کنفرانس مهندسی برق ایران

صفحه اصلی / سی و چهارمین کنفرانس بین المللی مهندسی برق

Artificial Intelligence and Deep Learning for Electric Motor Fault Diagnosis: Models, Methods and Applications

نویسندگان :

Melika Joharchi¹ Mojtaba Mirsalim² Gevork Gharehpetian³

1- دانشگاه صنعتی امیرکبیر 2- دانشگاه صنعتی امیرکبیر 3- دانشگاه صنعتی امیرکبیر

کلمات کلیدی :

Electric motors،fault diagnosis،condition monitoring،artificial intelligence،deep learning

چکیده :

Electric motors are critical components in industrial systems, and their unexpected failure can lead to significant economic losses and safety risks. Conventional fault diagnosis techniques, such as vibration analysis and motor current signature analysis, rely on expert-defined features and often struggle under variable operating conditions. Recent advances in artificial intelligence (AI) and deep learning have enabled data-driven approaches that automatically learn fault related patterns from sensor signals. This paper presents a focused review of AI and deep learning-based methods for electric motor fault diagnosis. Commonly used models, including convolutional neural networks (CNN), recurrent neural networks (RNN), autoencoders (AE), and Transformer-based architectures, are discussed in relation to typical motor fault types. A comparative perspective between traditional and AI-based approaches is provided, highlighting their respective strengths and limitations. Key challenges related to data availability, generalization, interpretability, and deployment are also examined, along with emerging research directions. The paper aims to provide practical insight into the application of intelligent diagnostic techniques for reliable motor condition monitoring.