An agent-based implementation of hidden Markov models for gas turbine condition monitoring

Andrew Kenyon, Victoria Catterson, Stephen McArthur, John Twiddle

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

This paper considers the use of a multi-agent system (MAS) incorporating hidden Markov models (HMMs) for the condition monitoring of gas turbine (GT) engines. Hidden Markov models utilizing a Gaussian probability distribution are proposed as an anomaly detection tool for gas turbines components. The use of this technique is shown to allow the modeling of the dynamics of GTs despite a lack of high frequency data. This allows the early detection of developing faults and avoids costly outages due to asset failure. These models are implemented as part of a MAS, using a proposed extension of an established power system ontology, for fault detection of gas turbines. The multi-agent system is shown to be applicable through a case study and comparison to an existing system utilizing historic data from a combined-cycle gas turbine plant provided by an industrial partner.
LanguageEnglish
Pages186-195
Number of pages10
JournalIEEE Transactions on Systems Man and Cybernetics: Systems
Volume44
Issue number2
Early online date14 Jun 2013
DOIs
Publication statusPublished - Feb 2014

Fingerprint

Condition monitoring
Hidden Markov models
Gas turbines
Multi agent systems
Turbine components
Fault detection
Outages
Probability distributions
Ontology
Turbines

Keywords

  • Markov models
  • gas turbine
  • multi agent system
  • hidden Markov models
  • multi-agent systems

Cite this

@article{047194a0468d493ba00504858c06e81c,
title = "An agent-based implementation of hidden Markov models for gas turbine condition monitoring",
abstract = "This paper considers the use of a multi-agent system (MAS) incorporating hidden Markov models (HMMs) for the condition monitoring of gas turbine (GT) engines. Hidden Markov models utilizing a Gaussian probability distribution are proposed as an anomaly detection tool for gas turbines components. The use of this technique is shown to allow the modeling of the dynamics of GTs despite a lack of high frequency data. This allows the early detection of developing faults and avoids costly outages due to asset failure. These models are implemented as part of a MAS, using a proposed extension of an established power system ontology, for fault detection of gas turbines. The multi-agent system is shown to be applicable through a case study and comparison to an existing system utilizing historic data from a combined-cycle gas turbine plant provided by an industrial partner.",
keywords = "Markov models, gas turbine, multi agent system, hidden Markov models, multi-agent systems",
author = "Andrew Kenyon and Victoria Catterson and Stephen McArthur and John Twiddle",
year = "2014",
month = "2",
doi = "10.1109/TSMC.2013.2251539",
language = "English",
volume = "44",
pages = "186--195",
journal = "IEEE Transactions on Systems Man and Cybernetics: Systems",
issn = "2168-2216",
publisher = "IEEE Advancing Technology for Humanity",
number = "2",

}

An agent-based implementation of hidden Markov models for gas turbine condition monitoring. / Kenyon, Andrew; Catterson, Victoria; McArthur, Stephen; Twiddle, John.

In: IEEE Transactions on Systems Man and Cybernetics: Systems, Vol. 44, No. 2, 02.2014, p. 186-195 .

Research output: Contribution to journalArticle

TY - JOUR

T1 - An agent-based implementation of hidden Markov models for gas turbine condition monitoring

AU - Kenyon, Andrew

AU - Catterson, Victoria

AU - McArthur, Stephen

AU - Twiddle, John

PY - 2014/2

Y1 - 2014/2

N2 - This paper considers the use of a multi-agent system (MAS) incorporating hidden Markov models (HMMs) for the condition monitoring of gas turbine (GT) engines. Hidden Markov models utilizing a Gaussian probability distribution are proposed as an anomaly detection tool for gas turbines components. The use of this technique is shown to allow the modeling of the dynamics of GTs despite a lack of high frequency data. This allows the early detection of developing faults and avoids costly outages due to asset failure. These models are implemented as part of a MAS, using a proposed extension of an established power system ontology, for fault detection of gas turbines. The multi-agent system is shown to be applicable through a case study and comparison to an existing system utilizing historic data from a combined-cycle gas turbine plant provided by an industrial partner.

AB - This paper considers the use of a multi-agent system (MAS) incorporating hidden Markov models (HMMs) for the condition monitoring of gas turbine (GT) engines. Hidden Markov models utilizing a Gaussian probability distribution are proposed as an anomaly detection tool for gas turbines components. The use of this technique is shown to allow the modeling of the dynamics of GTs despite a lack of high frequency data. This allows the early detection of developing faults and avoids costly outages due to asset failure. These models are implemented as part of a MAS, using a proposed extension of an established power system ontology, for fault detection of gas turbines. The multi-agent system is shown to be applicable through a case study and comparison to an existing system utilizing historic data from a combined-cycle gas turbine plant provided by an industrial partner.

KW - Markov models

KW - gas turbine

KW - multi agent system

KW - hidden Markov models

KW - multi-agent systems

UR - http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=5326

U2 - 10.1109/TSMC.2013.2251539

DO - 10.1109/TSMC.2013.2251539

M3 - Article

VL - 44

SP - 186

EP - 195

JO - IEEE Transactions on Systems Man and Cybernetics: Systems

T2 - IEEE Transactions on Systems Man and Cybernetics: Systems

JF - IEEE Transactions on Systems Man and Cybernetics: Systems

SN - 2168-2216

IS - 2

ER -