Online identification of nonlinear system using a new kernel algorithm

Author

Taouali, Okba

Author_Institution

Nat. Sch. of Eng. of Monastir, Univ. of Monastir, Monastir, Tunisia

fYear

2015

fDate

28-30 April 2015

Firstpage

480

Lastpage

485

Abstract

This paper proposes a new online kernel identification method of a nonlinear system in the Reproducing Kernel Hilbert Space. The proposed algorithm entitled online RKPLS-RN kernel method uses the Reduced Kernel Partial Least Square (RKPLS) technique in an offline phase to construct a RKHS model with reduced parameter number. Then the Regularized Network (RN) method is used on online phase to update the reduced parameters of the RKHS model. The considered measure of performance is the Normalized Means Square Error (NMSE). The proposed online kernel method is evaluated by handling the Cascades Tanks system.

Keywords

Hilbert spaces; identification; least squares approximations; mean square error methods; nonlinear systems; NMSE; RKPLS technique; RKPLS-RN kernel method; RN method; cascades tanks system; normalized means square error; online kernel identification method; online nonlinear system identification; parameter number; reduced kernel partial least square; regularized network; reproducing kernel Hilbert space; Control systems; Decision support systems; CSTR; Cascades Tanks; Kernel method; Learning machine; online identification;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems and Control (ICSC), 2015 4th International Conference on

Conference_Location

Sousse

Print_ISBN

978-1-4673-7108-7

Type

conf

DOI

10.1109/ICoSC.2015.7152789

Filename

7152789