Application of Fuzzy C-Means Clustering Approach to Partial Discharge Pattern Recognition of Cast-Resin Current Transformers

摘要

This paper proposes a fuzzy c-means (FCM)clustering based recognition method to identify the defects of cast-resin current transformers (CRCT) arising from partial discharge (PD). To identify the defects, field data are collected in this paper using a PD detecting system for the CRCT. The proposed FCM clustering based classifier builds the cluster centers according to distributions of the extracted feature vectors. Effectiveness and feasibility of the proposed method have been verified through the encouraging results obtained using comprehensive experimental data.