New adaptive moving window PCA for process monitoring

N. Ayech, C. Chakour, Mohamed-Faouzi Harkat

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

4 Citations (Scopus)

Abstract

Slow and normal process changes often occur in real processes, which lead to false alarms for a fixed-model monitoring approach. In this paper, two recursive PCA algorithms for adaptive process monitoring are studied. the first algorithm is based on Moving Window Principal Component Analysis (MWPCA), and the second is based on Forgetting Factors Principal Component Analysis (Recursive Weighted PCA). Furthermore, by changing the size and the shift of the window, also for the forgetting factor, we will see the influence of these changes on the monitoring performances. Then, adaptive forgetting factors will be used, for increasing the robustness against outliers. Using the same concept of varying forgetting factors, a new recursive algorithm for adaptive process monitoring based on Moving Window is proposed. By using the current model and the updated mean and covariance structures and an Adaptive Moving Window, a new model is derived recursively (AMWPCA). Based on the updated PCA representation the Q-statistic (SPE) (monitoring metric) is calculated and their control limits are updated. The feasibility and advantages of each algorithms is illustrated by application to Tennessee Eastman process.

Original languageEnglish
Title of host publication8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012
Pages606-611
Number of pages6
Volume8
EditionPART 1
DOIs
Publication statusPublished - 9 Oct 2012
Externally publishedYes
Event8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012 - Mexico City, Mexico
Duration: 29 Aug 201231 Aug 2012

Other

Other8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012
CountryMexico
CityMexico City
Period29/8/1231/8/12

Fingerprint

Process monitoring
Principal component analysis
Monitoring
Statistics

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Ayech, N., Chakour, C., & Harkat, M-F. (2012). New adaptive moving window PCA for process monitoring. In 8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012 (PART 1 ed., Vol. 8, pp. 606-611) https://doi.org/10.3182/20120829-3-MX-2028.00198

New adaptive moving window PCA for process monitoring. / Ayech, N.; Chakour, C.; Harkat, Mohamed-Faouzi.

8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012. Vol. 8 PART 1. ed. 2012. p. 606-611.

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

Ayech, N, Chakour, C & Harkat, M-F 2012, New adaptive moving window PCA for process monitoring. in 8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012. PART 1 edn, vol. 8, pp. 606-611, 8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012, Mexico City, Mexico, 29/8/12. https://doi.org/10.3182/20120829-3-MX-2028.00198
Ayech N, Chakour C, Harkat M-F. New adaptive moving window PCA for process monitoring. In 8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012. PART 1 ed. Vol. 8. 2012. p. 606-611 https://doi.org/10.3182/20120829-3-MX-2028.00198
Ayech, N. ; Chakour, C. ; Harkat, Mohamed-Faouzi. / New adaptive moving window PCA for process monitoring. 8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012. Vol. 8 PART 1. ed. 2012. pp. 606-611
@inproceedings{637b752c835c4851b0b444ffb57b7ddd,
title = "New adaptive moving window PCA for process monitoring",
abstract = "Slow and normal process changes often occur in real processes, which lead to false alarms for a fixed-model monitoring approach. In this paper, two recursive PCA algorithms for adaptive process monitoring are studied. the first algorithm is based on Moving Window Principal Component Analysis (MWPCA), and the second is based on Forgetting Factors Principal Component Analysis (Recursive Weighted PCA). Furthermore, by changing the size and the shift of the window, also for the forgetting factor, we will see the influence of these changes on the monitoring performances. Then, adaptive forgetting factors will be used, for increasing the robustness against outliers. Using the same concept of varying forgetting factors, a new recursive algorithm for adaptive process monitoring based on Moving Window is proposed. By using the current model and the updated mean and covariance structures and an Adaptive Moving Window, a new model is derived recursively (AMWPCA). Based on the updated PCA representation the Q-statistic (SPE) (monitoring metric) is calculated and their control limits are updated. The feasibility and advantages of each algorithms is illustrated by application to Tennessee Eastman process.",
author = "N. Ayech and C. Chakour and Mohamed-Faouzi Harkat",
year = "2012",
month = "10",
day = "9",
doi = "10.3182/20120829-3-MX-2028.00198",
language = "English",
isbn = "9783902823090",
volume = "8",
pages = "606--611",
booktitle = "8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012",
edition = "PART 1",

}

TY - GEN

T1 - New adaptive moving window PCA for process monitoring

AU - Ayech, N.

AU - Chakour, C.

AU - Harkat, Mohamed-Faouzi

PY - 2012/10/9

Y1 - 2012/10/9

N2 - Slow and normal process changes often occur in real processes, which lead to false alarms for a fixed-model monitoring approach. In this paper, two recursive PCA algorithms for adaptive process monitoring are studied. the first algorithm is based on Moving Window Principal Component Analysis (MWPCA), and the second is based on Forgetting Factors Principal Component Analysis (Recursive Weighted PCA). Furthermore, by changing the size and the shift of the window, also for the forgetting factor, we will see the influence of these changes on the monitoring performances. Then, adaptive forgetting factors will be used, for increasing the robustness against outliers. Using the same concept of varying forgetting factors, a new recursive algorithm for adaptive process monitoring based on Moving Window is proposed. By using the current model and the updated mean and covariance structures and an Adaptive Moving Window, a new model is derived recursively (AMWPCA). Based on the updated PCA representation the Q-statistic (SPE) (monitoring metric) is calculated and their control limits are updated. The feasibility and advantages of each algorithms is illustrated by application to Tennessee Eastman process.

AB - Slow and normal process changes often occur in real processes, which lead to false alarms for a fixed-model monitoring approach. In this paper, two recursive PCA algorithms for adaptive process monitoring are studied. the first algorithm is based on Moving Window Principal Component Analysis (MWPCA), and the second is based on Forgetting Factors Principal Component Analysis (Recursive Weighted PCA). Furthermore, by changing the size and the shift of the window, also for the forgetting factor, we will see the influence of these changes on the monitoring performances. Then, adaptive forgetting factors will be used, for increasing the robustness against outliers. Using the same concept of varying forgetting factors, a new recursive algorithm for adaptive process monitoring based on Moving Window is proposed. By using the current model and the updated mean and covariance structures and an Adaptive Moving Window, a new model is derived recursively (AMWPCA). Based on the updated PCA representation the Q-statistic (SPE) (monitoring metric) is calculated and their control limits are updated. The feasibility and advantages of each algorithms is illustrated by application to Tennessee Eastman process.

UR - http://www.scopus.com/inward/record.url?scp=84867090292&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84867090292&partnerID=8YFLogxK

U2 - 10.3182/20120829-3-MX-2028.00198

DO - 10.3182/20120829-3-MX-2028.00198

M3 - Conference contribution

SN - 9783902823090

VL - 8

SP - 606

EP - 611

BT - 8th IFAC Symposium on Fault Detection, Supervision and Safety of Technical Processes, SAFEPROCESS 2012

ER -