Study on method of modelling and controlling of magnetostrictive material

Author

An, Jinlong ; Yang, Qingxin ; Mazhengpin

Author_Institution

Key Lab. of Electromagn. Field & Electr. Apparautus Reliability, Hebei Univ. of Technol.

fYear

2006

fDate

Feb. 27 2006-March 3 2006

Firstpage

387

Lastpage

390

Abstract

Support vector machine is a learning technique based on the structural risk minimization principle, and it is also a kind of regression method with good generalization ability. This paper analyses the disadvantage of the nonlinear dynamical systems identification method based on neural networks, and presents a SVM method of modelling and controlling for magnetostrictive material. Simulation result indicates that this method has the better prediction precision than that of the approach based on the neural network. Therefore the present method can be used to the prediction control of magnetostrictive material

Keywords

magnetostrictive devices; minimisation; neurocontrollers; nonlinear dynamical systems; predictive control; regression analysis; support vector machines; magnetostrictive material; neural networks; nonlinear dynamical systems identification method; prediction control; regression method; structural risk minimization principle; support vector machine; Machine learning; Magnetic analysis; Magnetic materials; Magnetostriction; Neural networks; Nonlinear control systems; Nonlinear dynamical systems; Predictive models; Risk management; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Electromagnetic Compatibility, 2006. EMC-Zurich 2006. 17th International Zurich Symposium on

Conference_Location

Singapore

Print_ISBN

3-9522990-3-0

Type

conf

DOI

10.1109/EMCZUR.2006.214952

Filename

1629642