Fault Identification using Combined Adaptive Neuro-Fuzzy Inference System and Gustafson–Kessel Algorithm

Journal of Engineering Research

Amalina Abdullah¹  Channarong Banmongkol² 

[1] Chulalongkorn University;The University of Tokyo

关键词: Power System Protection;    Fault Identification;    Adaptive Neuro-Fuzzy Inference System;    Gustafson–Kessel Algorithm;   

DOI  :  

学科分类：社会科学、人文和艺术（综合）

来源: Kuwait University * Academic Publication Council

【 摘 要 】

Issues on detecting the occurrence of a fault, justifying the type, and estimating the exact location of the fault should be resolved to eliminate faults promptly and restore power supply with minimum interruption. Conventional approaches have contributed in assisting power utility in overcoming these issues. However, these approaches rely on line parameters and involve a few complex mathematical equations. In this paper, a new method for fault identification pertinent to classification and location is proposed by utilizing the combined adaptive neuro-fuzzy inference system (ANFIS) and Gustafson–Kessel (GK) clustering algorithm. The effectiveness and practicability of this method is demonstrated by simulation result. This method uses the GK fuzzy clustering algorithm to decide on the premise configuration and its parameter, and identifies its succeding parameter using orthogonal least square. The proposed method is independent of line parameter information and obtains high accuracy on estimation of fault locations.

【 授权许可】

Unknown