AN EVOLUTIONARY ADAPTATION STRATEGY FOR DYNAMIC SYSTEMS IDENTIFICATION

An Evolutionary adaptation strategy for discrete time dynamic systems identification is proposed. This adaptation strategy allows achieving the selection of a suitable dynamic order inside a given model structure and appropriate model parameters using non conventional objective functions. The propos...

Full description

Saved in:
Bibliographic Details
Published in:WSEAS Transactions on Systems and Control 2007-03, Vol.2 (3), p.261-266
Main Authors: Altamiranda, Edmary, Calderon, Rodrigo, Morles, Eliezer Colina
Format: Article
Language:English
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An Evolutionary adaptation strategy for discrete time dynamic systems identification is proposed. This adaptation strategy allows achieving the selection of a suitable dynamic order inside a given model structure and appropriate model parameters using non conventional objective functions. The proposed algorithm incorporates parametric crossover and parametric mutation along a weighted gradient direction [1]. The chromosomes are represented in terms of a real values string which facilitates the calculation of model parameters for different system dynamic orders and input-output delay. The performance of the strategy is illustrated with computer simulations using ARX model structures.
ISSN:1991-8763