Wiener system identification using polynomial non linear state space model

Author

Lamia, Sersour ; Djamah, Tounsia ; Hammar, Karima ; Bettayeb, Maamar

Author_Institution

Univ. M. Mammeri of Tizi Ouzou, Tizi Ouzou, Algeria

fYear

2015

fDate

25-27 May 2015

Firstpage

1

Lastpage

5

Abstract

This paper deals with identification of Wiener nonlinear systems. Such systems, consist of a linear dynamic block followed by a static non-linear subsystem. In this work, Polynomial Non Linear State Space(PNLSS) models are used to describe them. An output error identification method is performed, based on Levenberg-Marquardt algorithm; the parameters sensitivity functions are developed as a multivariable state space model. The method efficiency is investigated on numerical simulations in absence of noise and with noisy data for different signal to noise ratios.

Keywords

multivariable control systems; nonlinear control systems; polynomials; state-space methods; stochastic processes; Levenberg-Marquardt algorithm; PNLSS model; Wiener nonlinear system identification; linear dynamic block; multivariable state space model; output error identification method; parameter sensitivity functions; polynomial nonlinear state space model; static nonlinear subsystem; Computational modeling; Mathematical model; Numerical models; Polynomials; Sensitivity; Signal to noise ratio; Identification; Levenberg-Marquardt (LM) algorithm; Non-linear System; Polynomial Non Linear State Space(PNLSS) model; Wiener model;

fLanguage

English

Publisher

ieee

Conference_Titel

Control, Engineering & Information Technology (CEIT), 2015 3rd International Conference on

Conference_Location

Tlemcen

Type

conf

DOI

10.1109/CEIT.2015.7233069

Filename

7233069