Abstract:
Induction motors serve as critical components in numerous industrial applications, driving various machinery and processes. However, their reliable operation is often jeopardized by the occurrence of faults, leading to costly downtime, decreased productivity, and potential safety hazards. Traditional maintenance practices relying on scheduled inspections are often inefficient and prone to overlooking incipient faults, necessitating the development of advanced condition monitoring systems. Traditional fault detection systems for induction motors typically rely on periodic inspections and manual testing conducted by maintenance personnel. These systems often involve visual inspections of motor components, such as examining the condition of bearings, checking for signs of overheating, and inspecting electrical connections for loose connections or damage. Previous research has explored the integration of various methodologies including fuzzy logic, neural networks, and Fast Fourier Transform (FFT) for fault detection in induction motors. Fuzzy logic enables the modeling of imprecise or uncertain information, neural networks offer robust pattern recognition capabilities, and FFT facilitates the decomposition of signals into frequency components for fault signature analysis. The proposed idea is to use SVM in failure diagnostics. Theoretically, the SVM is an excellent classifier or regressor possessing a solid theoretical foundation. Practically, the SVM performs well in failure diagnostics, as shown in the cases presented. Finally, as failure diagnostics critically relies on feature extraction, this thesis considers feature extraction from the time domain
