Hammerstein model identification using genetic programming

Abstract

In this paper, we propose a Hammerstein model identification method using genetic programming. A Hammerstein model is composed of a nonlinear static block in series with a linear dynamic system. The aim of system identification is to give the optimal mathematical model both nonlinearity and linear dynamic system in an appropriate sense. Genetic programming is used to determine the structure of the nonlinear static block. Each individual in genetic programming represents a nonlinear function. The unknown parameters including those of the linear dynamic system model are estimated by the least square method. The fitness is evaluated as AIC (Akaike information criterion). AIC is calculated with the number of nodes in the genetic programming tree, the order of linear dynamic model and the accuracy of the model. The results of numerical studies indicate the usefulness of the proposed approach to Hammerstein model identification.