Takagi–Sugeno fuzzy based power system fault section diagnosis models via genetic learning adaptive GSK algorithm

Author's Department

Mathematics & Actuarial Science Department

Find in your Library

https://doi.org/10.1016/j.knosys.2022.109773

All Authors

Changsong Li, Guojiang Xiong, Xiaofan Fu, Ali Wagdy Mohamed, Xufeng Yuan, Mohammed Azmi Al-Betar, Ponnuthurai Nagaratnam Suganthan

Document Type

Research Article

Publication Title

Knowledge-Based Systems

Publication Date

Fall 11-1-2022

doi

10.1016/j.knosys.2022.109773

Abstract

To effectively deal with the operating uncertainties of protective relays and circuit breakers existing in the power system faults, an improved fault section diagnosis (FSD) method is proposed by using Takagi–Sugeno fuzzy neural networks (T–S FNN). In this method an optimal T–S FNN-based diagnosis model is built with the idea of distributed parallel processing for each section instead of the whole power system. To obtain accurate T–S FNN-based diagnosis models, a genetic learning adaptive gaining-sharing knowledge-based algorithm (GLAGSK) is designed to optimize their structure parameters and consequent parameters. GLAGSK combines an adaptive knowledge ratio and a genetic learning strategy to balance population diversity and convergence speed to boost the optimization ability. After a fault occurs, selective optimal T–S FNN-based diagnosis models are triggered according to the alarm information. They work in parallel to improve the fault diagnosis efficiency. Simulation results of three test systems including an actual fault event show that, compared with other peer algorithms, GLAGSK can obtain more accurate T–S FNN-based diagnosis models with faster global convergence. Besides, compared with the BP and RBF neural networks and other FSD methods, the proposed FSD method based on optimal T–S FNN-based diagnosis models can diagnose different complex faults successfully with higher fault credibility.

First Page

1

Last Page

14

This document is currently not available here.

Share

COinS