Online process monitoring based on kernel method

Radhia Fezai, Ines Jaffel, Okba Taouali, Mohamed-Faouzi Harkat, Nasreddine Bouguila

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

2 Citations (Scopus)

Abstract

This paper discusses the monitoring of dynamic process. In recent years, Kernel Principal component analysis (KPCA) has gained significant attention as a monitoring method of nonlinear systems. However, the fixed KPCA model limit its application for dynamic systems. For this purpose a new Variable Moving Window Kernel PCA (VMWKPCA) method is introduced to update the KPCA model. The basic idea of this technique is to vary the size of the moving window depending on the normal change of the process. Then the VMWKPCA method is performed for monitoring a Chemical reactor (CSTR). The simulation results proved that the new method is effective.

Original languageEnglish
Title of host publication2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages236-241
Number of pages6
ISBN (Electronic)9781509059874
DOIs
Publication statusPublished - 19 Oct 2017
Externally publishedYes
Event2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017 - Hammamet, Tunisia
Duration: 19 Jan 201721 Jan 2017

Other

Other2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017
CountryTunisia
CityHammamet
Period19/1/1721/1/17

Fingerprint

On-line Monitoring
Process Monitoring
Process monitoring
Kernel Methods
Kernel Principal Component Analysis
Principal component analysis
Kernel PCA
Monitoring
Chemical reactors
Chemical Reactors
Nonlinear systems
Dynamical systems
Dynamic Process
Dynamic Systems
Nonlinear Systems
Update
Vary
Model
Simulation

Keywords

  • fault detection
  • Kernel principal component analysis
  • MWKPCA
  • Principal Component Analysis
  • SPE
  • T
  • VMWKPCA

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Safety, Risk, Reliability and Quality
  • Control and Optimization

Cite this

Fezai, R., Jaffel, I., Taouali, O., Harkat, M-F., & Bouguila, N. (2017). Online process monitoring based on kernel method. In 2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017 (pp. 236-241). [8075663] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/CADIAG.2017.8075663

Online process monitoring based on kernel method. / Fezai, Radhia; Jaffel, Ines; Taouali, Okba; Harkat, Mohamed-Faouzi; Bouguila, Nasreddine.

2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 236-241 8075663.

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

Fezai, R, Jaffel, I, Taouali, O, Harkat, M-F & Bouguila, N 2017, Online process monitoring based on kernel method. in 2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017., 8075663, Institute of Electrical and Electronics Engineers Inc., pp. 236-241, 2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017, Hammamet, Tunisia, 19/1/17. https://doi.org/10.1109/CADIAG.2017.8075663
Fezai R, Jaffel I, Taouali O, Harkat M-F, Bouguila N. Online process monitoring based on kernel method. In 2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 236-241. 8075663 https://doi.org/10.1109/CADIAG.2017.8075663
Fezai, Radhia ; Jaffel, Ines ; Taouali, Okba ; Harkat, Mohamed-Faouzi ; Bouguila, Nasreddine. / Online process monitoring based on kernel method. 2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 236-241
@inproceedings{3023ad1d5a024bde8fd695e91316e1ac,
title = "Online process monitoring based on kernel method",
abstract = "This paper discusses the monitoring of dynamic process. In recent years, Kernel Principal component analysis (KPCA) has gained significant attention as a monitoring method of nonlinear systems. However, the fixed KPCA model limit its application for dynamic systems. For this purpose a new Variable Moving Window Kernel PCA (VMWKPCA) method is introduced to update the KPCA model. The basic idea of this technique is to vary the size of the moving window depending on the normal change of the process. Then the VMWKPCA method is performed for monitoring a Chemical reactor (CSTR). The simulation results proved that the new method is effective.",
keywords = "fault detection, Kernel principal component analysis, MWKPCA, Principal Component Analysis, SPE, T, VMWKPCA",
author = "Radhia Fezai and Ines Jaffel and Okba Taouali and Mohamed-Faouzi Harkat and Nasreddine Bouguila",
year = "2017",
month = "10",
day = "19",
doi = "10.1109/CADIAG.2017.8075663",
language = "English",
pages = "236--241",
booktitle = "2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Online process monitoring based on kernel method

AU - Fezai, Radhia

AU - Jaffel, Ines

AU - Taouali, Okba

AU - Harkat, Mohamed-Faouzi

AU - Bouguila, Nasreddine

PY - 2017/10/19

Y1 - 2017/10/19

N2 - This paper discusses the monitoring of dynamic process. In recent years, Kernel Principal component analysis (KPCA) has gained significant attention as a monitoring method of nonlinear systems. However, the fixed KPCA model limit its application for dynamic systems. For this purpose a new Variable Moving Window Kernel PCA (VMWKPCA) method is introduced to update the KPCA model. The basic idea of this technique is to vary the size of the moving window depending on the normal change of the process. Then the VMWKPCA method is performed for monitoring a Chemical reactor (CSTR). The simulation results proved that the new method is effective.

AB - This paper discusses the monitoring of dynamic process. In recent years, Kernel Principal component analysis (KPCA) has gained significant attention as a monitoring method of nonlinear systems. However, the fixed KPCA model limit its application for dynamic systems. For this purpose a new Variable Moving Window Kernel PCA (VMWKPCA) method is introduced to update the KPCA model. The basic idea of this technique is to vary the size of the moving window depending on the normal change of the process. Then the VMWKPCA method is performed for monitoring a Chemical reactor (CSTR). The simulation results proved that the new method is effective.

KW - fault detection

KW - Kernel principal component analysis

KW - MWKPCA

KW - Principal Component Analysis

KW - SPE

KW - T

KW - VMWKPCA

UR - http://www.scopus.com/inward/record.url?scp=85040012390&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85040012390&partnerID=8YFLogxK

U2 - 10.1109/CADIAG.2017.8075663

DO - 10.1109/CADIAG.2017.8075663

M3 - Conference contribution

SP - 236

EP - 241

BT - 2017 International Conference on Control, Automation and Diagnosis, ICCAD 2017

PB - Institute of Electrical and Electronics Engineers Inc.

ER -