A nonparametric fault isolation approach through hybrid novelty score

Gulanbaier Tuerhong, Seoung Bum Kim

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Isolating the variable or set of variables responsible for an out-of-control signal is a challenging task in multivariate statistical process control. Several fault isolation approaches have been proposed. However, all assumed a multivariate normal distribution on the process data, an assumption that limits their applicability in many situations. In the present study we propose a nonparametric fault isolation approach based on a hybrid novelty score (HNS). A simulation study was conducted to examine the performance of our proposed HNS-based fault isolation approach, and its results were compared with both parametric and nonparametric T2 decomposition approaches. The performance of our approach was superior in the simulation to either parametric or nonparametric T2 decompositions. This was especially true in nonnormal situations.

Original languageEnglish
Title of host publicationProceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011
Pages261-266
Number of pages6
DOIs
Publication statusPublished - 2011 Sep 22
Event2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011 - Beijing, China
Duration: 2011 Jul 102011 Jul 12

Other

Other2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011
CountryChina
CityBeijing
Period11/7/1011/7/12

Fingerprint

Decomposition
Statistical process control
Normal distribution

Keywords

  • Decomposition
  • Fault-isolation
  • Multivariate statistical process control
  • Nonparametric
  • Novelty score

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems

Cite this

Tuerhong, G., & Kim, S. B. (2011). A nonparametric fault isolation approach through hybrid novelty score. In Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011 (pp. 261-266). [5984094] https://doi.org/10.1109/ISI.2011.5984094

A nonparametric fault isolation approach through hybrid novelty score. / Tuerhong, Gulanbaier; Kim, Seoung Bum.

Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011. 2011. p. 261-266 5984094.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tuerhong, G & Kim, SB 2011, A nonparametric fault isolation approach through hybrid novelty score. in Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011., 5984094, pp. 261-266, 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011, Beijing, China, 11/7/10. https://doi.org/10.1109/ISI.2011.5984094
Tuerhong G, Kim SB. A nonparametric fault isolation approach through hybrid novelty score. In Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011. 2011. p. 261-266. 5984094 https://doi.org/10.1109/ISI.2011.5984094
Tuerhong, Gulanbaier ; Kim, Seoung Bum. / A nonparametric fault isolation approach through hybrid novelty score. Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011. 2011. pp. 261-266
@inproceedings{c7160861282644339aceff9fc9516668,
title = "A nonparametric fault isolation approach through hybrid novelty score",
abstract = "Isolating the variable or set of variables responsible for an out-of-control signal is a challenging task in multivariate statistical process control. Several fault isolation approaches have been proposed. However, all assumed a multivariate normal distribution on the process data, an assumption that limits their applicability in many situations. In the present study we propose a nonparametric fault isolation approach based on a hybrid novelty score (HNS). A simulation study was conducted to examine the performance of our proposed HNS-based fault isolation approach, and its results were compared with both parametric and nonparametric T2 decomposition approaches. The performance of our approach was superior in the simulation to either parametric or nonparametric T2 decompositions. This was especially true in nonnormal situations.",
keywords = "Decomposition, Fault-isolation, Multivariate statistical process control, Nonparametric, Novelty score",
author = "Gulanbaier Tuerhong and Kim, {Seoung Bum}",
year = "2011",
month = "9",
day = "22",
doi = "10.1109/ISI.2011.5984094",
language = "English",
isbn = "9781457700828",
pages = "261--266",
booktitle = "Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011",

}

TY - GEN

T1 - A nonparametric fault isolation approach through hybrid novelty score

AU - Tuerhong, Gulanbaier

AU - Kim, Seoung Bum

PY - 2011/9/22

Y1 - 2011/9/22

N2 - Isolating the variable or set of variables responsible for an out-of-control signal is a challenging task in multivariate statistical process control. Several fault isolation approaches have been proposed. However, all assumed a multivariate normal distribution on the process data, an assumption that limits their applicability in many situations. In the present study we propose a nonparametric fault isolation approach based on a hybrid novelty score (HNS). A simulation study was conducted to examine the performance of our proposed HNS-based fault isolation approach, and its results were compared with both parametric and nonparametric T2 decomposition approaches. The performance of our approach was superior in the simulation to either parametric or nonparametric T2 decompositions. This was especially true in nonnormal situations.

AB - Isolating the variable or set of variables responsible for an out-of-control signal is a challenging task in multivariate statistical process control. Several fault isolation approaches have been proposed. However, all assumed a multivariate normal distribution on the process data, an assumption that limits their applicability in many situations. In the present study we propose a nonparametric fault isolation approach based on a hybrid novelty score (HNS). A simulation study was conducted to examine the performance of our proposed HNS-based fault isolation approach, and its results were compared with both parametric and nonparametric T2 decomposition approaches. The performance of our approach was superior in the simulation to either parametric or nonparametric T2 decompositions. This was especially true in nonnormal situations.

KW - Decomposition

KW - Fault-isolation

KW - Multivariate statistical process control

KW - Nonparametric

KW - Novelty score

UR - http://www.scopus.com/inward/record.url?scp=80052892844&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052892844&partnerID=8YFLogxK

U2 - 10.1109/ISI.2011.5984094

DO - 10.1109/ISI.2011.5984094

M3 - Conference contribution

AN - SCOPUS:80052892844

SN - 9781457700828

SP - 261

EP - 266

BT - Proceedings of 2011 IEEE International Conference on Intelligence and Security Informatics, ISI 2011

ER -