Neuronal principal component analysis for nonlinear time-varying processes monitoring

C. Chakour, Mohamed-Faouzi Harkat, M. Djeghaba

Research output: Contribution to journalConference article

2 Citations (Scopus)

Abstract

This paper presents a new adaptive kernel principal component analysis algorithm (AKPCA) for nonlinear time-varying process monitoring. The basic idea is to use a neuronal principal component analysis based on the kernel version of the generalized Hebbian algorithm. The proposed algorithm follows a new methodology to update the KPCA model. At each time instant, when a new data is available, the KPCA model is updated accordingly without having to re-explore all previous data. By using the proposed algorithm, the performance of process monitoring is improved in two aspects; the speed computation and adaptation of the KPCA model, and the storage memory complexity. To identify faults in a dynamic process, the reconstruction based contribution approach is used and adapted in real time. The results for applying this algorithm on the Tennessee Eastman process shows its feasibility and advantageous performances.

Original languageEnglish
Pages (from-to)1408-1413
Number of pages6
JournalIFAC-PapersOnLine
Volume28
Issue number21
DOIs
Publication statusPublished - 1 Sep 2015
Externally publishedYes

Fingerprint

Process monitoring
Principal component analysis
Data storage equipment

Keywords

  • Dynamic process
  • Fault detection and isolation
  • Kernel principal component analysis

ASJC Scopus subject areas

  • Control and Systems Engineering

Cite this

Neuronal principal component analysis for nonlinear time-varying processes monitoring. / Chakour, C.; Harkat, Mohamed-Faouzi; Djeghaba, M.

In: IFAC-PapersOnLine, Vol. 28, No. 21, 01.09.2015, p. 1408-1413.

Research output: Contribution to journalConference article

@article{37b2447bbe344e4483bcb22bdb0ab1f7,
title = "Neuronal principal component analysis for nonlinear time-varying processes monitoring",
abstract = "This paper presents a new adaptive kernel principal component analysis algorithm (AKPCA) for nonlinear time-varying process monitoring. The basic idea is to use a neuronal principal component analysis based on the kernel version of the generalized Hebbian algorithm. The proposed algorithm follows a new methodology to update the KPCA model. At each time instant, when a new data is available, the KPCA model is updated accordingly without having to re-explore all previous data. By using the proposed algorithm, the performance of process monitoring is improved in two aspects; the speed computation and adaptation of the KPCA model, and the storage memory complexity. To identify faults in a dynamic process, the reconstruction based contribution approach is used and adapted in real time. The results for applying this algorithm on the Tennessee Eastman process shows its feasibility and advantageous performances.",
keywords = "Dynamic process, Fault detection and isolation, Kernel principal component analysis",
author = "C. Chakour and Mohamed-Faouzi Harkat and M. Djeghaba",
year = "2015",
month = "9",
day = "1",
doi = "10.1016/j.ifacol.2015.09.722",
language = "English",
volume = "28",
pages = "1408--1413",
journal = "IFAC-PapersOnLine",
issn = "2405-8963",
publisher = "IFAC Secretariat",
number = "21",

}

TY - JOUR

T1 - Neuronal principal component analysis for nonlinear time-varying processes monitoring

AU - Chakour, C.

AU - Harkat, Mohamed-Faouzi

AU - Djeghaba, M.

PY - 2015/9/1

Y1 - 2015/9/1

N2 - This paper presents a new adaptive kernel principal component analysis algorithm (AKPCA) for nonlinear time-varying process monitoring. The basic idea is to use a neuronal principal component analysis based on the kernel version of the generalized Hebbian algorithm. The proposed algorithm follows a new methodology to update the KPCA model. At each time instant, when a new data is available, the KPCA model is updated accordingly without having to re-explore all previous data. By using the proposed algorithm, the performance of process monitoring is improved in two aspects; the speed computation and adaptation of the KPCA model, and the storage memory complexity. To identify faults in a dynamic process, the reconstruction based contribution approach is used and adapted in real time. The results for applying this algorithm on the Tennessee Eastman process shows its feasibility and advantageous performances.

AB - This paper presents a new adaptive kernel principal component analysis algorithm (AKPCA) for nonlinear time-varying process monitoring. The basic idea is to use a neuronal principal component analysis based on the kernel version of the generalized Hebbian algorithm. The proposed algorithm follows a new methodology to update the KPCA model. At each time instant, when a new data is available, the KPCA model is updated accordingly without having to re-explore all previous data. By using the proposed algorithm, the performance of process monitoring is improved in two aspects; the speed computation and adaptation of the KPCA model, and the storage memory complexity. To identify faults in a dynamic process, the reconstruction based contribution approach is used and adapted in real time. The results for applying this algorithm on the Tennessee Eastman process shows its feasibility and advantageous performances.

KW - Dynamic process

KW - Fault detection and isolation

KW - Kernel principal component analysis

UR - http://www.scopus.com/inward/record.url?scp=84992521968&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84992521968&partnerID=8YFLogxK

U2 - 10.1016/j.ifacol.2015.09.722

DO - 10.1016/j.ifacol.2015.09.722

M3 - Conference article

AN - SCOPUS:84992521968

VL - 28

SP - 1408

EP - 1413

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

SN - 2405-8963

IS - 21

ER -