Copula based model for wind turbine power curve outlier rejection

Research output: Contribution to journalArticle

22 Citations (Scopus)

Abstract

Power curve measurements provide a conventional and effective means of assessing the performance of a wind turbine, both commercially and technically. Increasingly high wind penetration in power systems and offshore accessibility issues make it even more important to monitor the condition and performance of wind turbines based on timely and accurate wind speed and power measurements. Power curve data from Supervisory Control and Data Acquisition (SCADA) system records, however, often contains significant measurement deviations, which are commonly produced as a consequence of wind turbine operational transitions rather than stemming from physical degradation of the plant. To use this raw data for wind turbine condition monitoring purposes is thus likely to lead to high false alarm rates which would make the actual fault detection unreliable and potentially add unnecessarily to the costs of maintenance. To this end, this paper proposes a probabilistic outlier exclusion method developed around a Copula based joint probability model. This approach has the capability of capturing the complex multivariate nonlinear relation between parameters based on their univariate marginal distributions through the use of Copula; data points that deviate significantly from the consolidated power curve can then be removed depending on this derived joint probability distribution. After data filtering in this manner, it is shown how the resulting power curves are better defined and less subject to uncertainty, whilst broadly retaining the same dominant statistical characteristics. These improved power curves make subsequent condition monitoring more effective in the reliable detection of faults.
LanguageEnglish
Pages1677–1688
Number of pages12
JournalWind Energy
Volume17
Issue number11
Early online date27 Aug 2013
DOIs
Publication statusPublished - 1 Nov 2014

Fingerprint

Wind turbines
Condition monitoring
SCADA systems
Fault detection
Probability distributions
Degradation
Costs

Keywords

  • wind energy
  • wind turbine
  • power systems
  • power curve
  • outlier rejection
  • SCADA
  • copula model

Cite this

@article{4b49edd7d0cf470994dafcc4bbd195e5,
title = "Copula based model for wind turbine power curve outlier rejection",
abstract = "Power curve measurements provide a conventional and effective means of assessing the performance of a wind turbine, both commercially and technically. Increasingly high wind penetration in power systems and offshore accessibility issues make it even more important to monitor the condition and performance of wind turbines based on timely and accurate wind speed and power measurements. Power curve data from Supervisory Control and Data Acquisition (SCADA) system records, however, often contains significant measurement deviations, which are commonly produced as a consequence of wind turbine operational transitions rather than stemming from physical degradation of the plant. To use this raw data for wind turbine condition monitoring purposes is thus likely to lead to high false alarm rates which would make the actual fault detection unreliable and potentially add unnecessarily to the costs of maintenance. To this end, this paper proposes a probabilistic outlier exclusion method developed around a Copula based joint probability model. This approach has the capability of capturing the complex multivariate nonlinear relation between parameters based on their univariate marginal distributions through the use of Copula; data points that deviate significantly from the consolidated power curve can then be removed depending on this derived joint probability distribution. After data filtering in this manner, it is shown how the resulting power curves are better defined and less subject to uncertainty, whilst broadly retaining the same dominant statistical characteristics. These improved power curves make subsequent condition monitoring more effective in the reliable detection of faults.",
keywords = "wind energy, wind turbine, power systems, power curve, outlier rejection, SCADA, copula model",
author = "Yue Wang and Infield, {David G.} and Bruce Stephen and Galloway, {Stuart J.}",
year = "2014",
month = "11",
day = "1",
doi = "10.1002/we.1661",
language = "English",
volume = "17",
pages = "1677–1688",
journal = "Wind Energy",
issn = "1095-4244",
number = "11",

}

Copula based model for wind turbine power curve outlier rejection. / Wang, Yue; Infield, David G.; Stephen, Bruce; Galloway, Stuart J.

In: Wind Energy, Vol. 17, No. 11, 01.11.2014, p. 1677–1688.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Copula based model for wind turbine power curve outlier rejection

AU - Wang, Yue

AU - Infield, David G.

AU - Stephen, Bruce

AU - Galloway, Stuart J.

PY - 2014/11/1

Y1 - 2014/11/1

N2 - Power curve measurements provide a conventional and effective means of assessing the performance of a wind turbine, both commercially and technically. Increasingly high wind penetration in power systems and offshore accessibility issues make it even more important to monitor the condition and performance of wind turbines based on timely and accurate wind speed and power measurements. Power curve data from Supervisory Control and Data Acquisition (SCADA) system records, however, often contains significant measurement deviations, which are commonly produced as a consequence of wind turbine operational transitions rather than stemming from physical degradation of the plant. To use this raw data for wind turbine condition monitoring purposes is thus likely to lead to high false alarm rates which would make the actual fault detection unreliable and potentially add unnecessarily to the costs of maintenance. To this end, this paper proposes a probabilistic outlier exclusion method developed around a Copula based joint probability model. This approach has the capability of capturing the complex multivariate nonlinear relation between parameters based on their univariate marginal distributions through the use of Copula; data points that deviate significantly from the consolidated power curve can then be removed depending on this derived joint probability distribution. After data filtering in this manner, it is shown how the resulting power curves are better defined and less subject to uncertainty, whilst broadly retaining the same dominant statistical characteristics. These improved power curves make subsequent condition monitoring more effective in the reliable detection of faults.

AB - Power curve measurements provide a conventional and effective means of assessing the performance of a wind turbine, both commercially and technically. Increasingly high wind penetration in power systems and offshore accessibility issues make it even more important to monitor the condition and performance of wind turbines based on timely and accurate wind speed and power measurements. Power curve data from Supervisory Control and Data Acquisition (SCADA) system records, however, often contains significant measurement deviations, which are commonly produced as a consequence of wind turbine operational transitions rather than stemming from physical degradation of the plant. To use this raw data for wind turbine condition monitoring purposes is thus likely to lead to high false alarm rates which would make the actual fault detection unreliable and potentially add unnecessarily to the costs of maintenance. To this end, this paper proposes a probabilistic outlier exclusion method developed around a Copula based joint probability model. This approach has the capability of capturing the complex multivariate nonlinear relation between parameters based on their univariate marginal distributions through the use of Copula; data points that deviate significantly from the consolidated power curve can then be removed depending on this derived joint probability distribution. After data filtering in this manner, it is shown how the resulting power curves are better defined and less subject to uncertainty, whilst broadly retaining the same dominant statistical characteristics. These improved power curves make subsequent condition monitoring more effective in the reliable detection of faults.

KW - wind energy

KW - wind turbine

KW - power systems

KW - power curve

KW - outlier rejection

KW - SCADA

KW - copula model

UR - http://onlinelibrary.wiley.com/journal/10.1002/%28ISSN%291099-1824

U2 - 10.1002/we.1661

DO - 10.1002/we.1661

M3 - Article

VL - 17

SP - 1677

EP - 1688

JO - Wind Energy

T2 - Wind Energy

JF - Wind Energy

SN - 1095-4244

IS - 11

ER -