Online reduced kernel principal component analysis for process monitoring

Radhia Fezai, Majdi Mansouri, Okba Taouali, Mohamed-Faouzi Harkat, Nasreddine Bouguila

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

Kernel principal component analysis (KPCA), which is a nonlinear extension of principal component analysis (PCA), has gained significant attention as a monitoring method for nonlinear processes. However, KPCA cannot perform well for dynamic systems and when the training data set is large. Therefore, in this paper, an online reduced KPCA algorithm for process monitoring is proposed. The process monitoring performances are studied using two examples: a numerical example and Tennessee Eastman Process (TEP). The simulation results demonstrate the effectiveness of the proposed method when compared to the online KPCA method.

Original languageEnglish
Pages (from-to)1-11
Number of pages11
JournalJournal of Process Control
Volume61
DOIs
Publication statusPublished - 1 Jan 2018

Fingerprint

Kernel Principal Component Analysis
Process Monitoring
Process monitoring
Principal component analysis
Nonlinear Process
Principal Component Analysis
Dynamic Systems
Monitoring
Dynamical systems
Numerical Examples
Demonstrate
Simulation

Keywords

  • Dictionary
  • Dynamic process
  • Fault detection
  • Kernel PCA
  • Principal component analysis (PCA)
  • Reduced kernel PCA

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Modelling and Simulation
  • Computer Science Applications
  • Industrial and Manufacturing Engineering

Cite this

Online reduced kernel principal component analysis for process monitoring. / Fezai, Radhia; Mansouri, Majdi; Taouali, Okba; Harkat, Mohamed-Faouzi; Bouguila, Nasreddine.

In: Journal of Process Control, Vol. 61, 01.01.2018, p. 1-11.

Research output: Contribution to journalArticle

@article{7fd0030de09d4bad9ae30704ef0a61e7,
title = "Online reduced kernel principal component analysis for process monitoring",
abstract = "Kernel principal component analysis (KPCA), which is a nonlinear extension of principal component analysis (PCA), has gained significant attention as a monitoring method for nonlinear processes. However, KPCA cannot perform well for dynamic systems and when the training data set is large. Therefore, in this paper, an online reduced KPCA algorithm for process monitoring is proposed. The process monitoring performances are studied using two examples: a numerical example and Tennessee Eastman Process (TEP). The simulation results demonstrate the effectiveness of the proposed method when compared to the online KPCA method.",
keywords = "Dictionary, Dynamic process, Fault detection, Kernel PCA, Principal component analysis (PCA), Reduced kernel PCA",
author = "Radhia Fezai and Majdi Mansouri and Okba Taouali and Mohamed-Faouzi Harkat and Nasreddine Bouguila",
year = "2018",
month = "1",
day = "1",
doi = "10.1016/j.jprocont.2017.10.010",
language = "English",
volume = "61",
pages = "1--11",
journal = "Journal of Process Control",
issn = "0959-1524",
publisher = "Elsevier Limited",

}

TY - JOUR

T1 - Online reduced kernel principal component analysis for process monitoring

AU - Fezai, Radhia

AU - Mansouri, Majdi

AU - Taouali, Okba

AU - Harkat, Mohamed-Faouzi

AU - Bouguila, Nasreddine

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Kernel principal component analysis (KPCA), which is a nonlinear extension of principal component analysis (PCA), has gained significant attention as a monitoring method for nonlinear processes. However, KPCA cannot perform well for dynamic systems and when the training data set is large. Therefore, in this paper, an online reduced KPCA algorithm for process monitoring is proposed. The process monitoring performances are studied using two examples: a numerical example and Tennessee Eastman Process (TEP). The simulation results demonstrate the effectiveness of the proposed method when compared to the online KPCA method.

AB - Kernel principal component analysis (KPCA), which is a nonlinear extension of principal component analysis (PCA), has gained significant attention as a monitoring method for nonlinear processes. However, KPCA cannot perform well for dynamic systems and when the training data set is large. Therefore, in this paper, an online reduced KPCA algorithm for process monitoring is proposed. The process monitoring performances are studied using two examples: a numerical example and Tennessee Eastman Process (TEP). The simulation results demonstrate the effectiveness of the proposed method when compared to the online KPCA method.

KW - Dictionary

KW - Dynamic process

KW - Fault detection

KW - Kernel PCA

KW - Principal component analysis (PCA)

KW - Reduced kernel PCA

UR - http://www.scopus.com/inward/record.url?scp=85034268969&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85034268969&partnerID=8YFLogxK

U2 - 10.1016/j.jprocont.2017.10.010

DO - 10.1016/j.jprocont.2017.10.010

M3 - Article

VL - 61

SP - 1

EP - 11

JO - Journal of Process Control

JF - Journal of Process Control

SN - 0959-1524

ER -