Scientific Research and Essays

  • Abbreviation: Sci. Res. Essays
  • Language: English
  • ISSN: 1992-2248
  • DOI: 10.5897/SRE
  • Start Year: 2006
  • Published Articles: 2768

Full Length Research Paper

Recursive identification of Hammerstein ARMAX model with general discontinuous nonlinearity

  Sofien Rejeb1*, Raoudha Bel Hadj Ali2, Fayçal Ben Hmida1, Abdelkader Chaari1 and Hassani Messaoud2        
  1Unit C3S, High School of Sciences and Techniques of Tunis (ESSTT), 5 Av. Taha Hussein, BP 56, 1008 Tunis, Tunisia. 2Unit ATSI, National Engineering School of Monastir (ENIM), Rue Ibn EL Jazzar, 5019 Monastir, Tunisia.
Email: [email protected]

  •  Accepted: 22 August 2011
  •  Published: 31 October 2011

Abstract

 

This paper deals with the recursive parameter identification of the Hammerstein ARMAX (HARMAX) model with General model of discontinuous and asymmetric Nonlinearity (GNL) containing hysteresis, saturation, preload and dead-zone. This model contains different parameters the choice of which may generate several different nonlinearities. The linear and nonlinear structures are known, the linear block parameters are unknown. However the nonlinear block contains nine parameters only one of them is known. The estimation of the nonlinearity parameters, the internal variables relative to the nonlinearity as well as the linear model parameters are ensured by Recursive Extended Least Squares algorithm (RELS). The provided model is validated by simulation results.

 

Key words: Identification, discontinuous general nonlinearity, HARMAX model, parameter estimation, RELS algorithm.

Abbreviation

 

HARMAX, Hammerstein ARMAX; RELS, Recursive Extended Least Squares algorithm; GNL, General 
model of Nonlinearity; GNL, General Nonlinearity; ENL, Exact Nonlinearity; NMSE, Normalized Mean Squared Error.