Robust Fault Detection of Linear Systems using a Computationally Efficient Set-membership Method

Seyed Mojtaba Tabatabaeipour; Thomas Bak

doi:10.1109/CCA.2014.6981372

Robust Fault Detection of Linear Systems using a Computationally Efficient Set-membership Method

Seyed Mojtaba Tabatabaeipour, Thomas Bak

Publikation: Bidrag til bog/antologi/rapport/konference proceeding › Konferenceartikel i proceeding › Forskning › peer review

Abstract

In this paper, a computationally efficient set-membership method for robust fault detection of linear systems is proposed. The method computes an interval outerapproximation of the output of the system that is consistent with the model, the bounds on noise and disturbance, and the past measurements. If the output of the system does not belong to this interval, a fault is detected. To compute the output interval, we propose using support functions. Only two support functions for each output must be computed which results in a computationally efficient algorithm. Moreover, the method is trivially parallelizable. The method is demonstrated for fault detection of a hydraulic pitch actuator of a wind turbine. We show the effectiveness of the proposed method by comparing our results with two zonotope-based set-membership methods.

Published in:

Originalsprog	Engelsk
Titel	Control Applications (CCA), 2014 IEEE Conference on
Antal sider	6
Forlag	IEEE Press
Publikationsdato	2014
Sider	358-363
ISBN (Trykt)	978-1-4799-7408-5
DOI	https://doi.org/10.1109/CCA.2014.6981372
Status	Udgivet - 2014
Begivenhed	2014 IEEE Conference on Control Applications - Juan Les Antibes, Frankrig Varighed: 8 okt. 2014 → 10 okt. 2014 Konferencens nummer: 35444

Konference

Konference	2014 IEEE Conference on Control Applications
Nummer	35444
Land/Område	Frankrig
By	Juan Les Antibes
Periode	08/10/2014 → 10/10/2014

Navn	I E E E International Conference on Control Applications. Proceedings
ISSN	1085-1992

Adgang til dokumentet

10.1109/CCA.2014.6981372

AUB Link

Søg efter materialet i Aalborg Universitetsbiblioteks søgemaskine

NORCOWE: The Norwegian Centre for Offshore Wind Energy (NORCOWE)
Sørensen, J. D., Blaabjerg, F., Chen, Z., Bak, T., Knudsen, T., Jensen, T. N., Hu, W. & Nielsen, J. S.
<ingen navn>
01/01/2009 → 31/12/2017
Projekter: Projekt › Forskning

Citationsformater

@inproceedings{e6a64eedfe6a463d9c5403c55d83d50b,

title = "Robust Fault Detection of Linear Systems using a Computationally Efficient Set-membership Method",

abstract = "In this paper, a new computationally efficientset-membership method for robust fault detection of linearsystems is proposed. The method computes an intervalouter-approximation of the output of the system that isconsistent with the model, the bounds on noise anddisturbance, and the past measurements. If the output ofthe system does not belong to this interval, a fault isdetected. To compute the output interval, we propose usingsupport functions. Only two support functions for eachoutput must be computed which results in a computationallyefficient algorithm. Moreover, the method is triviallyparallelizable. The method is demonstrated for faultdetection of a hydraulic pitch actuator of a wind turbine.We show the effectiveness of the proposed method bycomparing our results with two zonotope-basedset-membership methods.",

author = "Tabatabaeipour, {Seyed Mojtaba} and Thomas Bak",

year = "2014",

doi = "10.1109/CCA.2014.6981372",

language = "English",

isbn = "978-1-4799-7408-5",

series = "I E E E International Conference on Control Applications. Proceedings",

publisher = "IEEE Press",

pages = "358--363",

booktitle = "Control Applications (CCA), 2014 IEEE Conference on",

note = "2014 IEEE Conference on Control Applications, CCA ; Conference date: 08-10-2014 Through 10-10-2014",

}

Tabatabaeipour, SM & Bak, T 2014, Robust Fault Detection of Linear Systems using a Computationally Efficient Set-membership Method. i Control Applications (CCA), 2014 IEEE Conference on. IEEE Press, I E E E International Conference on Control Applications. Proceedings, s. 358-363, 2014 IEEE Conference on Control Applications, Juan Les Antibes, Frankrig, 08/10/2014. https://doi.org/10.1109/CCA.2014.6981372

Robust Fault Detection of Linear Systems using a Computationally Efficient Set-membership Method. / Tabatabaeipour, Seyed Mojtaba; Bak, Thomas.
Control Applications (CCA), 2014 IEEE Conference on. IEEE Press, 2014. s. 358-363 (I E E E International Conference on Control Applications. Proceedings).

Publikation: Bidrag til bog/antologi/rapport/konference proceeding › Konferenceartikel i proceeding › Forskning › peer review

TY - GEN

T1 - Robust Fault Detection of Linear Systems using a Computationally Efficient Set-membership Method

AU - Tabatabaeipour, Seyed Mojtaba

AU - Bak, Thomas

N1 - Conference code: 35444

PY - 2014

Y1 - 2014

N2 - In this paper, a new computationally efficientset-membership method for robust fault detection of linearsystems is proposed. The method computes an intervalouter-approximation of the output of the system that isconsistent with the model, the bounds on noise anddisturbance, and the past measurements. If the output ofthe system does not belong to this interval, a fault isdetected. To compute the output interval, we propose usingsupport functions. Only two support functions for eachoutput must be computed which results in a computationallyefficient algorithm. Moreover, the method is triviallyparallelizable. The method is demonstrated for faultdetection of a hydraulic pitch actuator of a wind turbine.We show the effectiveness of the proposed method bycomparing our results with two zonotope-basedset-membership methods.

AB - In this paper, a new computationally efficientset-membership method for robust fault detection of linearsystems is proposed. The method computes an intervalouter-approximation of the output of the system that isconsistent with the model, the bounds on noise anddisturbance, and the past measurements. If the output ofthe system does not belong to this interval, a fault isdetected. To compute the output interval, we propose usingsupport functions. Only two support functions for eachoutput must be computed which results in a computationallyefficient algorithm. Moreover, the method is triviallyparallelizable. The method is demonstrated for faultdetection of a hydraulic pitch actuator of a wind turbine.We show the effectiveness of the proposed method bycomparing our results with two zonotope-basedset-membership methods.

U2 - 10.1109/CCA.2014.6981372

DO - 10.1109/CCA.2014.6981372

M3 - Article in proceeding

SN - 978-1-4799-7408-5

T3 - I E E E International Conference on Control Applications. Proceedings

SP - 358

EP - 363

BT - Control Applications (CCA), 2014 IEEE Conference on

PB - IEEE Press

T2 - 2014 IEEE Conference on Control Applications

Y2 - 8 October 2014 through 10 October 2014

ER -

Robust Fault Detection of Linear Systems using a Computationally Efficient Set-membership Method

Abstract

Konference

Adgang til dokumentet

AUB Link

Fingeraftryk

Projekter

NORCOWE: The Norwegian Centre for Offshore Wind Energy (NORCOWE)

Citationsformater