Non-parametric identification of linear parameter-varying spatially-interconnected systems using an LS-SVM approach

Q. Liu, J. Mohammadpour, R. Toth, N. Meskin

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)

Abstract

This paper considers a general approach for the identification of partial differential equation-governed spatially-distributed systems. Spatial discretization virtually divides a system into spatially-interconnected subsystems, which allows to define the identification problem at the subsystem level. Here we focus on such a distributed identification of spatially-interconnected systems with temporal/spatial varying properties, whose dynamics can be captured by temporal/spatial linear parameter-varying (LPV) models. Inaccurate selection of the functional dependencies of the model parameters on scheduling variables may lead to bias in the identified models. Hence, we propose a non-parametric identification approach via a least-squares support vector machine (LS-SVM) - `non-parametric' estimation is in the sense that the model dependence on the scheduling variables is not explicitly parametrized. The performance of the proposed approach is evaluated on an Euler-Bernoulli beam with varying thickness.
LanguageEnglish
Title of host publicationProceedings of the American Control Conference (ACC), 6-8 July 2016, Boston, Massachusetts
PublisherAmerican Automatic Control Council (AACC)
Pages4592-4597
ISBN (Print)978-1-4673-8682-1
DOIs
StatePublished - 2016
Event2016 American Control Conference (ACC 2016), July 6-8, 2016, Boston, MA, USA - Boston Marriott Copley Place, Boston, MA, United States
Duration: 6 Jul 20168 Jul 2016
http://acc2016.a2c2.org/

Conference

Conference2016 American Control Conference (ACC 2016), July 6-8, 2016, Boston, MA, USA
Abbreviated titleACC 2016
CountryUnited States
CityBoston, MA
Period6/07/168/07/16
Internet address

Fingerprint

Support vector machines
Large scale systems
Identification (control systems)
Scheduling
Partial differential equations

Cite this

Liu, Q., Mohammadpour, J., Toth, R., & Meskin, N. (2016). Non-parametric identification of linear parameter-varying spatially-interconnected systems using an LS-SVM approach. In Proceedings of the American Control Conference (ACC), 6-8 July 2016, Boston, Massachusetts (pp. 4592-4597). American Automatic Control Council (AACC). DOI: 10.1109/ACC.2016.7526076
Liu, Q. ; Mohammadpour, J. ; Toth, R. ; Meskin, N./ Non-parametric identification of linear parameter-varying spatially-interconnected systems using an LS-SVM approach. Proceedings of the American Control Conference (ACC), 6-8 July 2016, Boston, Massachusetts. American Automatic Control Council (AACC), 2016. pp. 4592-4597
@inproceedings{14677a64ee9b4638a3917db8a0bfa27a,
title = "Non-parametric identification of linear parameter-varying spatially-interconnected systems using an LS-SVM approach",
abstract = "This paper considers a general approach for the identification of partial differential equation-governed spatially-distributed systems. Spatial discretization virtually divides a system into spatially-interconnected subsystems, which allows to define the identification problem at the subsystem level. Here we focus on such a distributed identification of spatially-interconnected systems with temporal/spatial varying properties, whose dynamics can be captured by temporal/spatial linear parameter-varying (LPV) models. Inaccurate selection of the functional dependencies of the model parameters on scheduling variables may lead to bias in the identified models. Hence, we propose a non-parametric identification approach via a least-squares support vector machine (LS-SVM) - `non-parametric' estimation is in the sense that the model dependence on the scheduling variables is not explicitly parametrized. The performance of the proposed approach is evaluated on an Euler-Bernoulli beam with varying thickness.",
author = "Q. Liu and J. Mohammadpour and R. Toth and N. Meskin",
year = "2016",
doi = "10.1109/ACC.2016.7526076",
language = "English",
isbn = "978-1-4673-8682-1",
pages = "4592--4597",
booktitle = "Proceedings of the American Control Conference (ACC), 6-8 July 2016, Boston, Massachusetts",
publisher = "American Automatic Control Council (AACC)",

}

Liu, Q, Mohammadpour, J, Toth, R & Meskin, N 2016, Non-parametric identification of linear parameter-varying spatially-interconnected systems using an LS-SVM approach. in Proceedings of the American Control Conference (ACC), 6-8 July 2016, Boston, Massachusetts. American Automatic Control Council (AACC), pp. 4592-4597, 2016 American Control Conference (ACC 2016), July 6-8, 2016, Boston, MA, USA, Boston, MA, United States, 6/07/16. DOI: 10.1109/ACC.2016.7526076

Non-parametric identification of linear parameter-varying spatially-interconnected systems using an LS-SVM approach. / Liu, Q.; Mohammadpour, J.; Toth, R.; Meskin, N.

Proceedings of the American Control Conference (ACC), 6-8 July 2016, Boston, Massachusetts. American Automatic Control Council (AACC), 2016. p. 4592-4597.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Non-parametric identification of linear parameter-varying spatially-interconnected systems using an LS-SVM approach

AU - Liu,Q.

AU - Mohammadpour,J.

AU - Toth,R.

AU - Meskin,N.

PY - 2016

Y1 - 2016

N2 - This paper considers a general approach for the identification of partial differential equation-governed spatially-distributed systems. Spatial discretization virtually divides a system into spatially-interconnected subsystems, which allows to define the identification problem at the subsystem level. Here we focus on such a distributed identification of spatially-interconnected systems with temporal/spatial varying properties, whose dynamics can be captured by temporal/spatial linear parameter-varying (LPV) models. Inaccurate selection of the functional dependencies of the model parameters on scheduling variables may lead to bias in the identified models. Hence, we propose a non-parametric identification approach via a least-squares support vector machine (LS-SVM) - `non-parametric' estimation is in the sense that the model dependence on the scheduling variables is not explicitly parametrized. The performance of the proposed approach is evaluated on an Euler-Bernoulli beam with varying thickness.

AB - This paper considers a general approach for the identification of partial differential equation-governed spatially-distributed systems. Spatial discretization virtually divides a system into spatially-interconnected subsystems, which allows to define the identification problem at the subsystem level. Here we focus on such a distributed identification of spatially-interconnected systems with temporal/spatial varying properties, whose dynamics can be captured by temporal/spatial linear parameter-varying (LPV) models. Inaccurate selection of the functional dependencies of the model parameters on scheduling variables may lead to bias in the identified models. Hence, we propose a non-parametric identification approach via a least-squares support vector machine (LS-SVM) - `non-parametric' estimation is in the sense that the model dependence on the scheduling variables is not explicitly parametrized. The performance of the proposed approach is evaluated on an Euler-Bernoulli beam with varying thickness.

U2 - 10.1109/ACC.2016.7526076

DO - 10.1109/ACC.2016.7526076

M3 - Conference contribution

SN - 978-1-4673-8682-1

SP - 4592

EP - 4597

BT - Proceedings of the American Control Conference (ACC), 6-8 July 2016, Boston, Massachusetts

PB - American Automatic Control Council (AACC)

ER -

Liu Q, Mohammadpour J, Toth R, Meskin N. Non-parametric identification of linear parameter-varying spatially-interconnected systems using an LS-SVM approach. In Proceedings of the American Control Conference (ACC), 6-8 July 2016, Boston, Massachusetts. American Automatic Control Council (AACC). 2016. p. 4592-4597. Available from, DOI: 10.1109/ACC.2016.7526076