Fault Detection and Diagnosis in a Continuous Process with Uncertainly Normal Situations

Yoshifumi Tsuge, Kazushi Mori, Shigeyuki Tateno, Hisayoshi Matsuyama

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Fault detection and diagnosis of a continuous process with input fluctuations such as load fluctuations can not be simply performed because of the difficulty of detecting the abnormality in the process, where normal values of state variables are uncertainly time-varying due to input fluctuations. In this study, normal values of output variables are estimated by the auto regressive exogenous model (ARX model). The measured value of an output variable is classified as 5-range signs (+, +?, 0, -?, -) by simultaneously performing two sequential probability ratio tests (SPRTs) based on the error residual between estimated and measured values : one test examines whether it is normal or higher than normal, the other examines whether it is normal or lower than normal. A combination of signs given to all the output variables is called a “pattern,” which is considered to represent an abnormal situation occurring in the process. Then, the fault diagnosis algorithm based on signed directed graph can deduce the fault origin that causes the pattern. The effectiveness of this approach is demonstrated through experiments with a tank-pipeline system.

Original languageEnglish
Pages (from-to)565-572
Number of pages8
Journalkagaku kogaku ronbunshu
Volume21
Issue number3
DOIs
Publication statusPublished - Jan 1 1995

Fingerprint

Fault detection
Failure analysis
Directed graphs
Pipelines
Experiments

All Science Journal Classification (ASJC) codes

  • Chemistry(all)
  • Chemical Engineering(all)

Cite this

Fault Detection and Diagnosis in a Continuous Process with Uncertainly Normal Situations. / Tsuge, Yoshifumi; Mori, Kazushi; Tateno, Shigeyuki; Matsuyama, Hisayoshi.

In: kagaku kogaku ronbunshu, Vol. 21, No. 3, 01.01.1995, p. 565-572.

Research output: Contribution to journalArticle

Tsuge, Yoshifumi ; Mori, Kazushi ; Tateno, Shigeyuki ; Matsuyama, Hisayoshi. / Fault Detection and Diagnosis in a Continuous Process with Uncertainly Normal Situations. In: kagaku kogaku ronbunshu. 1995 ; Vol. 21, No. 3. pp. 565-572.
@article{959b8b6bf12b4ab6ba956f3de3e10d6a,
title = "Fault Detection and Diagnosis in a Continuous Process with Uncertainly Normal Situations",
abstract = "Fault detection and diagnosis of a continuous process with input fluctuations such as load fluctuations can not be simply performed because of the difficulty of detecting the abnormality in the process, where normal values of state variables are uncertainly time-varying due to input fluctuations. In this study, normal values of output variables are estimated by the auto regressive exogenous model (ARX model). The measured value of an output variable is classified as 5-range signs (+, +?, 0, -?, -) by simultaneously performing two sequential probability ratio tests (SPRTs) based on the error residual between estimated and measured values : one test examines whether it is normal or higher than normal, the other examines whether it is normal or lower than normal. A combination of signs given to all the output variables is called a “pattern,” which is considered to represent an abnormal situation occurring in the process. Then, the fault diagnosis algorithm based on signed directed graph can deduce the fault origin that causes the pattern. The effectiveness of this approach is demonstrated through experiments with a tank-pipeline system.",
author = "Yoshifumi Tsuge and Kazushi Mori and Shigeyuki Tateno and Hisayoshi Matsuyama",
year = "1995",
month = "1",
day = "1",
doi = "10.1252/kakoronbunshu.21.565",
language = "English",
volume = "21",
pages = "565--572",
journal = "Kagaku Kogaku Ronbunshu",
issn = "0386-216X",
publisher = "The Society of Chemical Engineers, Japan",
number = "3",

}

TY - JOUR

T1 - Fault Detection and Diagnosis in a Continuous Process with Uncertainly Normal Situations

AU - Tsuge, Yoshifumi

AU - Mori, Kazushi

AU - Tateno, Shigeyuki

AU - Matsuyama, Hisayoshi

PY - 1995/1/1

Y1 - 1995/1/1

N2 - Fault detection and diagnosis of a continuous process with input fluctuations such as load fluctuations can not be simply performed because of the difficulty of detecting the abnormality in the process, where normal values of state variables are uncertainly time-varying due to input fluctuations. In this study, normal values of output variables are estimated by the auto regressive exogenous model (ARX model). The measured value of an output variable is classified as 5-range signs (+, +?, 0, -?, -) by simultaneously performing two sequential probability ratio tests (SPRTs) based on the error residual between estimated and measured values : one test examines whether it is normal or higher than normal, the other examines whether it is normal or lower than normal. A combination of signs given to all the output variables is called a “pattern,” which is considered to represent an abnormal situation occurring in the process. Then, the fault diagnosis algorithm based on signed directed graph can deduce the fault origin that causes the pattern. The effectiveness of this approach is demonstrated through experiments with a tank-pipeline system.

AB - Fault detection and diagnosis of a continuous process with input fluctuations such as load fluctuations can not be simply performed because of the difficulty of detecting the abnormality in the process, where normal values of state variables are uncertainly time-varying due to input fluctuations. In this study, normal values of output variables are estimated by the auto regressive exogenous model (ARX model). The measured value of an output variable is classified as 5-range signs (+, +?, 0, -?, -) by simultaneously performing two sequential probability ratio tests (SPRTs) based on the error residual between estimated and measured values : one test examines whether it is normal or higher than normal, the other examines whether it is normal or lower than normal. A combination of signs given to all the output variables is called a “pattern,” which is considered to represent an abnormal situation occurring in the process. Then, the fault diagnosis algorithm based on signed directed graph can deduce the fault origin that causes the pattern. The effectiveness of this approach is demonstrated through experiments with a tank-pipeline system.

UR - http://www.scopus.com/inward/record.url?scp=33646335894&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646335894&partnerID=8YFLogxK

U2 - 10.1252/kakoronbunshu.21.565

DO - 10.1252/kakoronbunshu.21.565

M3 - Article

AN - SCOPUS:33646335894

VL - 21

SP - 565

EP - 572

JO - Kagaku Kogaku Ronbunshu

JF - Kagaku Kogaku Ronbunshu

SN - 0386-216X

IS - 3

ER -