A Comprehensive Survey of Electrical Signal supported Condition Monitoring Methodology for Detecting with Diagnosing Faults in Induction Motor

The seamless and efficient process of 3 phase induction motor (IM) is crucial for modern industries. This paper begins by reviewing traditional time-domain and frequency-domain analyses of trembling and current signature used toward diagnose various IM faults, using data from dissimilar faulty IMs in a laboratory setting. It then discusses the benefits and challenges of these conventional methods. Following this, the paper summarizes current research and advancements in signal-based automation techniques for condition monitoring, focusing on detecting and diagnosing different electrical and mechanical faults in IMs. Recently, artificial intelligence (AI) techniques have been increasingly utilized for IM and other machinery fault diagnosis. The paper provides a detailed review of advancements in AI-based fault analysis, including trendy methods, which are being integrated with traditional monitoring approaches. The focus is primarily on techniques developed in the past decade. The review covers system signals, conventional and advanced signal processing methodologies, and emphasizes selecting effective arithmetical features, AI methods, & associated teaching and testing strategies for IM fault diagnostics. In addition, the paper talk about current advancement, identifies research gaps, and explores outlook directions in IM fault monitoring and analysis.