WCSE 2016
ISBN: 978-981-11-0008-6 DOI: 10.18178/wcse.2016.06.126

Dynamic Modeling on the MIMO System with Linear Programming Support Vector Regression and Its Application

Yanfang Yu, Li Jiang

Abstract— Few researches have been done for dynamic modeling on the multiple-input multiple-output system in actual industry so far. Therefore, this paper proposed a novel dynamic model to solve the multiple-input multiple-output problem. It respectively mapped the delayed output and the primal input into different nonlinear space, associated the kernel trick with the support vector regression technique to estimate the nonlinear dynamic problem. Simulation results of dynamic estimation of COOH content and degree of polymerization in the poly ethylene terephthalate production demonstrated that the present algorithm exactly approached the output of test set and reduced the error of test set respectively to the magnitude of 10-7 and 10-8.

Index Terms— linear programming support vector regression, dynamic modeling, kernel trick.

Yanfang Yu, Li Jiang
Shanghai Institute of Technology, CHINA

[Download]


Cite: Yanfang Yu, Li Jiang, "Dynamic Modeling on the MIMO System with Linear Programming Support Vector Regression and Its Application," Proceedings of 2016 6th International Workshop on Computer Science and Engineering, pp. 710-714, Tokyo, 17-19 June, 2016.