Singular value decomposition based feature extraction approaches for classifying faults of induction motors

- We propose two robust feature extraction methods for fault classification on an induction motor.
- We propose a short-time energy (STE) plus singular value decomposition (SVD) technique in the time-domain analysis.
- We propose a discrete cosine transform (DCT) plus SVD technique in the frequency-domain analysis.
- We utilize multi-layer support vector machines (MLSVMs) to identify faults from the healthy ones.