Statistical fault detection using PCA-based GLR hypothesis testing

Fouzi Harrou, Mohamed Nounou, Hazem Nounou, Muddu Madakyaru

Research output: Contribution to journalArticle

48 Citations (Scopus)

Abstract

Safe process operation requires effective fault detection (FD) methods that can identify faults in various process parameters. In the absence of a process model, principal component analysis (PCA) has been successfully used as a data-based FD technique for highly correlated process variables. Some of the PCA detection indices include the T2 or Q statistics, which have their advantages and disadvantages. When a process model is available, however, the generalized likelihood ratio (GLR) test, which is a statistical hypothesis testing method, has shown good fault detection abilities. In this work, a PCA-based GLR fault detection algorithm is developed to exploit the advantages of the GLR test in the absence of a process model. In fact, PCA is used to provide a modeling framework for the develop fault detection algorithm. The PCA-based GLR fault detection algorithm provides optimal properties by maximizing the detection probability of faults for a given false alarm rate. The performance of the PCA-based GLR fault detection algorithm is illustrated and compared to conventional fault detection methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of these examples clearly show the effectiveness of the developed algorithm over conventional methods.

Original languageEnglish
Pages (from-to)129-139
Number of pages11
JournalJournal of Loss Prevention in the Process Industries
Volume26
Issue number1
DOIs
Publication statusPublished - Jan 2013

Fingerprint

Principal Component Analysis
Fault detection
Principal component analysis
principal component analysis
Testing
testing
Hypothesis testing
Likelihood ratio
methodology
Statistics
statistics
Process model

Keywords

  • Fault detection for process safety
  • GLR test
  • Principal component analysis
  • Statistical hypothesis testing

ASJC Scopus subject areas

  • Food Science
  • Energy Engineering and Power Technology
  • Chemical Engineering(all)
  • Management Science and Operations Research
  • Safety, Risk, Reliability and Quality
  • Control and Systems Engineering
  • Industrial and Manufacturing Engineering

Cite this

Statistical fault detection using PCA-based GLR hypothesis testing. / Harrou, Fouzi; Nounou, Mohamed; Nounou, Hazem; Madakyaru, Muddu.

In: Journal of Loss Prevention in the Process Industries, Vol. 26, No. 1, 01.2013, p. 129-139.

Research output: Contribution to journalArticle

@article{ffca290a3b0f4808b67b8073e80fdf3e,
title = "Statistical fault detection using PCA-based GLR hypothesis testing",
abstract = "Safe process operation requires effective fault detection (FD) methods that can identify faults in various process parameters. In the absence of a process model, principal component analysis (PCA) has been successfully used as a data-based FD technique for highly correlated process variables. Some of the PCA detection indices include the T2 or Q statistics, which have their advantages and disadvantages. When a process model is available, however, the generalized likelihood ratio (GLR) test, which is a statistical hypothesis testing method, has shown good fault detection abilities. In this work, a PCA-based GLR fault detection algorithm is developed to exploit the advantages of the GLR test in the absence of a process model. In fact, PCA is used to provide a modeling framework for the develop fault detection algorithm. The PCA-based GLR fault detection algorithm provides optimal properties by maximizing the detection probability of faults for a given false alarm rate. The performance of the PCA-based GLR fault detection algorithm is illustrated and compared to conventional fault detection methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of these examples clearly show the effectiveness of the developed algorithm over conventional methods.",
keywords = "Fault detection for process safety, GLR test, Principal component analysis, Statistical hypothesis testing",
author = "Fouzi Harrou and Mohamed Nounou and Hazem Nounou and Muddu Madakyaru",
year = "2013",
month = "1",
doi = "10.1016/j.jlp.2012.10.003",
language = "English",
volume = "26",
pages = "129--139",
journal = "Journal of Loss Prevention in the Process Industries",
issn = "0950-4230",
publisher = "Elsevier BV",
number = "1",

}

TY - JOUR

T1 - Statistical fault detection using PCA-based GLR hypothesis testing

AU - Harrou, Fouzi

AU - Nounou, Mohamed

AU - Nounou, Hazem

AU - Madakyaru, Muddu

PY - 2013/1

Y1 - 2013/1

N2 - Safe process operation requires effective fault detection (FD) methods that can identify faults in various process parameters. In the absence of a process model, principal component analysis (PCA) has been successfully used as a data-based FD technique for highly correlated process variables. Some of the PCA detection indices include the T2 or Q statistics, which have their advantages and disadvantages. When a process model is available, however, the generalized likelihood ratio (GLR) test, which is a statistical hypothesis testing method, has shown good fault detection abilities. In this work, a PCA-based GLR fault detection algorithm is developed to exploit the advantages of the GLR test in the absence of a process model. In fact, PCA is used to provide a modeling framework for the develop fault detection algorithm. The PCA-based GLR fault detection algorithm provides optimal properties by maximizing the detection probability of faults for a given false alarm rate. The performance of the PCA-based GLR fault detection algorithm is illustrated and compared to conventional fault detection methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of these examples clearly show the effectiveness of the developed algorithm over conventional methods.

AB - Safe process operation requires effective fault detection (FD) methods that can identify faults in various process parameters. In the absence of a process model, principal component analysis (PCA) has been successfully used as a data-based FD technique for highly correlated process variables. Some of the PCA detection indices include the T2 or Q statistics, which have their advantages and disadvantages. When a process model is available, however, the generalized likelihood ratio (GLR) test, which is a statistical hypothesis testing method, has shown good fault detection abilities. In this work, a PCA-based GLR fault detection algorithm is developed to exploit the advantages of the GLR test in the absence of a process model. In fact, PCA is used to provide a modeling framework for the develop fault detection algorithm. The PCA-based GLR fault detection algorithm provides optimal properties by maximizing the detection probability of faults for a given false alarm rate. The performance of the PCA-based GLR fault detection algorithm is illustrated and compared to conventional fault detection methods through two simulated examples, one using synthetic data and the other using simulated continuously stirred tank reactor (CSTR) data. The results of these examples clearly show the effectiveness of the developed algorithm over conventional methods.

KW - Fault detection for process safety

KW - GLR test

KW - Principal component analysis

KW - Statistical hypothesis testing

UR - http://www.scopus.com/inward/record.url?scp=84872500828&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872500828&partnerID=8YFLogxK

U2 - 10.1016/j.jlp.2012.10.003

DO - 10.1016/j.jlp.2012.10.003

M3 - Article

VL - 26

SP - 129

EP - 139

JO - Journal of Loss Prevention in the Process Industries

JF - Journal of Loss Prevention in the Process Industries

SN - 0950-4230

IS - 1

ER -