Convergence analysis of the alternating RELS algorithm for the identification of the reduced complexity Volterra model
by Imen Laamiri; Anis Khouaja; Hassani Messaoud
International Journal of Modelling, Identification and Control (IJMIC), Vol. 24, No. 4, 2015

Abstract: In this paper, we provide a sufficient convergence condition of the alternating recursive extended least squares (RELS) algorithm proposed for the identification of the reduced complexity Volterra model describing stochastic nonlinear systems. The reduced Volterra model used is the third order PARAFC-Volterra model provided using the parallel factor (PARAFAC) tensor decomposition of the Volterra kernels of order higher than two of the classical Volterra model. The recursive stochastic algorithm alternating recursive extended least squares (ARELS) consists of the execution in an alternating way of the classical RELS algorithm developed to identify the linear stochastic input-output models. The ARELS convergence was proved using the ordinary differential equation (ODE) method. It is noted that the ARELS algorithm convergence cannot be ensured when the disturbance acting on the system to be identified has specific features. The ARELS algorithm is tested using Monte Carlo simulation under the determined sufficient convergence condition. To raise the supremacy of the ARELS algorithm compared to the classical ARLS algorithm, we compare their performance on a nonlinear satellite channel and a benchmark system continuous stirred tank reactor (CSTR). Moreover, the efficiency of the proposed identification approach is proven on an experimental coupled tanks system (CTS).

Online publication date: Wed, 11-Nov-2015

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Modelling, Identification and Control (IJMIC):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com