Premium
Hybrid identification of fuzzy rule‐based models
Author(s) -
Oh SungKwun,
Pedrycz Witold,
Park ByoungJun
Publication year - 2002
Publication title -
international journal of intelligent systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.291
H-Index - 87
eISSN - 1098-111X
pISSN - 0884-8173
DOI - 10.1002/int.1004
Subject(s) - adaptive neuro fuzzy inference system , computer science , fuzzy logic , defuzzification , neuro fuzzy , weighting , identification (biology) , fuzzy rule , fuzzy set operations , fuzzy control system , fuzzy classification , genetic algorithm , membership function , data mining , mathematical optimization , fuzzy number , artificial intelligence , algorithm , mathematics , machine learning , fuzzy set , medicine , botany , biology , radiology
Abstract In this study, we propose a hybrid identification algorithm for a class of fuzzy rule‐based systems. Therule‐based fuzzy modeling concerns structure optimization and parameter identification using the fuzzyinference methods and hybrid structure combined with two methods of optimization theories for nonlinear systems.Two types of inference methods of a fuzzy model concern a simplified and linear type of inference. The proposedhybrid optimal identification algorithm is carried out using a combination of genetic algorithms and an improvedcomplex method. The genetic algorithms determine initial parameters of the membership function of the premise partof the fuzzy rules. In the sequel, the improved complex method (being in essence a powerful auto‐tuningalgorithm) leads to fine‐tuning of the parameters of the respective membership functions. An aggregateperformance index with a weighting factor is proposed in order to achieve a balance between performance of thefuzzy model obtained for the training and testing data. Numerical examples are included to evaluate the performanceof the proposed model. They are also contrasted with the performance of the fuzzy models existing in theliterature. © 2002 John Wiley & Sons, Inc.