Local module identification in dynamic networks using regularized kernel-based methods

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

Abstract

In order to identify a specific system (module) of interest embedded in a dynamic network, one typically has to formulate a multi-input single-output (MISO) identification problem which requires to identify all modules in the MISO structure, and determine their model order. While the former task poses the problem of estimating a large number of parameters that are of no interest to the experimenter, the latter task may result computationally challenging in large-size networks. To avoid these issues and increase the accuracy of the identified module of interest, we use regularized kernel-based methods. Keeping a parametric model for the module of interest, we model the impulse response of the remaining modules in the MISO structure as zero mean Gaussian vectors with covariance matrix (kernel) given by the first-order stable spline kernel, accounting also for the noise model affecting the output of the target model. Using an Empirical Bayes (EB) approach, the target-module parameters are estimated by maximizing the marginal likelihood of the module output. The related optimization problem is solved using the Expectation-Maximization (EM) algorithm. Numerical experiments illustrate the potentials of the introduced method in comparison with the state-of-the-art techniques for local identification.
LanguageEnglish
Title of host publication2018 IEEE Conference on Decision and Control, CDC 2018
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages4713-4718
Number of pages6
ISBN (Electronic)978-1-5386-1395-5
ISBN (Print)978-1-5386-1396-2
DOIs
StatePublished - 18 Jan 2019
Event57th IEEE Conference on Decision and Control, CDC 2018 - Miami, United States
Duration: 17 Dec 201819 Dec 2018
Conference number: 57

Conference

Conference57th IEEE Conference on Decision and Control, CDC 2018
Abbreviated titleCDC 2018
CountryUnited States
CityMiami
Period17/12/1819/12/18

Cite this

Ramaswamy, K. R., Bottegal, G., & Van den Hof, P. M. J. (2019). Local module identification in dynamic networks using regularized kernel-based methods. In 2018 IEEE Conference on Decision and Control, CDC 2018 (pp. 4713-4718). [8619436] Piscataway: Institute of Electrical and Electronics Engineers. DOI: 10.1109/CDC.2018.8619436
Ramaswamy, Karthik R. ; Bottegal, Giulio ; Van den Hof, Paul M.J./ Local module identification in dynamic networks using regularized kernel-based methods. 2018 IEEE Conference on Decision and Control, CDC 2018. Piscataway : Institute of Electrical and Electronics Engineers, 2019. pp. 4713-4718
@inproceedings{e2b18b6042b746e1bd4b06d6ca50cff7,
title = "Local module identification in dynamic networks using regularized kernel-based methods",
abstract = "In order to identify a specific system (module) of interest embedded in a dynamic network, one typically has to formulate a multi-input single-output (MISO) identification problem which requires to identify all modules in the MISO structure, and determine their model order. While the former task poses the problem of estimating a large number of parameters that are of no interest to the experimenter, the latter task may result computationally challenging in large-size networks. To avoid these issues and increase the accuracy of the identified module of interest, we use regularized kernel-based methods. Keeping a parametric model for the module of interest, we model the impulse response of the remaining modules in the MISO structure as zero mean Gaussian vectors with covariance matrix (kernel) given by the first-order stable spline kernel, accounting also for the noise model affecting the output of the target model. Using an Empirical Bayes (EB) approach, the target-module parameters are estimated by maximizing the marginal likelihood of the module output. The related optimization problem is solved using the Expectation-Maximization (EM) algorithm. Numerical experiments illustrate the potentials of the introduced method in comparison with the state-of-the-art techniques for local identification.",
author = "Ramaswamy, {Karthik R.} and Giulio Bottegal and {Van den Hof}, {Paul M.J.}",
year = "2019",
month = "1",
day = "18",
doi = "10.1109/CDC.2018.8619436",
language = "English",
isbn = "978-1-5386-1396-2",
pages = "4713--4718",
booktitle = "2018 IEEE Conference on Decision and Control, CDC 2018",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Ramaswamy, KR, Bottegal, G & Van den Hof, PMJ 2019, Local module identification in dynamic networks using regularized kernel-based methods. in 2018 IEEE Conference on Decision and Control, CDC 2018., 8619436, Institute of Electrical and Electronics Engineers, Piscataway, pp. 4713-4718, 57th IEEE Conference on Decision and Control, CDC 2018, Miami, United States, 17/12/18. DOI: 10.1109/CDC.2018.8619436

Local module identification in dynamic networks using regularized kernel-based methods. / Ramaswamy, Karthik R.; Bottegal, Giulio; Van den Hof, Paul M.J.

2018 IEEE Conference on Decision and Control, CDC 2018. Piscataway : Institute of Electrical and Electronics Engineers, 2019. p. 4713-4718 8619436.

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

TY - GEN

T1 - Local module identification in dynamic networks using regularized kernel-based methods

AU - Ramaswamy,Karthik R.

AU - Bottegal,Giulio

AU - Van den Hof,Paul M.J.

PY - 2019/1/18

Y1 - 2019/1/18

N2 - In order to identify a specific system (module) of interest embedded in a dynamic network, one typically has to formulate a multi-input single-output (MISO) identification problem which requires to identify all modules in the MISO structure, and determine their model order. While the former task poses the problem of estimating a large number of parameters that are of no interest to the experimenter, the latter task may result computationally challenging in large-size networks. To avoid these issues and increase the accuracy of the identified module of interest, we use regularized kernel-based methods. Keeping a parametric model for the module of interest, we model the impulse response of the remaining modules in the MISO structure as zero mean Gaussian vectors with covariance matrix (kernel) given by the first-order stable spline kernel, accounting also for the noise model affecting the output of the target model. Using an Empirical Bayes (EB) approach, the target-module parameters are estimated by maximizing the marginal likelihood of the module output. The related optimization problem is solved using the Expectation-Maximization (EM) algorithm. Numerical experiments illustrate the potentials of the introduced method in comparison with the state-of-the-art techniques for local identification.

AB - In order to identify a specific system (module) of interest embedded in a dynamic network, one typically has to formulate a multi-input single-output (MISO) identification problem which requires to identify all modules in the MISO structure, and determine their model order. While the former task poses the problem of estimating a large number of parameters that are of no interest to the experimenter, the latter task may result computationally challenging in large-size networks. To avoid these issues and increase the accuracy of the identified module of interest, we use regularized kernel-based methods. Keeping a parametric model for the module of interest, we model the impulse response of the remaining modules in the MISO structure as zero mean Gaussian vectors with covariance matrix (kernel) given by the first-order stable spline kernel, accounting also for the noise model affecting the output of the target model. Using an Empirical Bayes (EB) approach, the target-module parameters are estimated by maximizing the marginal likelihood of the module output. The related optimization problem is solved using the Expectation-Maximization (EM) algorithm. Numerical experiments illustrate the potentials of the introduced method in comparison with the state-of-the-art techniques for local identification.

UR - http://www.scopus.com/inward/record.url?scp=85062175611&partnerID=8YFLogxK

U2 - 10.1109/CDC.2018.8619436

DO - 10.1109/CDC.2018.8619436

M3 - Conference contribution

SN - 978-1-5386-1396-2

SP - 4713

EP - 4718

BT - 2018 IEEE Conference on Decision and Control, CDC 2018

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Ramaswamy KR, Bottegal G, Van den Hof PMJ. Local module identification in dynamic networks using regularized kernel-based methods. In 2018 IEEE Conference on Decision and Control, CDC 2018. Piscataway: Institute of Electrical and Electronics Engineers. 2019. p. 4713-4718. 8619436. Available from, DOI: 10.1109/CDC.2018.8619436