Fault diagnosis of nonlinear processes based on structured adaptive kernel PCA

Chouaib Chakour, Mohamed-Faouzi Harkat, Messaoud Djeghaba

Research output: Contribution to conferencePaper

Abstract

In this paper a new algorithm for adaptive kernel principal component analysis (AKPCA) is proposed for dynamic process monitoring. The proposed AKPCA algorithm combine two existing algorithms, the recursive weighted PCA (RWPCA) and the moving window kernel PCA algorithms. For fault detection and isolation, a set of structured residuals is generated by using a partial AKPCA models. Each partial AKPCA model is performed on subsets of variables. The structured residuals are utilized in composing an isolation scheme, according to a properly designed incidence matrix. The results for applying this algorithm on the nonlinear time varying processes of the Tennessee Eastman shows its feasibility and advantageous performances.

Original languageEnglish
Pages61-66
Number of pages6
DOIs
Publication statusPublished - 1 Jan 2013
Externally publishedYes
Event2013 3rd International Conference on Systems and Control, ICSC 2013 - Algiers, Algeria
Duration: 29 Oct 201331 Oct 2013

Other

Other2013 3rd International Conference on Systems and Control, ICSC 2013
CountryAlgeria
CityAlgiers
Period29/10/1331/10/13

Fingerprint

Failure analysis
Principal component analysis
Process monitoring
Set theory
Fault detection

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Chakour, C., Harkat, M-F., & Djeghaba, M. (2013). Fault diagnosis of nonlinear processes based on structured adaptive kernel PCA. 61-66. Paper presented at 2013 3rd International Conference on Systems and Control, ICSC 2013, Algiers, Algeria. https://doi.org/10.1109/ICoSC.2013.6750836

Fault diagnosis of nonlinear processes based on structured adaptive kernel PCA. / Chakour, Chouaib; Harkat, Mohamed-Faouzi; Djeghaba, Messaoud.

2013. 61-66 Paper presented at 2013 3rd International Conference on Systems and Control, ICSC 2013, Algiers, Algeria.

Research output: Contribution to conferencePaper

Chakour, C, Harkat, M-F & Djeghaba, M 2013, 'Fault diagnosis of nonlinear processes based on structured adaptive kernel PCA' Paper presented at 2013 3rd International Conference on Systems and Control, ICSC 2013, Algiers, Algeria, 29/10/13 - 31/10/13, pp. 61-66. https://doi.org/10.1109/ICoSC.2013.6750836
Chakour C, Harkat M-F, Djeghaba M. Fault diagnosis of nonlinear processes based on structured adaptive kernel PCA. 2013. Paper presented at 2013 3rd International Conference on Systems and Control, ICSC 2013, Algiers, Algeria. https://doi.org/10.1109/ICoSC.2013.6750836
Chakour, Chouaib ; Harkat, Mohamed-Faouzi ; Djeghaba, Messaoud. / Fault diagnosis of nonlinear processes based on structured adaptive kernel PCA. Paper presented at 2013 3rd International Conference on Systems and Control, ICSC 2013, Algiers, Algeria.6 p.
@conference{f0eb3ed285184008a4103d90585de982,
title = "Fault diagnosis of nonlinear processes based on structured adaptive kernel PCA",
abstract = "In this paper a new algorithm for adaptive kernel principal component analysis (AKPCA) is proposed for dynamic process monitoring. The proposed AKPCA algorithm combine two existing algorithms, the recursive weighted PCA (RWPCA) and the moving window kernel PCA algorithms. For fault detection and isolation, a set of structured residuals is generated by using a partial AKPCA models. Each partial AKPCA model is performed on subsets of variables. The structured residuals are utilized in composing an isolation scheme, according to a properly designed incidence matrix. The results for applying this algorithm on the nonlinear time varying processes of the Tennessee Eastman shows its feasibility and advantageous performances.",
author = "Chouaib Chakour and Mohamed-Faouzi Harkat and Messaoud Djeghaba",
year = "2013",
month = "1",
day = "1",
doi = "10.1109/ICoSC.2013.6750836",
language = "English",
pages = "61--66",
note = "2013 3rd International Conference on Systems and Control, ICSC 2013 ; Conference date: 29-10-2013 Through 31-10-2013",

}

TY - CONF

T1 - Fault diagnosis of nonlinear processes based on structured adaptive kernel PCA

AU - Chakour, Chouaib

AU - Harkat, Mohamed-Faouzi

AU - Djeghaba, Messaoud

PY - 2013/1/1

Y1 - 2013/1/1

N2 - In this paper a new algorithm for adaptive kernel principal component analysis (AKPCA) is proposed for dynamic process monitoring. The proposed AKPCA algorithm combine two existing algorithms, the recursive weighted PCA (RWPCA) and the moving window kernel PCA algorithms. For fault detection and isolation, a set of structured residuals is generated by using a partial AKPCA models. Each partial AKPCA model is performed on subsets of variables. The structured residuals are utilized in composing an isolation scheme, according to a properly designed incidence matrix. The results for applying this algorithm on the nonlinear time varying processes of the Tennessee Eastman shows its feasibility and advantageous performances.

AB - In this paper a new algorithm for adaptive kernel principal component analysis (AKPCA) is proposed for dynamic process monitoring. The proposed AKPCA algorithm combine two existing algorithms, the recursive weighted PCA (RWPCA) and the moving window kernel PCA algorithms. For fault detection and isolation, a set of structured residuals is generated by using a partial AKPCA models. Each partial AKPCA model is performed on subsets of variables. The structured residuals are utilized in composing an isolation scheme, according to a properly designed incidence matrix. The results for applying this algorithm on the nonlinear time varying processes of the Tennessee Eastman shows its feasibility and advantageous performances.

UR - http://www.scopus.com/inward/record.url?scp=84896598844&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84896598844&partnerID=8YFLogxK

U2 - 10.1109/ICoSC.2013.6750836

DO - 10.1109/ICoSC.2013.6750836

M3 - Paper

SP - 61

EP - 66

ER -