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

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review

Abstract

In this paper, a new computationally efficient
set-membership method for robust fault detection of linear
systems is proposed. The method computes an interval
outer-approximation 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.

Original language	English
Title of host publication	Control Applications (CCA), 2014 IEEE Conference on
Number of pages	6
Publisher	IEEE Press
Publication date	2014
Pages	358-363
ISBN (Print)	978-1-4799-7408-5
DOIs	https://doi.org/10.1109/CCA.2014.6981372
Publication status	Published - 2014
Event	2014 IEEE Conference on Control Applications - Juan Les Antibes, France Duration: 8 Oct 2014 → 10 Oct 2014 Conference number: 35444

Conference

Conference	2014 IEEE Conference on Control Applications
Number	35444
Country/Territory	France
City	Juan Les Antibes
Period	08/10/2014 → 10/10/2014

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

Access to Document

10.1109/CCA.2014.6981372

AUB Link

Search for the material in Aalborg University Library's search engine

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
Project: Research

Cite this

@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. in Control Applications (CCA), 2014 IEEE Conference on. IEEE Press, I E E E International Conference on Control Applications. Proceedings, pp. 358-363, 2014 IEEE Conference on Control Applications, Juan Les Antibes, France, 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. p. 358-363 (I E E E International Conference on Control Applications. Proceedings).

Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › 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

Conference

Access to Document

AUB Link

Fingerprint

Projects

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

Cite this