Kernel Principal Component Analysis Improvement based on Data-Reduction via Class Interval

Mohammed Tahar Habib Kaib; Abdelmalek Kouadri; Mohamed Faouzi Harkat; Abderazak Bensmail; Majdi Mansouri; Mohamed Nounou

doi:10.1016/j.ifacol.2024.07.249

Kernel Principal Component Analysis Improvement based on Data-Reduction via Class Interval

Mohammed Tahar Habib Kaib, Abdelmalek Kouadri, Mohamed Faouzi Harkat, Abderazak Bensmail, Majdi Mansouri, Mohamed Nounou

Research output: Contribution to journal › Article › peer-review

Abstract

Kernel Principal Component Analysis (KPCA) is an effective nonlinear extension of the Principal Component Analysis for fault detection. For large-sized data, KPCA may drop its detection performance, occupy more storage space for the monitoring model, and take more execution time in the online part. Reduced KPCA pre-processes the training data before applying the KPCA method, the proposed approach selects samples based on class interval to reduce the number of observations in the training data set while maintaining decent detection performance. This approach is applied to the Tennessee Eastman Process and then compared to some of the existing approaches. Copyright (c) 2024 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)

Original language	English
Pages (from-to)	390-395
Number of pages	6
Journal	IFAC-PapersOnLine
Volume	58
Issue number	4
DOIs	https://doi.org/10.1016/j.ifacol.2024.07.249
Publication status	Published - 1 Jun 2024
Externally published	Yes
Event	12th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes, SAFEPROCESS 2024 - Ferrara, Italy Duration: 4 Jun 2024 → 7 Jun 2024

Keywords

Data-driven techniques
Fault Detection (FD)
Histogram
Kernel Principal Component Analysis (KPCA)
Principal Component Analysis (PCA)
Tennessee Eastman Process

Access to Document

10.1016/j.ifacol.2024.07.249

Cite this

@article{f13233f7fc3c4310a987ef419a592947,

title = "Kernel Principal Component Analysis Improvement based on Data-Reduction via Class Interval",

abstract = "Kernel Principal Component Analysis (KPCA) is an effective nonlinear extension of the Principal Component Analysis for fault detection. For large-sized data, KPCA may drop its detection performance, occupy more storage space for the monitoring model, and take more execution time in the online part. Reduced KPCA pre-processes the training data before applying the KPCA method, the proposed approach selects samples based on class interval to reduce the number of observations in the training data set while maintaining decent detection performance. This approach is applied to the Tennessee Eastman Process and then compared to some of the existing approaches. Copyright (c) 2024 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)",

keywords = "Data-driven techniques, Fault Detection (FD), Histogram, Kernel Principal Component Analysis (KPCA), Principal Component Analysis (PCA), Tennessee Eastman Process",

author = "{Habib Kaib}, {Mohammed Tahar} and Abdelmalek Kouadri and Harkat, {Mohamed Faouzi} and Abderazak Bensmail and Majdi Mansouri and Mohamed Nounou",

note = "Publisher Copyright: {\textcopyright} 2024 The Authors. This is an open access article under the CC BY-NC-ND license.; 12th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes, SAFEPROCESS 2024 ; Conference date: 04-06-2024 Through 07-06-2024",

year = "2024",

month = jun,

day = "1",

doi = "10.1016/j.ifacol.2024.07.249",

language = "English",

volume = "58",

pages = "390--395",

journal = "IFAC-PapersOnLine",

issn = "2405-8963",

publisher = "Elsevier B.V.",

number = "4",

}

TY - JOUR

T1 - Kernel Principal Component Analysis Improvement based on Data-Reduction via Class Interval

AU - Habib Kaib, Mohammed Tahar

AU - Kouadri, Abdelmalek

AU - Harkat, Mohamed Faouzi

AU - Bensmail, Abderazak

AU - Mansouri, Majdi

AU - Nounou, Mohamed

PY - 2024/6/1

Y1 - 2024/6/1

N2 - Kernel Principal Component Analysis (KPCA) is an effective nonlinear extension of the Principal Component Analysis for fault detection. For large-sized data, KPCA may drop its detection performance, occupy more storage space for the monitoring model, and take more execution time in the online part. Reduced KPCA pre-processes the training data before applying the KPCA method, the proposed approach selects samples based on class interval to reduce the number of observations in the training data set while maintaining decent detection performance. This approach is applied to the Tennessee Eastman Process and then compared to some of the existing approaches. Copyright (c) 2024 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)

AB - Kernel Principal Component Analysis (KPCA) is an effective nonlinear extension of the Principal Component Analysis for fault detection. For large-sized data, KPCA may drop its detection performance, occupy more storage space for the monitoring model, and take more execution time in the online part. Reduced KPCA pre-processes the training data before applying the KPCA method, the proposed approach selects samples based on class interval to reduce the number of observations in the training data set while maintaining decent detection performance. This approach is applied to the Tennessee Eastman Process and then compared to some of the existing approaches. Copyright (c) 2024 The Authors. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0/)

KW - Data-driven techniques

KW - Fault Detection (FD)

KW - Histogram

KW - Kernel Principal Component Analysis (KPCA)

KW - Principal Component Analysis (PCA)

KW - Tennessee Eastman Process

UR - http://www.scopus.com/inward/record.url?scp=85202898347&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2024.07.249

DO - 10.1016/j.ifacol.2024.07.249

M3 - Article

AN - SCOPUS:85202898347

SN - 2405-8963

VL - 58

SP - 390

EP - 395

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

IS - 4

T2 - 12th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes, SAFEPROCESS 2024

Y2 - 4 June 2024 through 7 June 2024

ER -

Kernel Principal Component Analysis Improvement based on Data-Reduction via Class Interval

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this