI to detect the bearing fault, the present study is performed. A comprehensive study is conducted by considering various diagnosis methods from ML and AI for detecting a minor bearing fault (hole and scratch). This study helps in understanding the difference between the diagnosis approach and their effectiveness in detecting an IM bearing fault. It is accomplished through FFT (fast Fourier transform) analysis of the load current and the extracted features are used to train the algorithm. The application is extended by comparing the result of ML and AI, and then explaining the specific purpose of use.