A measurement-based technique for incipient anomaly detection

Handle URI:
http://hdl.handle.net/10754/621301
Title:
A measurement-based technique for incipient anomaly detection
Authors:
Harrou, Fouzi; Sun, Ying ( 0000-0001-6703-4270 )
Abstract:
Fault detection is essential for safe operation of various engineering systems. Principal component analysis (PCA) has been widely used in monitoring highly correlated process variables. Conventional PCA-based methods, nevertheless, often fail to detect small or incipient faults. In this paper, we develop new PCA-based monitoring charts, combining PCA with multivariate memory control charts, such as the multivariate cumulative sum (MCUSUM) and multivariate exponentially weighted moving average (MEWMA) monitoring schemes. The multivariate control charts with memory are sensitive to small and moderate faults in the process mean, which significantly improves the performance of PCA methods and widen their applicability in practice. Using simulated data, we demonstrate that the proposed PCA-based MEWMA and MCUSUM control charts are more effective in detecting small shifts in the mean of the multivariate process variables, and outperform the conventional PCA-based monitoring charts. © 2015 IEEE.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Citation:
Harrou F, Sun Y (2015) A measurement-based technique for incipient anomaly detection. 2015 15th International Conference on Intelligent Systems Design and Applications (ISDA). Available: http://dx.doi.org/10.1109/ISDA.2015.7489200.
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
2015 15th International Conference on Intelligent Systems Design and Applications (ISDA)
Conference/Event name:
15th International Conference on Intelligent Systems Design and Applications, ISDA 2015
Issue Date:
13-Jun-2016
DOI:
10.1109/ISDA.2015.7489200
Type:
Conference Paper
Appears in Collections:
Conference Papers; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorHarrou, Fouzien
dc.contributor.authorSun, Yingen
dc.date.accessioned2016-11-03T06:57:12Z-
dc.date.available2016-11-03T06:57:12Z-
dc.date.issued2016-06-13en
dc.identifier.citationHarrou F, Sun Y (2015) A measurement-based technique for incipient anomaly detection. 2015 15th International Conference on Intelligent Systems Design and Applications (ISDA). Available: http://dx.doi.org/10.1109/ISDA.2015.7489200.en
dc.identifier.doi10.1109/ISDA.2015.7489200en
dc.identifier.urihttp://hdl.handle.net/10754/621301-
dc.description.abstractFault detection is essential for safe operation of various engineering systems. Principal component analysis (PCA) has been widely used in monitoring highly correlated process variables. Conventional PCA-based methods, nevertheless, often fail to detect small or incipient faults. In this paper, we develop new PCA-based monitoring charts, combining PCA with multivariate memory control charts, such as the multivariate cumulative sum (MCUSUM) and multivariate exponentially weighted moving average (MEWMA) monitoring schemes. The multivariate control charts with memory are sensitive to small and moderate faults in the process mean, which significantly improves the performance of PCA methods and widen their applicability in practice. Using simulated data, we demonstrate that the proposed PCA-based MEWMA and MCUSUM control charts are more effective in detecting small shifts in the mean of the multivariate process variables, and outperform the conventional PCA-based monitoring charts. © 2015 IEEE.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.titleA measurement-based technique for incipient anomaly detectionen
dc.typeConference Paperen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journal2015 15th International Conference on Intelligent Systems Design and Applications (ISDA)en
dc.conference.date14 December 2015 through 16 December 2015en
dc.conference.name15th International Conference on Intelligent Systems Design and Applications, ISDA 2015en
kaust.authorHarrou, Fouzien
kaust.authorSun, Yingen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.