Title: Least squares support kernel machines (LS-SKM) for identification

Authors: Mounira Tarhouni; Salah Zidi; Kaouther Laabidi; M. Ksouri-Lahmari

Addresses: Unit of Research Analysis and Control of Systems (ACS, ENIT), BP 37, Le Belvedere 1002, Tunis, Tunisia. ' LAGIS (USTL, Lille), Villeneuve d'Ascq 59650 Lille, France. ' Unit of Research Analysis and Control of Systems (ACS, ENIT), BP 37, Le Belvedere 1002, Tunis, Tunisia. ' Unit of Research Analysis and Control of Systems (ACS, ENIT), BP 37, Le Belvedere 1002, Tunis, Tunisia

Abstract: This paper presents a novel approach for non-linear systems identification called 'least squares support kernel machines (LS-SKM)'. Instead of using a least squares support vector machines (LS-SVM) with a single kernel function, the proposed approach combines several kernels in order to take advantage of their performances and also reflects the fact that practical learning problems often involve multiple, heterogeneous data sources. The idea is to divide the regressor vector in several regressor vectors, and, for each vector a kernel function is used. The choice of kernel function and the corresponding parameters is an important task which is related to the non-linear system degrees. A constrained particle swarm optimisation (CPSO) is used to give solution for the determination of optimised kernel parameters. Two examples are presented for qualitative comparison with the classical LS-SVM. The results reveal the accuracy and the robustness of the obtained model based on our proposed hybrid method.

Keywords: least squares; support kernel machines; LS-SKM; support vector machines; LS-SVM; SKM; SVM; nonlinear system identification; constrained PSO; particle swarm optimisation; CPSO.

DOI: 10.1504/IJMIC.2012.048641

International Journal of Modelling, Identification and Control, 2012 Vol.17 No.1, pp.68 - 77

Published online: 17 Dec 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article