GA optimization of generalized OBF TS fuzzy models with global and local estimation approaches

Anderson V. Medeiros*, Wagner C. Amaral, Ricardo J.G.B. Campello

*Kontaktforfatter

Publikation: Kapitel i bog/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

Abstract

OBF (Orthonormal Basis Function) Fuzzy models have shown to be a promising approach to the areas of nonlinear system identification and control since they exhibit several advantages over those dynamic model topologies usually adopted in the literature. A more general architecture, called Generalized OBF Takagi-Sugeno Fuzzy Model, was introduced in previous work and provided the mathematical interpretation that was missing to the former OBF fuzzy models. In spite of its clear mathematical meaning, however, the identification of this new generalized model is not a trivial task. This paper discusses the use of a genetic algorithm (GA) especially designed for this task, where a fitness function based on the Akaike information criterion plays a key role by considering both model accuracy and parsimony aspects. The hybridization of the GA with classical estimation algorithms is also investigated. Specifically, two different hybridization approaches (with global and local least squares) are evaluated in the modeling of a real nonlinear magnetic levitation system.

OriginalsprogEngelsk
Titel2006 IEEE International Conference on Fuzzy Systems
ForlagIEEE
Publikationsdato2006
Sider1835-1842
Artikelnummer1681955
ISBN (Trykt)0780394887, 9780780394889
DOI
StatusUdgivet - 2006
Udgivet eksterntJa
Begivenhed2006 IEEE International Conference on Fuzzy Systems - Vancouver, BC, Canada
Varighed: 16. jul. 200621. jul. 2006

Konference

Konference2006 IEEE International Conference on Fuzzy Systems
Land/OmrådeCanada
ByVancouver, BC
Periode16/07/200621/07/2006
NavnIEEE International Conference on Fuzzy Systems
ISSN1098-7584

Fingeraftryk

Dyk ned i forskningsemnerne om 'GA optimization of generalized OBF TS fuzzy models with global and local estimation approaches'. Sammen danner de et unikt fingeraftryk.

Citationsformater