T-S fuzzy modelling using advanced genetic algorithms

  • Authors:

  • Pavel Šišpera;Miroslav Pokorný;Jan Roupec

  • Affiliations:

  • Department of Measurement and Control, Faculty of Electrical Engineering and Computer Science, VSB Technical University Ostrava, Ostrava, Czech Republic;Department of Measurement and Control, Faculty of Electrical Engineering and Computer Science, VSB Technical University Ostrava, Ostrava, Czech Republic;Department of Automatic Control and Computer Science, Faculty of Mechanical Engineering, Technical University of Brno, Brno, Czech Republic

  • Venue:
  • ACMOS'05 Proceedings of the 7th WSEAS international conference on Automatic control, modeling and simulation
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper introduces a soft-computing oriented approach to Takagi-Sugeno fuzzy modelling using the evolutionary principles. 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 redundant genes application are used. The presented algorithm allows also the determination of the relevant inputs variables of fuzzy model from theirs potential candidates. To clarify the advantages of the proposed approaches the numerical example of modelling of fuzzy non-linear system is also introduced.