Condition Monitoring and Fault Diagnosis of Induction Motor Faults using Deep Learning : A Review

Abstract

Induction motors are widely used in different sectors for various industrial applications and their continuous operation is critical for ensuring the continuity of production processes and saving costs. Condition monitoring and fault diagnosis of induction motors have been of great interest to researchers and practitioners for many years. Deep learning, as an emerging machine learning technique, has shown great potential in this field. This paper reviews the recent research on condition monitoring and fault diagnosis of induction motors using deep learning. The paper introduces the types of faults in induction motors and mentions some traditional and shallow machine learning methods that have been used for condition monitoring and fault diagnosis. The second section summarizes the research on deep learning-based approaches for motor fault diagnosis, including various deep learning architectures and techniques such as autoencoders (AEs), deep belief networks (DBNs), recurrent neural networks (RNNs), generative adversarial networks (GANs), and convolutional neural networks (CNNs). Finally, the paper discusses the challenges encountered in using deep learning techniques, solutions to these problems and highlights opportunities for future research in this field.