Iterated Robust kernel Fuzzy Principal Component Analysis and application to fault detection

Raoudha Baklouti, Majdi Mansouri, Mohamed Nounou, Hazem Nounou, Ahmed Ben Hamida

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

In this paper, we propose an Iterated Robust kernel Fuzzy Principal Component Analysis (IRkFPCA), which is the method that attempts to combine the advantages of the state of art methods and use a more accurate multi-objective function for jointly reducing the modeling errors, optimizing the robustness to outliers and improving the time complexity since it does not require the storage and inversion of the covariance matrix to obtain a memory-efficient approximation of kernel PCA. This proposed technique computes iteratively the principal components, which are used for modeling and fault detection. The detection stage is related to the evaluation of residuals, also known as detection indices, which are signals that reveal the fault presence. Those indices are obtained from the analysis of the difference between the process measurements and their estimations using the IRkFPCA technique. The performance of the proposed method is illustrated and compared to Iterated kernel Principal Component Analysis (IkPCA) and Iterated Principal Component Analysis (IPCA) methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of the comparative studies reveal that the developed IRkFPCA method provides a better performance in terms of modeling and fault detection accuracies than the Iterated Robust Fuzzy Principal Component Analysis (IRFPCA) and Iterated kernel Principal Component Analysis (IkPCA) methods; while both methods provide improved accuracy over the Iterated Principal Component Analysis (IPCA) method.

Original languageEnglish
Pages (from-to)34-49
Number of pages16
JournalJournal of Computational Science
Volume15
DOIs
Publication statusPublished - 1 Jul 2016

Fingerprint

Fault Detection
Fault detection
Principal component analysis
Principal Component Analysis
kernel
Kernel Principal Component Analysis
Kernel PCA
Modeling Error
Principal Components
Synthetic Data
Covariance matrix
Modeling
Reactor
Time Complexity
Outlier
Comparative Study
Inversion
Fault
Objective function
Robustness

Keywords

  • Fault detection
  • Iterated robust fuzzy
  • Kernel principal component analysis
  • Modeling

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)
  • Modelling and Simulation

Cite this

Iterated Robust kernel Fuzzy Principal Component Analysis and application to fault detection. / Baklouti, Raoudha; Mansouri, Majdi; Nounou, Mohamed; Nounou, Hazem; Hamida, Ahmed Ben.

In: Journal of Computational Science, Vol. 15, 01.07.2016, p. 34-49.

Research output: Contribution to journalArticle

@article{fcf25884c4234069aeea25393ac6a1ca,
title = "Iterated Robust kernel Fuzzy Principal Component Analysis and application to fault detection",
abstract = "In this paper, we propose an Iterated Robust kernel Fuzzy Principal Component Analysis (IRkFPCA), which is the method that attempts to combine the advantages of the state of art methods and use a more accurate multi-objective function for jointly reducing the modeling errors, optimizing the robustness to outliers and improving the time complexity since it does not require the storage and inversion of the covariance matrix to obtain a memory-efficient approximation of kernel PCA. This proposed technique computes iteratively the principal components, which are used for modeling and fault detection. The detection stage is related to the evaluation of residuals, also known as detection indices, which are signals that reveal the fault presence. Those indices are obtained from the analysis of the difference between the process measurements and their estimations using the IRkFPCA technique. The performance of the proposed method is illustrated and compared to Iterated kernel Principal Component Analysis (IkPCA) and Iterated Principal Component Analysis (IPCA) methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of the comparative studies reveal that the developed IRkFPCA method provides a better performance in terms of modeling and fault detection accuracies than the Iterated Robust Fuzzy Principal Component Analysis (IRFPCA) and Iterated kernel Principal Component Analysis (IkPCA) methods; while both methods provide improved accuracy over the Iterated Principal Component Analysis (IPCA) method.",
keywords = "Fault detection, Iterated robust fuzzy, Kernel principal component analysis, Modeling",
author = "Raoudha Baklouti and Majdi Mansouri and Mohamed Nounou and Hazem Nounou and Hamida, {Ahmed Ben}",
year = "2016",
month = "7",
day = "1",
doi = "10.1016/j.jocs.2015.11.005",
language = "English",
volume = "15",
pages = "34--49",
journal = "Journal of Computational Science",
issn = "1877-7503",
publisher = "Elsevier",

}

TY - JOUR

T1 - Iterated Robust kernel Fuzzy Principal Component Analysis and application to fault detection

AU - Baklouti, Raoudha

AU - Mansouri, Majdi

AU - Nounou, Mohamed

AU - Nounou, Hazem

AU - Hamida, Ahmed Ben

PY - 2016/7/1

Y1 - 2016/7/1

N2 - In this paper, we propose an Iterated Robust kernel Fuzzy Principal Component Analysis (IRkFPCA), which is the method that attempts to combine the advantages of the state of art methods and use a more accurate multi-objective function for jointly reducing the modeling errors, optimizing the robustness to outliers and improving the time complexity since it does not require the storage and inversion of the covariance matrix to obtain a memory-efficient approximation of kernel PCA. This proposed technique computes iteratively the principal components, which are used for modeling and fault detection. The detection stage is related to the evaluation of residuals, also known as detection indices, which are signals that reveal the fault presence. Those indices are obtained from the analysis of the difference between the process measurements and their estimations using the IRkFPCA technique. The performance of the proposed method is illustrated and compared to Iterated kernel Principal Component Analysis (IkPCA) and Iterated Principal Component Analysis (IPCA) methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of the comparative studies reveal that the developed IRkFPCA method provides a better performance in terms of modeling and fault detection accuracies than the Iterated Robust Fuzzy Principal Component Analysis (IRFPCA) and Iterated kernel Principal Component Analysis (IkPCA) methods; while both methods provide improved accuracy over the Iterated Principal Component Analysis (IPCA) method.

AB - In this paper, we propose an Iterated Robust kernel Fuzzy Principal Component Analysis (IRkFPCA), which is the method that attempts to combine the advantages of the state of art methods and use a more accurate multi-objective function for jointly reducing the modeling errors, optimizing the robustness to outliers and improving the time complexity since it does not require the storage and inversion of the covariance matrix to obtain a memory-efficient approximation of kernel PCA. This proposed technique computes iteratively the principal components, which are used for modeling and fault detection. The detection stage is related to the evaluation of residuals, also known as detection indices, which are signals that reveal the fault presence. Those indices are obtained from the analysis of the difference between the process measurements and their estimations using the IRkFPCA technique. The performance of the proposed method is illustrated and compared to Iterated kernel Principal Component Analysis (IkPCA) and Iterated Principal Component Analysis (IPCA) methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of the comparative studies reveal that the developed IRkFPCA method provides a better performance in terms of modeling and fault detection accuracies than the Iterated Robust Fuzzy Principal Component Analysis (IRFPCA) and Iterated kernel Principal Component Analysis (IkPCA) methods; while both methods provide improved accuracy over the Iterated Principal Component Analysis (IPCA) method.

KW - Fault detection

KW - Iterated robust fuzzy

KW - Kernel principal component analysis

KW - Modeling

UR - http://www.scopus.com/inward/record.url?scp=84949681832&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949681832&partnerID=8YFLogxK

U2 - 10.1016/j.jocs.2015.11.005

DO - 10.1016/j.jocs.2015.11.005

M3 - Article

AN - SCOPUS:84949681832

VL - 15

SP - 34

EP - 49

JO - Journal of Computational Science

JF - Journal of Computational Science

SN - 1877-7503

ER -