On-line Alert Systems for Production Plants: A Conflict Based Approach

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

4 Citationer (Scopus)
300 Downloads (Pure)

Resumé

We present a new methodology for detecting faults and abnormal behavior in production plants. The methodology stems from a joint project with a Danish energy consortium. During the course of the project we encountered several problems that we believe are common for projects of this type. Most notably, there was a lack of both knowledge and data concerning possible faults, and it therefore turned out to be infeasible to learn/construct a standard classification model for doing fault detection. As an alternative we propose a method for doing on-line fault detection using only a model of normal system operation. Faults are detected by measuring the conflict between the model and the sensor readings, and knowledge about the possible faults is therefore not required. We illustrate the proposed method using real-world data from a coal driven power plant as well as simulated data from an oil production facility.
Udgivelsesdato: JUL
OriginalsprogEngelsk
TidsskriftInternational Journal of Approximate Reasoning
Vol/bind45
Udgave nummer2
Sider (fra-til)255-270
ISSN0888-613X
DOI
StatusUdgivet - 2007

Fingerprint

Online systems
Fault
Fault detection
Fault Detection
Methodology
Power plants
Power Plant
Coal
Sensors
Model
Sensor
Conflict
Alternatives
Energy
Knowledge

Citer dette

@article{8fbd8d209c2c11db8ed6000ea68e967b,
title = "On-line Alert Systems for Production Plants: A Conflict Based Approach",
abstract = "We present a new methodology for detecting faults and abnormal behavior in production plants. The methodology stems from a joint project with a Danish energy consortium. During the course of the project we encountered several problems that we believe are common for projects of this type. Most notably, there was a lack of both knowledge and data concerning possible faults, and it therefore turned out to be infeasible to learn/construct a standard classification model for doing fault detection. As an alternative we propose a method for doing on-line fault detection using only a model of normal system operation. Faults are detected by measuring the conflict between the model and the sensor readings, and knowledge about the possible faults is therefore not required. We illustrate the proposed method using real-world data from a coal driven power plant as well as simulated data from an oil production facility.",
author = "Nielsen, {Thomas Dyhre} and Jensen, {Finn V.}",
year = "2007",
doi = "10.1016/j.ijar.2006.06.010",
language = "English",
volume = "45",
pages = "255--270",
journal = "International Journal of Approximate Reasoning",
issn = "0888-613X",
publisher = "Elsevier",
number = "2",

}

On-line Alert Systems for Production Plants: A Conflict Based Approach. / Nielsen, Thomas Dyhre; Jensen, Finn V.

I: International Journal of Approximate Reasoning, Bind 45, Nr. 2, 2007, s. 255-270.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - On-line Alert Systems for Production Plants: A Conflict Based Approach

AU - Nielsen, Thomas Dyhre

AU - Jensen, Finn V.

PY - 2007

Y1 - 2007

N2 - We present a new methodology for detecting faults and abnormal behavior in production plants. The methodology stems from a joint project with a Danish energy consortium. During the course of the project we encountered several problems that we believe are common for projects of this type. Most notably, there was a lack of both knowledge and data concerning possible faults, and it therefore turned out to be infeasible to learn/construct a standard classification model for doing fault detection. As an alternative we propose a method for doing on-line fault detection using only a model of normal system operation. Faults are detected by measuring the conflict between the model and the sensor readings, and knowledge about the possible faults is therefore not required. We illustrate the proposed method using real-world data from a coal driven power plant as well as simulated data from an oil production facility.

AB - We present a new methodology for detecting faults and abnormal behavior in production plants. The methodology stems from a joint project with a Danish energy consortium. During the course of the project we encountered several problems that we believe are common for projects of this type. Most notably, there was a lack of both knowledge and data concerning possible faults, and it therefore turned out to be infeasible to learn/construct a standard classification model for doing fault detection. As an alternative we propose a method for doing on-line fault detection using only a model of normal system operation. Faults are detected by measuring the conflict between the model and the sensor readings, and knowledge about the possible faults is therefore not required. We illustrate the proposed method using real-world data from a coal driven power plant as well as simulated data from an oil production facility.

U2 - 10.1016/j.ijar.2006.06.010

DO - 10.1016/j.ijar.2006.06.010

M3 - Journal article

VL - 45

SP - 255

EP - 270

JO - International Journal of Approximate Reasoning

JF - International Journal of Approximate Reasoning

SN - 0888-613X

IS - 2

ER -