Model reduction for spatio-temporal systems based on KPCA and LS-SVRM

Author

Ai Ling ; Zhu Yi ; San Ye

Author_Institution

Control & Simulation Center, Harbin Inst. of Technol., Harbin, China

fYear

2013

fDate

25-27 May 2013

Firstpage

4618

Lastpage

4622

Abstract

A novel model reduction technology based on Kernel Principal Component Analysis (KPCA) and Least squares support vector regression machine (LS-SVRM) for spatio-temporal systems. The process of the method is divided into two stages: Firstly, Kernel PCA is used for feature extraction and the necessary instructions for kernel parameter selected is given; Subsequently the temporal dynamics is constructed by LS-SVRM approach; In final, the accuracy and efficiency of the proposed method are verified and contrast to KL-LS-SVRM through the case of a tubular reactor.

Keywords

principal component analysis; reduced order systems; regression analysis; support vector machines; KPCA; LS-SVRM approach; feature extraction; kernel parameter; kernel principal component analysis; least squares support vector regression machine; novel model reduction technology; spatio-temporal systems; temporal dynamics; tubular reactor; Distributed parameter systems; Inductors; Kernel; Mathematical model; Principal component analysis; Process control; Reduced order systems; KPCA; LS-SVRM; Model Reduction; Spatio-temporal Systems; tubular reactor;

fLanguage

English

Publisher

ieee

Conference_Titel

Control and Decision Conference (CCDC), 2013 25th Chinese

Conference_Location

Guiyang

Print_ISBN

978-1-4673-5533-9

Type

conf

DOI

10.1109/CCDC.2013.6561769

Filename

6561769