Parameter Estimation for Hammerstein Nonlinear Controlled Auto-Regression Models

Author

Fan, Wei ; Ding, Feng ; Shi, Yang

Author_Institution

Jiangnan Univ., Wuxi

fYear

2007

fDate

18-21 Aug. 2007

Firstpage

1007

Lastpage

1012

Abstract

A recursive least-squares identification algorithm is developed for Hammerstein nonlinear models, which consist of memoryless nonlinear blocks followed by linear dynamical systems described by controlled auto-regression (CAR) models. Convergence analysis of the proposed algorithms indicates that the parameter estimation error consistently converges to zero under proper conditions. The simulation results show that the proposed algorithm is effective.

Keywords

autoregressive processes; convergence; least squares approximations; linear systems; nonlinear control systems; recursive estimation; time-varying systems; Hammerstein nonlinear controlled; auto-regression models; convergence analysis; linear dynamical systems; memoryless nonlinear blocks; parameter estimation error; recursive least-squares identification algorithm; Algorithm design and analysis; Automation; Control system synthesis; Convergence; Finite impulse response filter; Iterative algorithms; Least squares approximation; Logistics; Nonlinear control systems; Parameter estimation; Hammerstein models; Recursive identification; least squares; martingale convergence theorem; parameter estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

Automation and Logistics, 2007 IEEE International Conference on

Conference_Location

Jinan

Print_ISBN

978-1-4244-1531-1

Type

conf

DOI

10.1109/ICAL.2007.4338714

Filename

4338714