Resumé

To optimally plan maintenance of wind turbine blades, knowledge of the degradation processes and the remaining useful life is essential. In this paper, a method is proposed for calibration of a Markov deterioration model based on past inspection data for a range of blades, and updating of the model for a specific wind turbine blade, whenever information is available from inspections
and/or condition monitoring. Dynamic Bayesian networks are used to obtain probabilities of inspection outcomes for a maximum likelihood estimation of the transition probabilities in the Markov model, and are used again when updating the model for a specific blade using observations. The method is illustrated using indicative data from a database containing data from inspections of wind turbine blades.
OriginalsprogEngelsk
Artikelnummer664
TidsskriftEnergies
Vol/bind10
Udgave nummer5
Antal sider13
ISSN1996-1073
DOI
StatusUdgivet - 2017

Fingerprint

Turbine Blade
Wind Turbine
Bayesian Estimation
Wind turbines
Turbomachine blades
Inspection
Blade
Updating
Dynamic Bayesian Networks
Condition Monitoring
Deterioration
Transition Probability
Maximum Likelihood Estimation
Markov Model
Maximum likelihood estimation
Maintenance
Condition monitoring
Degradation
Calibration
Bayesian networks

Emneord

  • Remaining useful life
  • Wind turbine blades
  • Hidden Markov model
  • Dynamic Bayesian networks

Citer dette

@article{4b06ffe5902a4af0beb5fe3412fc355a,
title = "Bayesian Estimation of Remaining Useful Life for Wind Turbine Blades",
abstract = "To optimally plan maintenance of wind turbine blades, knowledge of the degradation processes and the remaining useful life is essential. In this paper, a method is proposed for calibration of a Markov deterioration model based on past inspection data for a range of blades, and updating of the model for a specific wind turbine blade, whenever information is available from inspectionsand/or condition monitoring. Dynamic Bayesian networks are used to obtain probabilities of inspection outcomes for a maximum likelihood estimation of the transition probabilities in the Markov model, and are used again when updating the model for a specific blade using observations. The method is illustrated using indicative data from a database containing data from inspections of wind turbine blades.",
keywords = "Remaining useful life, Wind turbine blades, Hidden Markov model, Dynamic Bayesian networks, Remaining useful life, Wind turbine blades, Hidden Markov model, Dynamic Bayesian networks",
author = "Nielsen, {Jannie S{\o}nderk{\ae}r} and S{\o}rensen, {John Dalsgaard}",
year = "2017",
doi = "10.3390/en10050664",
language = "English",
volume = "10",
journal = "Energies",
issn = "1996-1073",
publisher = "M D P I AG",
number = "5",

}

Bayesian Estimation of Remaining Useful Life for Wind Turbine Blades. / Nielsen, Jannie Sønderkær; Sørensen, John Dalsgaard.

I: Energies, Bind 10, Nr. 5, 664, 2017.

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

TY - JOUR

T1 - Bayesian Estimation of Remaining Useful Life for Wind Turbine Blades

AU - Nielsen, Jannie Sønderkær

AU - Sørensen, John Dalsgaard

PY - 2017

Y1 - 2017

N2 - To optimally plan maintenance of wind turbine blades, knowledge of the degradation processes and the remaining useful life is essential. In this paper, a method is proposed for calibration of a Markov deterioration model based on past inspection data for a range of blades, and updating of the model for a specific wind turbine blade, whenever information is available from inspectionsand/or condition monitoring. Dynamic Bayesian networks are used to obtain probabilities of inspection outcomes for a maximum likelihood estimation of the transition probabilities in the Markov model, and are used again when updating the model for a specific blade using observations. The method is illustrated using indicative data from a database containing data from inspections of wind turbine blades.

AB - To optimally plan maintenance of wind turbine blades, knowledge of the degradation processes and the remaining useful life is essential. In this paper, a method is proposed for calibration of a Markov deterioration model based on past inspection data for a range of blades, and updating of the model for a specific wind turbine blade, whenever information is available from inspectionsand/or condition monitoring. Dynamic Bayesian networks are used to obtain probabilities of inspection outcomes for a maximum likelihood estimation of the transition probabilities in the Markov model, and are used again when updating the model for a specific blade using observations. The method is illustrated using indicative data from a database containing data from inspections of wind turbine blades.

KW - Remaining useful life

KW - Wind turbine blades

KW - Hidden Markov model

KW - Dynamic Bayesian networks

KW - Remaining useful life

KW - Wind turbine blades

KW - Hidden Markov model

KW - Dynamic Bayesian networks

U2 - 10.3390/en10050664

DO - 10.3390/en10050664

M3 - Journal article

VL - 10

JO - Energies

JF - Energies

SN - 1996-1073

IS - 5

M1 - 664

ER -