2-ℒ nonlinear system identification via recurrent neural networks

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

This paper proposes an ℒ2-ℒ identification scheme as a new robust identification method for nonlinear systems via recurrent neural networks. Based on linear matrix inequality (LMI) formulation, for the first time, the ℒ2-ℒ learning algorithm is presented to reduce the effect of disturbance to an ℒ2-ℒ induced norm constraint. New stability results, such as boundedness, input-to-state stability (ISS), and convergence, are established in some senses. It is shown that the design of the ℒ2-ℒ identification method can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed identification scheme.

Original languageEnglish
Pages (from-to)543-552
Number of pages10
JournalNonlinear Dynamics
Volume62
Issue number3
DOIs
Publication statusPublished - 2010 Nov 1

Fingerprint

Nonlinear System Identification
Identification Scheme
Recurrent neural networks
Recurrent Neural Networks
Nonlinear systems
Identification (control systems)
Stability and Convergence
Linear matrix inequalities
Learning algorithms
Matrix Inequality
Linear Inequalities
Learning Algorithm
Boundedness
Nonlinear Systems
Disturbance
Norm
Numerical Examples
Formulation
Demonstrate
Design

Keywords

  • ℒ-ℒ identification
  • Input-to-state stability (ISS)
  • Linear matrix inequality (LMI)
  • Recurrent neural networks
  • Weight learning law

ASJC Scopus subject areas

  • Applied Mathematics
  • Mechanical Engineering
  • Aerospace Engineering
  • Ocean Engineering
  • Electrical and Electronic Engineering
  • Control and Systems Engineering

Cite this

2-ℒ nonlinear system identification via recurrent neural networks. / Ahn, Choon Ki.

In: Nonlinear Dynamics, Vol. 62, No. 3, 01.11.2010, p. 543-552.

Research output: Contribution to journalArticle

@article{7482a9487c904685960ea62e26e76f6b,
title = "ℒ2-ℒ∞ nonlinear system identification via recurrent neural networks",
abstract = "This paper proposes an ℒ2-ℒ∞ identification scheme as a new robust identification method for nonlinear systems via recurrent neural networks. Based on linear matrix inequality (LMI) formulation, for the first time, the ℒ2-ℒ∞ learning algorithm is presented to reduce the effect of disturbance to an ℒ2-ℒ∞ induced norm constraint. New stability results, such as boundedness, input-to-state stability (ISS), and convergence, are established in some senses. It is shown that the design of the ℒ2-ℒ∞ identification method can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed identification scheme.",
keywords = "ℒ-ℒ identification, Input-to-state stability (ISS), Linear matrix inequality (LMI), Recurrent neural networks, Weight learning law",
author = "Ahn, {Choon Ki}",
year = "2010",
month = "11",
day = "1",
doi = "10.1007/s11071-010-9741-3",
language = "English",
volume = "62",
pages = "543--552",
journal = "Nonlinear Dynamics",
issn = "0924-090X",
publisher = "Springer Netherlands",
number = "3",

}

TY - JOUR

T1 - ℒ2-ℒ∞ nonlinear system identification via recurrent neural networks

AU - Ahn, Choon Ki

PY - 2010/11/1

Y1 - 2010/11/1

N2 - This paper proposes an ℒ2-ℒ∞ identification scheme as a new robust identification method for nonlinear systems via recurrent neural networks. Based on linear matrix inequality (LMI) formulation, for the first time, the ℒ2-ℒ∞ learning algorithm is presented to reduce the effect of disturbance to an ℒ2-ℒ∞ induced norm constraint. New stability results, such as boundedness, input-to-state stability (ISS), and convergence, are established in some senses. It is shown that the design of the ℒ2-ℒ∞ identification method can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed identification scheme.

AB - This paper proposes an ℒ2-ℒ∞ identification scheme as a new robust identification method for nonlinear systems via recurrent neural networks. Based on linear matrix inequality (LMI) formulation, for the first time, the ℒ2-ℒ∞ learning algorithm is presented to reduce the effect of disturbance to an ℒ2-ℒ∞ induced norm constraint. New stability results, such as boundedness, input-to-state stability (ISS), and convergence, are established in some senses. It is shown that the design of the ℒ2-ℒ∞ identification method can be achieved by solving LMIs, which can be easily facilitated by using some standard numerical packages. A numerical example is presented to demonstrate the validity of the proposed identification scheme.

KW - ℒ-ℒ identification

KW - Input-to-state stability (ISS)

KW - Linear matrix inequality (LMI)

KW - Recurrent neural networks

KW - Weight learning law

UR - http://www.scopus.com/inward/record.url?scp=78049475423&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=78049475423&partnerID=8YFLogxK

U2 - 10.1007/s11071-010-9741-3

DO - 10.1007/s11071-010-9741-3

M3 - Article

AN - SCOPUS:78049475423

VL - 62

SP - 543

EP - 552

JO - Nonlinear Dynamics

JF - Nonlinear Dynamics

SN - 0924-090X

IS - 3

ER -