Fuzzy Ontology Reasoning for Power Transformer Fault Diagnosis

Abstract

This paper presents a novel fuzzy ontology reasoner for power transformer fault diagnosis under a multi-agent framework. The developed ontology provides a comprehensive knowledge base as part of a multi-agent system to enable imprecision reasoning. It is the first time that a fuzzy ontology model is developed for accurate power transformer fault diagnosis. It aims to develop an improved ontology model for transformer fault diagnosis by applying the fuzzy ontology. The proposed technique deals with the imprecision situation using the fuzzy ontology, in order to build an ontology-based knowledge representation for accurate power transformer fault diagnosis. The proposed system is tested with actual transformer online data to demonstrate the functionality of the developed fuzzy ontology, which can identify the faults that are unidentifiable using a basic ontology model, and this can significantly improve the overall accuracy for transformer fault diagnosis under a multi-agent framework.