Fault Diagnosis of Power Systems Using Intuitionistic Fuzzy Spiking Neural P Systems

In this paper, intuitionistic fuzzy spiking neural P (IFSNP) systems as a variant are proposed by integrating intuitionistic fuzzy logic into original spiking neural P systems. Compared with a common fuzzy set, intuitionistic fuzzy set can more finely describe the uncertainty due to its membership and non-membership degrees. Therefore, IFSNP systems are very suitable to deal with fault diagnosis of power systems, specially with incomplete and uncertain alarm messages. The fault modeling method and fuzzy reasoning algorithm based on IFSNP systems are discussed. Two examples are used to demonstrate the availability and effectiveness of IFSNP systems for fault diagnosis of power systems. Case studies involve single fault, complex fault, and multiple faults with protection device failures and incorrect tripping signals.

I. Introduction

The power system consists of many system elements, such as generators, transformers, bus bars and transmission lines, which are protected by protection systems comprised of protective relays (PRs), circuit breakers (CRs) and communication equipment. The supervisory control and data acquisition (SCADA) system is equipped together with electric power systems. Fault diagnosis of power systems is a process of discriminating the faulted system elements by tripping of protective relays and circuit breakers. When a fault event occurs, it can lead to a large amount of alarm messages in SCADA system. The alarm messages must be analyzed by dispatchers according to their operating experiences in order to identify the faults. However, the received data is often incomplete and tripping of protective relays and circuit breakers is sometime uncertain. Therefore, fault diagnosis is a difficult and complicated task since circuit breakers may fail to operate the multiple faults with the incomplete and uncertain alarm messages.