A nonlinear system identification method based on local linear PLS method

Takashi Shikimori, Hideo Muroi, Shuichi Adachi

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

Abstract

Nonlinear system identification is one of the most important topics in system identification theory. In this paper, a new nonlinear system identification method using Partial Least-Squares (PLS) method is proposed, which is called a local linear PLS method because it is based on local models. The proposed method consists of two steps. First step is to identify local linear models by using the conventional Recursive Least-Squares (RLS) method. Second step is to identify a virtual system which describes a nonlinearity of the identified object, by PLS method. The effectiveness of the proposed method is shown through numerical simulations.

Original languageEnglish
Title of host publicationProceedings of the IASTED International Conference on Intelligent Systems and Control
Pages41-47
Number of pages7
DOIs
Publication statusPublished - 2011
Event13th IASTED International Conference on Intelligent Systems and Control, ISC 2011 - Cambridge, United Kingdom
Duration: 2011 Jul 112011 Jul 13

Other

Other13th IASTED International Conference on Intelligent Systems and Control, ISC 2011
CountryUnited Kingdom
CityCambridge
Period11/7/1111/7/13

Fingerprint

Nonlinear System Identification
Linear Least Squares
Partial Least Squares
Least Square Method
Nonlinear systems
Identification (control systems)
Recursive Method
System Identification
Linear Model
Computer simulation
Nonlinearity
Numerical Simulation

Keywords

  • Identification
  • Local linear model
  • Nonlinear systems
  • Partial least-squares (PLS) method
  • Recursive least-squares (RLS) method

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Control and Systems Engineering
  • Modelling and Simulation
  • Theoretical Computer Science

Cite this

Shikimori, T., Muroi, H., & Adachi, S. (2011). A nonlinear system identification method based on local linear PLS method. In Proceedings of the IASTED International Conference on Intelligent Systems and Control (pp. 41-47) https://doi.org/10.2316/P.2011.744-026

A nonlinear system identification method based on local linear PLS method. / Shikimori, Takashi; Muroi, Hideo; Adachi, Shuichi.

Proceedings of the IASTED International Conference on Intelligent Systems and Control. 2011. p. 41-47.

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

Shikimori, T, Muroi, H & Adachi, S 2011, A nonlinear system identification method based on local linear PLS method. in Proceedings of the IASTED International Conference on Intelligent Systems and Control. pp. 41-47, 13th IASTED International Conference on Intelligent Systems and Control, ISC 2011, Cambridge, United Kingdom, 11/7/11. https://doi.org/10.2316/P.2011.744-026
Shikimori T, Muroi H, Adachi S. A nonlinear system identification method based on local linear PLS method. In Proceedings of the IASTED International Conference on Intelligent Systems and Control. 2011. p. 41-47 https://doi.org/10.2316/P.2011.744-026
Shikimori, Takashi ; Muroi, Hideo ; Adachi, Shuichi. / A nonlinear system identification method based on local linear PLS method. Proceedings of the IASTED International Conference on Intelligent Systems and Control. 2011. pp. 41-47
@inproceedings{d4b23668bd1a4505a1b2cd940d22fbc1,
title = "A nonlinear system identification method based on local linear PLS method",
abstract = "Nonlinear system identification is one of the most important topics in system identification theory. In this paper, a new nonlinear system identification method using Partial Least-Squares (PLS) method is proposed, which is called a local linear PLS method because it is based on local models. The proposed method consists of two steps. First step is to identify local linear models by using the conventional Recursive Least-Squares (RLS) method. Second step is to identify a virtual system which describes a nonlinearity of the identified object, by PLS method. The effectiveness of the proposed method is shown through numerical simulations.",
keywords = "Identification, Local linear model, Nonlinear systems, Partial least-squares (PLS) method, Recursive least-squares (RLS) method",
author = "Takashi Shikimori and Hideo Muroi and Shuichi Adachi",
year = "2011",
doi = "10.2316/P.2011.744-026",
language = "English",
isbn = "9780889868892",
pages = "41--47",
booktitle = "Proceedings of the IASTED International Conference on Intelligent Systems and Control",

}

TY - GEN

T1 - A nonlinear system identification method based on local linear PLS method

AU - Shikimori, Takashi

AU - Muroi, Hideo

AU - Adachi, Shuichi

PY - 2011

Y1 - 2011

N2 - Nonlinear system identification is one of the most important topics in system identification theory. In this paper, a new nonlinear system identification method using Partial Least-Squares (PLS) method is proposed, which is called a local linear PLS method because it is based on local models. The proposed method consists of two steps. First step is to identify local linear models by using the conventional Recursive Least-Squares (RLS) method. Second step is to identify a virtual system which describes a nonlinearity of the identified object, by PLS method. The effectiveness of the proposed method is shown through numerical simulations.

AB - Nonlinear system identification is one of the most important topics in system identification theory. In this paper, a new nonlinear system identification method using Partial Least-Squares (PLS) method is proposed, which is called a local linear PLS method because it is based on local models. The proposed method consists of two steps. First step is to identify local linear models by using the conventional Recursive Least-Squares (RLS) method. Second step is to identify a virtual system which describes a nonlinearity of the identified object, by PLS method. The effectiveness of the proposed method is shown through numerical simulations.

KW - Identification

KW - Local linear model

KW - Nonlinear systems

KW - Partial least-squares (PLS) method

KW - Recursive least-squares (RLS) method

UR - http://www.scopus.com/inward/record.url?scp=80455149772&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80455149772&partnerID=8YFLogxK

U2 - 10.2316/P.2011.744-026

DO - 10.2316/P.2011.744-026

M3 - Conference contribution

AN - SCOPUS:80455149772

SN - 9780889868892

SP - 41

EP - 47

BT - Proceedings of the IASTED International Conference on Intelligent Systems and Control

ER -