A neural network approach to improving identification of nonlinear polynomial models

Takayuki Furuzuki, Yinshi Li, Kotaro Hirasawa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

This paper presents a neural network approach to improving the identification of nonlinear polynomial model. The idea is to realize the identification in two steps. In the first step, a quasi-ARX neural network is first used to approximate the system under study; then a reasonable number of important monomial terms are selected as candidates, by introducing an importance index based on a Taylor expansion of the identified quasi-ARX neural network; In the second step, Genetic algorithm (GA) is applied to the selected important terms to further determine a set of significant terms to include in the polynomial model. In this way, the whole identification algorithm is implemented very efficiently. Numerical simulations are carried out to demonstrate the effectiveness of the proposed two-step identification method.

Original languageEnglish
Title of host publicationProceedings of the SICE Annual Conference
Pages1662-1667
Number of pages6
Publication statusPublished - 2005
EventSICE Annual Conference 2005 - Okayama
Duration: 2005 Aug 82005 Aug 10

Other

OtherSICE Annual Conference 2005
CityOkayama
Period05/8/805/8/10

Fingerprint

Neural networks
Genetic algorithms
Computer simulation
Statistical Models

Keywords

  • Genetic algorithm
  • Identification
  • Neural network
  • Nonlinear polynomial model
  • Poly-nomial expansion

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Furuzuki, T., Li, Y., & Hirasawa, K. (2005). A neural network approach to improving identification of nonlinear polynomial models. In Proceedings of the SICE Annual Conference (pp. 1662-1667)

A neural network approach to improving identification of nonlinear polynomial models. / Furuzuki, Takayuki; Li, Yinshi; Hirasawa, Kotaro.

Proceedings of the SICE Annual Conference. 2005. p. 1662-1667.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Furuzuki, T, Li, Y & Hirasawa, K 2005, A neural network approach to improving identification of nonlinear polynomial models. in Proceedings of the SICE Annual Conference. pp. 1662-1667, SICE Annual Conference 2005, Okayama, 05/8/8.
Furuzuki T, Li Y, Hirasawa K. A neural network approach to improving identification of nonlinear polynomial models. In Proceedings of the SICE Annual Conference. 2005. p. 1662-1667
Furuzuki, Takayuki ; Li, Yinshi ; Hirasawa, Kotaro. / A neural network approach to improving identification of nonlinear polynomial models. Proceedings of the SICE Annual Conference. 2005. pp. 1662-1667
@inproceedings{9f55e4bc44fc4a748fc2992db4ea5d36,
title = "A neural network approach to improving identification of nonlinear polynomial models",
abstract = "This paper presents a neural network approach to improving the identification of nonlinear polynomial model. The idea is to realize the identification in two steps. In the first step, a quasi-ARX neural network is first used to approximate the system under study; then a reasonable number of important monomial terms are selected as candidates, by introducing an importance index based on a Taylor expansion of the identified quasi-ARX neural network; In the second step, Genetic algorithm (GA) is applied to the selected important terms to further determine a set of significant terms to include in the polynomial model. In this way, the whole identification algorithm is implemented very efficiently. Numerical simulations are carried out to demonstrate the effectiveness of the proposed two-step identification method.",
keywords = "Genetic algorithm, Identification, Neural network, Nonlinear polynomial model, Poly-nomial expansion",
author = "Takayuki Furuzuki and Yinshi Li and Kotaro Hirasawa",
year = "2005",
language = "English",
pages = "1662--1667",
booktitle = "Proceedings of the SICE Annual Conference",

}

TY - GEN

T1 - A neural network approach to improving identification of nonlinear polynomial models

AU - Furuzuki, Takayuki

AU - Li, Yinshi

AU - Hirasawa, Kotaro

PY - 2005

Y1 - 2005

N2 - This paper presents a neural network approach to improving the identification of nonlinear polynomial model. The idea is to realize the identification in two steps. In the first step, a quasi-ARX neural network is first used to approximate the system under study; then a reasonable number of important monomial terms are selected as candidates, by introducing an importance index based on a Taylor expansion of the identified quasi-ARX neural network; In the second step, Genetic algorithm (GA) is applied to the selected important terms to further determine a set of significant terms to include in the polynomial model. In this way, the whole identification algorithm is implemented very efficiently. Numerical simulations are carried out to demonstrate the effectiveness of the proposed two-step identification method.

AB - This paper presents a neural network approach to improving the identification of nonlinear polynomial model. The idea is to realize the identification in two steps. In the first step, a quasi-ARX neural network is first used to approximate the system under study; then a reasonable number of important monomial terms are selected as candidates, by introducing an importance index based on a Taylor expansion of the identified quasi-ARX neural network; In the second step, Genetic algorithm (GA) is applied to the selected important terms to further determine a set of significant terms to include in the polynomial model. In this way, the whole identification algorithm is implemented very efficiently. Numerical simulations are carried out to demonstrate the effectiveness of the proposed two-step identification method.

KW - Genetic algorithm

KW - Identification

KW - Neural network

KW - Nonlinear polynomial model

KW - Poly-nomial expansion

UR - http://www.scopus.com/inward/record.url?scp=33645314087&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33645314087&partnerID=8YFLogxK

M3 - Conference contribution

SP - 1662

EP - 1667

BT - Proceedings of the SICE Annual Conference

ER -