Wavelet Based Kernel Fisher Classifier For Ferroresonance Identification

This paper presents a Wavelet Kernel Fisher Classifier (WKFC) for ferroresonance detection. Using this method Ferroresonance can be discriminate from other transients such as capacitor switching, load switching and transformer switching. Wavelet transform is used for decomposition of signals, feature selection is done by Kernel Principal Component Analysis (KPCA). The Fisher classifier is applied to multiresolution featured KPCA mapped data. Ferroresonance data and other transients were obtained by simulation using EMTP program. Results show that the proposed procedure is efficient in identifying ferroresonance from other events.