Bayesian topology identification of linear dynamic networks

Shengling Shi; Giulio Bottegal; Paul M.J. Van den Hof

doi:10.23919/ECC.2019.8795766

Bayesian topology identification of linear dynamic networks

Shengling Shi, Giulio Bottegal, Paul M.J. Van den Hof

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

14 Citations (Scopus)

1 Downloads (Pure)

Abstract

In networks of dynamic systems, one challenge is to identify the interconnection structure on the basis of measured signals. Inspired by a Bayesian approach in [1], in this paper, we explore a Bayesian model selection method for identifying the connectivity of networks of transfer functions, without the need to estimate the dynamics. The algorithm employs a Bayesian measure and a forward-backward search algorithm. To obtain the Bayesian measure, the impulse responses of network modules are modeled as Gaussian processes and the hyperparameters are estimated by marginal likelihood maximization using the expectation-maximization algorithm. Numerical results demonstrate the effectiveness of this method.

Original language	English
Title of host publication	2019 18th European Control Conference, ECC 2019
Place of Publication	Piscataway
Publisher	Institute of Electrical and Electronics Engineers
Pages	2814-2819
Number of pages	6
ISBN (Electronic)	978-3-907144-00-8
DOIs	https://doi.org/10.23919/ECC.2019.8795766
Publication status	Published - 1 Jun 2019
Event	18th European Control Conference, ECC 2019 - Naples, Italy, Naples, Italy Duration: 25 Jun 2019 → 28 Jun 2019 Conference number: 18 https://www.ifac-control.org/events/european-control-conference-in-cooperation-with-ifac-ecc-2019

Conference

Conference	18th European Control Conference, ECC 2019
Abbreviated title	ECC 2019
Country/Territory	Italy
City	Naples
Period	25/06/19 → 28/06/19
Other	18th European Control Conference (ECC 2019) (in cooperation with IFAC)
Internet address	https://www.ifac-control.org/events/european-control-conference-in-cooperation-with-ifac-ecc-2019

Access to Document

10.23919/ECC.2019.8795766

Cite this

@inproceedings{8e5a10e3e42d495ea533ea6c78be2247,

title = "Bayesian topology identification of linear dynamic networks",

abstract = "In networks of dynamic systems, one challenge is to identify the interconnection structure on the basis of measured signals. Inspired by a Bayesian approach in [1], in this paper, we explore a Bayesian model selection method for identifying the connectivity of networks of transfer functions, without the need to estimate the dynamics. The algorithm employs a Bayesian measure and a forward-backward search algorithm. To obtain the Bayesian measure, the impulse responses of network modules are modeled as Gaussian processes and the hyperparameters are estimated by marginal likelihood maximization using the expectation-maximization algorithm. Numerical results demonstrate the effectiveness of this method.",

author = "Shengling Shi and Giulio Bottegal and {Van den Hof}, {Paul M.J.}",

year = "2019",

month = jun,

day = "1",

doi = "10.23919/ECC.2019.8795766",

language = "English",

pages = "2814--2819",

booktitle = "2019 18th European Control Conference, ECC 2019",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "18th European Control Conference, ECC 2019, ECC 2019 ; Conference date: 25-06-2019 Through 28-06-2019",

url = "https://www.ifac-control.org/events/european-control-conference-in-cooperation-with-ifac-ecc-2019",

}

Bayesian topology identification of linear dynamic networks. / Shi, Shengling; Bottegal, Giulio; Van den Hof, Paul M.J.
2019 18th European Control Conference, ECC 2019. Piscataway: Institute of Electrical and Electronics Engineers, 2019. p. 2814-2819 8795766.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Bayesian topology identification of linear dynamic networks

AU - Shi, Shengling

AU - Bottegal, Giulio

AU - Van den Hof, Paul M.J.

N1 - Conference code: 18

PY - 2019/6/1

Y1 - 2019/6/1

N2 - In networks of dynamic systems, one challenge is to identify the interconnection structure on the basis of measured signals. Inspired by a Bayesian approach in [1], in this paper, we explore a Bayesian model selection method for identifying the connectivity of networks of transfer functions, without the need to estimate the dynamics. The algorithm employs a Bayesian measure and a forward-backward search algorithm. To obtain the Bayesian measure, the impulse responses of network modules are modeled as Gaussian processes and the hyperparameters are estimated by marginal likelihood maximization using the expectation-maximization algorithm. Numerical results demonstrate the effectiveness of this method.

AB - In networks of dynamic systems, one challenge is to identify the interconnection structure on the basis of measured signals. Inspired by a Bayesian approach in [1], in this paper, we explore a Bayesian model selection method for identifying the connectivity of networks of transfer functions, without the need to estimate the dynamics. The algorithm employs a Bayesian measure and a forward-backward search algorithm. To obtain the Bayesian measure, the impulse responses of network modules are modeled as Gaussian processes and the hyperparameters are estimated by marginal likelihood maximization using the expectation-maximization algorithm. Numerical results demonstrate the effectiveness of this method.

UR - http://www.scopus.com/inward/record.url?scp=85071583759&partnerID=8YFLogxK

U2 - 10.23919/ECC.2019.8795766

DO - 10.23919/ECC.2019.8795766

M3 - Conference contribution

AN - SCOPUS:85071583759

SP - 2814

EP - 2819

BT - 2019 18th European Control Conference, ECC 2019

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

T2 - 18th European Control Conference, ECC 2019

Y2 - 25 June 2019 through 28 June 2019

ER -

Bayesian topology identification of linear dynamic networks

Abstract

Conference

Access to Document

Other files and links

Fingerprint