Subspace identification with moment matching

Research output: Contribution to journalArticle

Abstract

We propose a deterministic identification method that involves a priori information characterized as moments of a transfer function. The moments are expressed in terms of the solution to a Sylvester matrix equation. The Sylvester equation is incorporated with a conventional subspace identification method, and a problem for moment-constrained subspace identification is formulated. Since the identification problem is in a class of nonlinear optimization problems, it cannot be efficiently solved in numerical computation. Application of a change-of-variable technique reduces the problem to least squares optimization, and the solution provides a state-space model that involves the prespecified moments. Finally, the effectiveness of the proposed method is illustrated in numerical simulations.

LanguageEnglish
Pages22-32
Number of pages11
JournalAutomatica
Volume99
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Transfer functions
Computer simulation

Keywords

  • Frequency-response characteristics
  • Least-squares method
  • Subspace method
  • System identification

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Cite this

Subspace identification with moment matching. / Inoue, Masaki.

In: Automatica, Vol. 99, 01.01.2019, p. 22-32.

Research output: Contribution to journalArticle

@article{d206e810030442f8afa5e32a2d560a5d,
title = "Subspace identification with moment matching",
abstract = "We propose a deterministic identification method that involves a priori information characterized as moments of a transfer function. The moments are expressed in terms of the solution to a Sylvester matrix equation. The Sylvester equation is incorporated with a conventional subspace identification method, and a problem for moment-constrained subspace identification is formulated. Since the identification problem is in a class of nonlinear optimization problems, it cannot be efficiently solved in numerical computation. Application of a change-of-variable technique reduces the problem to least squares optimization, and the solution provides a state-space model that involves the prespecified moments. Finally, the effectiveness of the proposed method is illustrated in numerical simulations.",
keywords = "Frequency-response characteristics, Least-squares method, Subspace method, System identification",
author = "Masaki Inoue",
year = "2019",
month = "1",
day = "1",
doi = "10.1016/j.automatica.2018.10.002",
language = "English",
volume = "99",
pages = "22--32",
journal = "Automatica",
issn = "0005-1098",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Subspace identification with moment matching

AU - Inoue, Masaki

PY - 2019/1/1

Y1 - 2019/1/1

N2 - We propose a deterministic identification method that involves a priori information characterized as moments of a transfer function. The moments are expressed in terms of the solution to a Sylvester matrix equation. The Sylvester equation is incorporated with a conventional subspace identification method, and a problem for moment-constrained subspace identification is formulated. Since the identification problem is in a class of nonlinear optimization problems, it cannot be efficiently solved in numerical computation. Application of a change-of-variable technique reduces the problem to least squares optimization, and the solution provides a state-space model that involves the prespecified moments. Finally, the effectiveness of the proposed method is illustrated in numerical simulations.

AB - We propose a deterministic identification method that involves a priori information characterized as moments of a transfer function. The moments are expressed in terms of the solution to a Sylvester matrix equation. The Sylvester equation is incorporated with a conventional subspace identification method, and a problem for moment-constrained subspace identification is formulated. Since the identification problem is in a class of nonlinear optimization problems, it cannot be efficiently solved in numerical computation. Application of a change-of-variable technique reduces the problem to least squares optimization, and the solution provides a state-space model that involves the prespecified moments. Finally, the effectiveness of the proposed method is illustrated in numerical simulations.

KW - Frequency-response characteristics

KW - Least-squares method

KW - Subspace method

KW - System identification

UR - http://www.scopus.com/inward/record.url?scp=85055736596&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85055736596&partnerID=8YFLogxK

U2 - 10.1016/j.automatica.2018.10.002

DO - 10.1016/j.automatica.2018.10.002

M3 - Article

VL - 99

SP - 22

EP - 32

JO - Automatica

T2 - Automatica

JF - Automatica

SN - 0005-1098

ER -