Publication detail

Genetic Algorithm Utilization in Fuzzy Regression Modelling

POKORNÝ, M. ŽELASKO, P. ROUPEC, J.

Czech title

Použití genetického algoritmu ve fuzzy regresním modelování

English title

Genetic Algorithm Utilization in Fuzzy Regression Modelling

Type

conference paper

Language

en

Original abstract

This paper introduces a soft-computing oriented approach to Takagi-Sugeno fuzzy modelling using the evolutionary principles. The presented algorithm allows determination of relevant input variables of fuzzy model from their potential candidates. Genetic algorithms are applied to optimize fuzzy input variables space through genetic fuzzy clustering procedure and to identify the fuzzy model. Some advanced procedures e.g. individuals lifetime limitation and a shade zone of genes are used. To clarify the advantages of the proposed approaches the numerical example of modellin of fuzzy non-linear system is presented.

Czech abstract

Referát uvádí přístup Takagi-Sugenova soft-computingového fuzzy modelování s použitím evolučních přístupů. Pro optimalizaci diverzifikace vstupního fuzzy prostoru a identifikaci modelu je použita fuzzy-genetická shlukovací procedura. Použitý genetický algoritmus využívá některé moderní operátory jako limitaci doby životnosti jedince v populaci nebo bitové redundance v chromozomech. Uvedený algoritmus umožňuje rovněž stanovit redundantní vstupní proměnné fuzzy modelu. Pro ověření kvality funkce navrženého algoritmu je uveden numerický příklad modelování nelineární soustavy.

English abstract

This paper introduces a soft-computing oriented approach to Takagi-Sugeno fuzzy modelling using the evolutionary principles. The presented algorithm allows determination of relevant input variables of fuzzy model from their potential candidates. Genetic algorithms are applied to optimize fuzzy input variables space through genetic fuzzy clustering procedure and to identify the fuzzy model. Some advanced procedures e.g. individuals lifetime limitation and a shade zone of genes are used. To clarify the advantages of the proposed approaches the numerical example of modellin of fuzzy non-linear system is presented.

Keywords in Czech

Genetický algoritmnus; identifikace fuzzy modelu

Keywords in English

Genetic algorithm;fuzzy model identification

Released

29.08.2004

Location

Awaji, Japan

Book

Proceedings of Taiwan-Japan Symposium 2004 On Fuzzy Systems & Innovational Computing

Edition number

1

Pages count

8

BIBTEX


@inproceedings{BUT20755,
  author="Miroslav {Pokorný} and Petr {Želasko} and Jan {Roupec},
  title="Genetic Algorithm Utilization in Fuzzy Regression Modelling",
  booktitle="Proceedings of Taiwan-Japan Symposium 2004 On Fuzzy Systems & Innovational Computing",
  year="2004",
  month="August",
  address="Awaji, Japan"
}