On the Application of PCA Technique to Fault Diagnosis

S. Ding; P. Zhang; E. Ding; S. Yin; A. Naik; P. Deng; W. Gui

doi:10.1016/S1007-0214(10)70043-2

On the Application of PCA Technique to Fault Diagnosis

S. Ding^*
, P. Zhang
, E. Ding
, S. Yin
, A. Naik
, P. Deng
, W. Gui

^*Corresponding author for this work

University of Duisburg-Essen
Gelsenkirchen University of Applied Sciences
Central South University

Research output: Contribution to journal › Article › peer-review

171 Scopus citations

Abstract

In this paper, we briefly address the application of the standard principal component analysis (PCA) technique to fault detection and identification. Based on an analysis of the existing test statistic, we propose a new test statistic, which is similar to the Hawkin's T_H² statistic but without the numerical drawback. In comparison with the SPE index, the threshold setting associated with the new statistic is computationally simpler. Our further study is dedicated to the analysis of fault sensitivity. We consider the off-set and scaling faults, and evaluate the test statistic by viewing its sensitivity to the faults. Our final study focuses on identifying off-set and scaling faults. To this end, two algorithms are proposed. This paper also includes some critical remarks on the application of the PCA technique to fault diagnosis.

Original language	English
Pages (from-to)	138-144
Number of pages	7
Journal	Tsinghua Science and Technology
Volume	15
Issue number	2
DOIs	https://doi.org/10.1016/S1007-0214(10)70043-2
State	Published - Apr 2010
Externally published	Yes

Keywords

fault diagnosis
multivariate analysis
principal component analysis (PCA)
process monitoring

Access to Document

10.1016/S1007-0214(10)70043-2

Cite this

@article{d12c37defe134ff0a99816d3f0c0736f,

title = "On the Application of PCA Technique to Fault Diagnosis",

abstract = "In this paper, we briefly address the application of the standard principal component analysis (PCA) technique to fault detection and identification. Based on an analysis of the existing test statistic, we propose a new test statistic, which is similar to the Hawkin's TH2 statistic but without the numerical drawback. In comparison with the SPE index, the threshold setting associated with the new statistic is computationally simpler. Our further study is dedicated to the analysis of fault sensitivity. We consider the off-set and scaling faults, and evaluate the test statistic by viewing its sensitivity to the faults. Our final study focuses on identifying off-set and scaling faults. To this end, two algorithms are proposed. This paper also includes some critical remarks on the application of the PCA technique to fault diagnosis.",

keywords = "fault diagnosis, multivariate analysis, principal component analysis (PCA), process monitoring",

author = "S. Ding and P. Zhang and E. Ding and S. Yin and A. Naik and P. Deng and W. Gui",

year = "2010",

month = apr,

doi = "10.1016/S1007-0214(10)70043-2",

language = "英语",

volume = "15",

pages = "138--144",

journal = "Tsinghua Science and Technology",

issn = "1007-0214",

publisher = "Tsinghua University Press",

number = "2",

}

TY - JOUR

T1 - On the Application of PCA Technique to Fault Diagnosis

AU - Ding, S.

AU - Zhang, P.

AU - Ding, E.

AU - Yin, S.

AU - Naik, A.

AU - Deng, P.

AU - Gui, W.

PY - 2010/4

Y1 - 2010/4

N2 - In this paper, we briefly address the application of the standard principal component analysis (PCA) technique to fault detection and identification. Based on an analysis of the existing test statistic, we propose a new test statistic, which is similar to the Hawkin's TH2 statistic but without the numerical drawback. In comparison with the SPE index, the threshold setting associated with the new statistic is computationally simpler. Our further study is dedicated to the analysis of fault sensitivity. We consider the off-set and scaling faults, and evaluate the test statistic by viewing its sensitivity to the faults. Our final study focuses on identifying off-set and scaling faults. To this end, two algorithms are proposed. This paper also includes some critical remarks on the application of the PCA technique to fault diagnosis.

AB - In this paper, we briefly address the application of the standard principal component analysis (PCA) technique to fault detection and identification. Based on an analysis of the existing test statistic, we propose a new test statistic, which is similar to the Hawkin's TH2 statistic but without the numerical drawback. In comparison with the SPE index, the threshold setting associated with the new statistic is computationally simpler. Our further study is dedicated to the analysis of fault sensitivity. We consider the off-set and scaling faults, and evaluate the test statistic by viewing its sensitivity to the faults. Our final study focuses on identifying off-set and scaling faults. To this end, two algorithms are proposed. This paper also includes some critical remarks on the application of the PCA technique to fault diagnosis.

KW - fault diagnosis

KW - multivariate analysis

KW - principal component analysis (PCA)

KW - process monitoring

UR - https://www.scopus.com/pages/publications/77950342431

U2 - 10.1016/S1007-0214(10)70043-2

DO - 10.1016/S1007-0214(10)70043-2

M3 - 文章

AN - SCOPUS:77950342431

SN - 1007-0214

VL - 15

SP - 138

EP - 144

JO - Tsinghua Science and Technology

JF - Tsinghua Science and Technology

IS - 2

ER -

On the Application of PCA Technique to Fault Diagnosis

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this