GA optimization of ladder-structured GOBF models

Abstract

A new technique for systems identification using ladder-structured generalized orthonormal basis function model is presented. In this approach the model poles and the number of functions are optimized using a genetic algorithm. A fitness function based on the Akaike information criterion considering model accuracy and model parsimony provides optimal number of functions and poles of the system model. Simulated and a real examples illustrate the performance of the proposed technique.