Stator and Bearing Faults Diagnosis of Induction Motor

This research presents an intelligent and automated diagnostic framework for early detection of stator and bearing faults in three-phase induction motors (IMs), which are vital components in industrial, commercial, and residential systems. Combining experimental data with advanced AI techniques like fuzzy logic, neural networks, and support vector machines, the study develops high-accuracy models for stator fault classification. For bearing fault diagnosis, a multi-stage methodology is introduced using statistical time-domain features, KPCA-SVM classifiers, and a novel Adaptive Modified Morlet Wavelet (AMMW) transform. The proposed techniques demonstrate excellent performance even in noisy conditions, offering a highly reliable solution for real-time IM fault monitoring and predictive maintenance.