Partial discharge pattern recognition for cast-resin current transformer

Abstract

The applications of genetic algorithm (GA) and fuzzy c-means (FCM) clustering approach to recognize partial discharge (PD) patterns of cast-resin current transformer (CRCT) are proposed in this paper. The PD patterns are collected by a PD detecting system in the laboratory. Several statistical methods are used on the phase-related distributions in this paper to extract the features for clustering. After the feature extraction procedure, we employ GA for selection of optimal feature combination. Based on the optimal features selected by GA, the PD pattern represented by feature vectors are clustered through the FCM scheme with reasonable discrimination. To verify the effectiveness of the proposed technique, the experimental results and the analysis using 250 sets of field-test PD patterns from five artificial defect types of CRCTs are used in this paper. It has been shown that through the features extraction and optimal vector selection procedure, the extracted statistical featuring vectors can significantly reduce the size of the PD pattern database. Also, the FCM based PD pattern recognition scheme is very effective for clustering the defects of CRCT.