Fault Diagnosis of Electric Power Systems Based on Fuzzy Petri Nets

TLDR

Fuzzy Petri Nets (FPNs) are introduced as a framework for fault diagnosis models. The study uses FPNs to construct fault diagnosis models that can accurately identify faults despite incomplete and uncertain alarm data. The authors build FPN-based fault diagnosis models, validate them through logical testifications, and demonstrate feasibility with simulation examples. Seven case studies confirm that the models correctly diagnose faulted elements, even with substantial incomplete and uncertain alarm information.

Abstract

In this paper, Fuzzy Petri Nets (FPN) is used as a modeling tool to build fault diagnosis models aimed to accurately diagnose faults when some incomplete and uncertain alarm information of protective relays and circuit breakers is detected. In order to understand the significance of fault diagnosis models more conveniently, the definition and structure of FPNs are necessarily introduced at first. Then, models of fault diagnosis based on FPN are built, and their corresponding logical testifications are carried out. Finally, the validity and feasibility of this method is illustrated by simulation examples. It is shown from seven cases that the faulted system elements can be diagnosed correctly by use of these models, and a satisfying result can also be achieved even in the situation with large amount of incomplete and uncertain alarm information.