| 202D1 |
|---|
| Identification of Hammerstein Systems Using Automatic Choosing Function Model Designed by GA |
| AUTHORS |
| Tomohiro Hachino, Katsuhisa Deguchi , Hitoshi Takata (Kagoshima University) |
| ABSTRACT |
| Many practical systems have inherently nonlinear characteristics such as saturation, dead-zone, etc. Identification of nonlinear systems is of practical importance for precise analysis, prediction or control design. Hammerstein model is one of the block oriented models with a nonlinear static part followed by a linear dynamic part, and often used for nonlinear system identification. In this paper an identification method of Hammerstein systems is proposed by using an automatic choosing function (ACF) model and genetic algorithm(GA). The data region of the input signals is divided into some subdomains and unknown nonlinear static part to be estimated is approximately represented by a linear local equation on each subdomain. These local equations are united into a single one by the ACF smoothly. The connection coefficients of the ACF and the system parameters of the linear dynamic part are estimated by the linear least-squares method. The accuracy of this identification method depends sensitively on the ACF model structure such as the widths of the subdomains, the shape of the ACF. These elements are determined properly by using the GA, which is a probabilistic search procedure based on the mechanics of natural selection and natural genetics. Simulation results are shown to illustrate the propose method. |
Back to Program
Time Table
Top of This Site
Annual Conference 2002