A novel fuzzy cause-and-effect-networks based methodology for a distribution system's fault diagnosis

Abstract

Distribution systems gained importance due to their proximity to huge numbers of consumers. Power distribution system operation is essential for consumers and equipment safety. In case of abnormal events, accurate and fast fault diagnosis (identifying fault's location and type) is a vital issue to retrieve a sound distribution system's operation. Thus, in this paper, a novel on-line distribution systems fault diagnosis methodology using fuzzy-cause-and-effect-networks (FCE-Nets) is presented. Both, a fault's location and type are identified using simple matrix operations, Fuzzy logic data base, and if-then-rules. Two case studies are carried out to evaluate the proposed methodology. The obtained results are compared with two different methods: Expert System (ES) with Artificial Neural Networks (ANNs) and cause-and-effect-network (CE-NETs) to evaluate the proposed method. The comparison outcomes are discussed and conclusions are reported.