Fault detection in processes represented by PLS models using an EWMA control scheme

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

Abstract

Fault detection is important for effective and safe process operation. Partial least squares (PLS) has been used successfully in fault detection for multivariate processes with highly correlated variables. However, the conventional PLS-based detection metrics, such as the Hotelling's T2 and the Q statistics are not well suited to detect small faults because they only use information about the process in the most recent observation. Exponentially weighed moving average (EWMA), however, has been shown to be more sensitive to small shifts in the mean of process variables. In this paper, a PLS-based EWMA fault detection method is proposed for monitoring processes represented by PLS models. The performance of the proposed method is compared with that of the traditional PLS-based fault detection method through a simulated example involving various fault scenarios that could be encountered in real processes. The simulation results clearly show the effectiveness of the proposed method over the conventional PLS method.

Original languageEnglish
Title of host publicationInternational Conference on Control, Decision and Information Technologies, CoDIT 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages152-157
Number of pages6
DOIs
Publication statusPublished - 18 Oct 2016
Event3rd International Conference on Control, Decision and Information Technologies, CoDIT 2016 - Saint Julian's, Malta
Duration: 6 Apr 20168 Apr 2016

Other

Other3rd International Conference on Control, Decision and Information Technologies, CoDIT 2016
CountryMalta
CitySaint Julian's
Period6/4/168/4/16

Fingerprint

Partial Least Squares
Moving Average
Fault Detection
Fault detection
Information use
Process monitoring
Fault
Model
Hotelling's T2
Process Monitoring
Statistics
Least Square Method
Partial least squares
Moving average
Metric
Scenarios
Simulation

Keywords

  • Databased fault detection
  • EWMA
  • Mean shift
  • Partial least squares
  • Process monitoring

ASJC Scopus subject areas

  • Decision Sciences (miscellaneous)
  • Control and Optimization
  • Computer Science Applications
  • Information Systems

Cite this

Harrou, F., Nounou, M., & Nounou, H. (2016). Fault detection in processes represented by PLS models using an EWMA control scheme. In International Conference on Control, Decision and Information Technologies, CoDIT 2016 (pp. 152-157). [7593552] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CoDIT.2016.7593552

Fault detection in processes represented by PLS models using an EWMA control scheme. / Harrou, Fouzi; Nounou, Mohamed; Nounou, Hazem.

International Conference on Control, Decision and Information Technologies, CoDIT 2016. Institute of Electrical and Electronics Engineers Inc., 2016. p. 152-157 7593552.

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

Harrou, F, Nounou, M & Nounou, H 2016, Fault detection in processes represented by PLS models using an EWMA control scheme. in International Conference on Control, Decision and Information Technologies, CoDIT 2016., 7593552, Institute of Electrical and Electronics Engineers Inc., pp. 152-157, 3rd International Conference on Control, Decision and Information Technologies, CoDIT 2016, Saint Julian's, Malta, 6/4/16. https://doi.org/10.1109/CoDIT.2016.7593552
Harrou F, Nounou M, Nounou H. Fault detection in processes represented by PLS models using an EWMA control scheme. In International Conference on Control, Decision and Information Technologies, CoDIT 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 152-157. 7593552 https://doi.org/10.1109/CoDIT.2016.7593552
Harrou, Fouzi ; Nounou, Mohamed ; Nounou, Hazem. / Fault detection in processes represented by PLS models using an EWMA control scheme. International Conference on Control, Decision and Information Technologies, CoDIT 2016. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 152-157
@inproceedings{7212b72a350e4fbe96c2fe81c2b44963,
title = "Fault detection in processes represented by PLS models using an EWMA control scheme",
abstract = "Fault detection is important for effective and safe process operation. Partial least squares (PLS) has been used successfully in fault detection for multivariate processes with highly correlated variables. However, the conventional PLS-based detection metrics, such as the Hotelling's T2 and the Q statistics are not well suited to detect small faults because they only use information about the process in the most recent observation. Exponentially weighed moving average (EWMA), however, has been shown to be more sensitive to small shifts in the mean of process variables. In this paper, a PLS-based EWMA fault detection method is proposed for monitoring processes represented by PLS models. The performance of the proposed method is compared with that of the traditional PLS-based fault detection method through a simulated example involving various fault scenarios that could be encountered in real processes. The simulation results clearly show the effectiveness of the proposed method over the conventional PLS method.",
keywords = "Databased fault detection, EWMA, Mean shift, Partial least squares, Process monitoring",
author = "Fouzi Harrou and Mohamed Nounou and Hazem Nounou",
year = "2016",
month = "10",
day = "18",
doi = "10.1109/CoDIT.2016.7593552",
language = "English",
pages = "152--157",
booktitle = "International Conference on Control, Decision and Information Technologies, CoDIT 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Fault detection in processes represented by PLS models using an EWMA control scheme

AU - Harrou, Fouzi

AU - Nounou, Mohamed

AU - Nounou, Hazem

PY - 2016/10/18

Y1 - 2016/10/18

N2 - Fault detection is important for effective and safe process operation. Partial least squares (PLS) has been used successfully in fault detection for multivariate processes with highly correlated variables. However, the conventional PLS-based detection metrics, such as the Hotelling's T2 and the Q statistics are not well suited to detect small faults because they only use information about the process in the most recent observation. Exponentially weighed moving average (EWMA), however, has been shown to be more sensitive to small shifts in the mean of process variables. In this paper, a PLS-based EWMA fault detection method is proposed for monitoring processes represented by PLS models. The performance of the proposed method is compared with that of the traditional PLS-based fault detection method through a simulated example involving various fault scenarios that could be encountered in real processes. The simulation results clearly show the effectiveness of the proposed method over the conventional PLS method.

AB - Fault detection is important for effective and safe process operation. Partial least squares (PLS) has been used successfully in fault detection for multivariate processes with highly correlated variables. However, the conventional PLS-based detection metrics, such as the Hotelling's T2 and the Q statistics are not well suited to detect small faults because they only use information about the process in the most recent observation. Exponentially weighed moving average (EWMA), however, has been shown to be more sensitive to small shifts in the mean of process variables. In this paper, a PLS-based EWMA fault detection method is proposed for monitoring processes represented by PLS models. The performance of the proposed method is compared with that of the traditional PLS-based fault detection method through a simulated example involving various fault scenarios that could be encountered in real processes. The simulation results clearly show the effectiveness of the proposed method over the conventional PLS method.

KW - Databased fault detection

KW - EWMA

KW - Mean shift

KW - Partial least squares

KW - Process monitoring

UR - http://www.scopus.com/inward/record.url?scp=84995426769&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84995426769&partnerID=8YFLogxK

U2 - 10.1109/CoDIT.2016.7593552

DO - 10.1109/CoDIT.2016.7593552

M3 - Conference contribution

SP - 152

EP - 157

BT - International Conference on Control, Decision and Information Technologies, CoDIT 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -