Identification and frequency domain analysis of non-stationary and nonlinear systems using time-varying NARMAX models

Fei He, Hua Liang Wei, Stephen A. Billings

Research output: Contribution to journalArticle

12 Citations (Scopus)
3 Downloads (Pure)

Abstract

This paper introduces a new approach for nonlinear and non-stationary (time-varying) system identification based on time-varying nonlinear autoregressive moving average with exogenous variable (TV-NARMAX) models. The challenging model structure selection and parameter tracking problems are solved by combining a multiwavelet basis function expansion of the time-varying parameters with an orthogonal least squares algorithm. Numerical examples demonstrate that the proposed approach can track rapid time-varying effects in nonlinear systems more accurately than the standard recursive algorithms. Based on the identified time domain model, a new frequency domain analysis approach is introduced based on a time-varying generalised frequency response function (TV-GFRF) concept, which enables the analysis of nonlinear, non-stationary systems in the frequency domain. Features in the TV-GFRFs which depend on the TV-NARMAX model structure and time-varying parameters are investigated. It is shown that the high-dimension...
Original languageEnglish
Pages (from-to)2087-2100
Number of pages14
JournalInternational Journal of Systems Science
Volume46
Issue number11
Early online date19 Nov 2013
DOIs
Publication statusPublished - 18 Aug 2015
Externally publishedYes

Fingerprint

Frequency Domain Analysis
Frequency domain analysis
Model structures
Nonlinear systems
Time-varying
Nonlinear Systems
Time-varying Parameters
Autoregressive Moving Average
Time varying systems
Frequency response
Identification (control systems)
Orthogonal Least Squares
Frequency Response Function
Multiwavelet
Domain Model
Time-varying Systems
Least Square Algorithm
Recursive Algorithm
System Identification
Model

Bibliographical note

This is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Systems Science on 19/11/13, available online: http://www.tandfonline.com/10.1080/00207721.2013.860202

Keywords

  • generalised frequency response functions
  • nonlinear and non-stationary systems
  • system identification
  • timevarying systems
  • wavelet basis functions

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Theoretical Computer Science
  • Computer Science Applications

Cite this

Identification and frequency domain analysis of non-stationary and nonlinear systems using time-varying NARMAX models. / He, Fei; Wei, Hua Liang; Billings, Stephen A.

In: International Journal of Systems Science, Vol. 46, No. 11, 18.08.2015, p. 2087-2100.

Research output: Contribution to journalArticle

@article{ec06b3a7435847749543192ca4e239f9,
title = "Identification and frequency domain analysis of non-stationary and nonlinear systems using time-varying NARMAX models",
abstract = "This paper introduces a new approach for nonlinear and non-stationary (time-varying) system identification based on time-varying nonlinear autoregressive moving average with exogenous variable (TV-NARMAX) models. The challenging model structure selection and parameter tracking problems are solved by combining a multiwavelet basis function expansion of the time-varying parameters with an orthogonal least squares algorithm. Numerical examples demonstrate that the proposed approach can track rapid time-varying effects in nonlinear systems more accurately than the standard recursive algorithms. Based on the identified time domain model, a new frequency domain analysis approach is introduced based on a time-varying generalised frequency response function (TV-GFRF) concept, which enables the analysis of nonlinear, non-stationary systems in the frequency domain. Features in the TV-GFRFs which depend on the TV-NARMAX model structure and time-varying parameters are investigated. It is shown that the high-dimension...",
keywords = "generalised frequency response functions, nonlinear and non-stationary systems, system identification, timevarying systems, wavelet basis functions",
author = "Fei He and Wei, {Hua Liang} and Billings, {Stephen A.}",
note = "This is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Systems Science on 19/11/13, available online: http://www.tandfonline.com/10.1080/00207721.2013.860202",
year = "2015",
month = "8",
day = "18",
doi = "10.1080/00207721.2013.860202",
language = "English",
volume = "46",
pages = "2087--2100",
journal = "International Journal of Systems Science",
issn = "0020-7721",
publisher = "Taylor & Francis",
number = "11",

}

TY - JOUR

T1 - Identification and frequency domain analysis of non-stationary and nonlinear systems using time-varying NARMAX models

AU - He, Fei

AU - Wei, Hua Liang

AU - Billings, Stephen A.

N1 - This is an Accepted Manuscript of an article published by Taylor & Francis in International Journal of Systems Science on 19/11/13, available online: http://www.tandfonline.com/10.1080/00207721.2013.860202

PY - 2015/8/18

Y1 - 2015/8/18

N2 - This paper introduces a new approach for nonlinear and non-stationary (time-varying) system identification based on time-varying nonlinear autoregressive moving average with exogenous variable (TV-NARMAX) models. The challenging model structure selection and parameter tracking problems are solved by combining a multiwavelet basis function expansion of the time-varying parameters with an orthogonal least squares algorithm. Numerical examples demonstrate that the proposed approach can track rapid time-varying effects in nonlinear systems more accurately than the standard recursive algorithms. Based on the identified time domain model, a new frequency domain analysis approach is introduced based on a time-varying generalised frequency response function (TV-GFRF) concept, which enables the analysis of nonlinear, non-stationary systems in the frequency domain. Features in the TV-GFRFs which depend on the TV-NARMAX model structure and time-varying parameters are investigated. It is shown that the high-dimension...

AB - This paper introduces a new approach for nonlinear and non-stationary (time-varying) system identification based on time-varying nonlinear autoregressive moving average with exogenous variable (TV-NARMAX) models. The challenging model structure selection and parameter tracking problems are solved by combining a multiwavelet basis function expansion of the time-varying parameters with an orthogonal least squares algorithm. Numerical examples demonstrate that the proposed approach can track rapid time-varying effects in nonlinear systems more accurately than the standard recursive algorithms. Based on the identified time domain model, a new frequency domain analysis approach is introduced based on a time-varying generalised frequency response function (TV-GFRF) concept, which enables the analysis of nonlinear, non-stationary systems in the frequency domain. Features in the TV-GFRFs which depend on the TV-NARMAX model structure and time-varying parameters are investigated. It is shown that the high-dimension...

KW - generalised frequency response functions

KW - nonlinear and non-stationary systems

KW - system identification

KW - timevarying systems

KW - wavelet basis functions

UR - http://www.scopus.com/inward/record.url?scp=84928697291&partnerID=8YFLogxK

U2 - 10.1080/00207721.2013.860202

DO - 10.1080/00207721.2013.860202

M3 - Article

VL - 46

SP - 2087

EP - 2100

JO - International Journal of Systems Science

JF - International Journal of Systems Science

SN - 0020-7721

IS - 11

ER -