Application of statistical wind models for system impacts

D. Hill, D. Mcmillan, K. Bell, D. Infield, G. W. Ault

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

Abstract

The UK government has provided an incentive mechanism for renewable electricity that is delivering a significantly increasing penetration of wind power in the electricity supply mix, and this growth is likely to continue in the
near to medium term. There is a real and pressing need to assess the impacts of increasing amounts of wind power on the UK power system. Statistical models are presented that characterize the temporal and spatial nature of windspeeds across the UK in a more comprehensive way than hitherto expressed. AutoRegressive Moving Average models (ARMA), often used for predictive purposes on shorter time-scales, are developed to characterize the windspeed field. A detrending method to allow for non-stationarity of the data is presented, developed specifically to model annual trends and a seasonally dependent
diurnal effect, noted to be present across sites studied. Vector auto-regressive (VAR) models extend the work by incorporating spatiotemporal correlations between the different sites. Results are presented demonstrating the effectiveness of the proposed approach to wind modelling and synthesis. In future work, these wind synthesis procedures will be used as input to wind and
power system time domain modeling with a view to an improved understanding of how a substantial UK wind penetration will impact on grid operation, thus providing a powerful tool for operational and planning purposes.
LanguageEnglish
Title of host publicationUniversities Power Engineering Conference (UPEC), 2009 Proceedings of the 44th International
PublisherIEEE
Number of pages5
ISBN (Print)978-1-4244-6823-2
Publication statusPublished - 11 Mar 2010
EventThe 44th International Universities' Power Engineering Conference - Glasgow, United Kingdom
Duration: 1 Sep 20094 Sep 2009

Conference

ConferenceThe 44th International Universities' Power Engineering Conference
CountryUnited Kingdom
CityGlasgow
Period1/09/094/09/09

Fingerprint

Wind power
Electricity
Planning
Statistical Models

Keywords

  • vector auto-regression
  • wind power
  • ARMA modelling
  • windspeed prediction
  • atmospheric modeling
  • autocorrelation
  • autoregressive processes
  • government
  • large scale integration
  • power system modeling
  • wind forecasting
  • wind energy
  • predictive models
  • power system planning

Cite this

Hill, D., Mcmillan, D., Bell, K., Infield, D., & Ault, G. W. (2010). Application of statistical wind models for system impacts. In Universities Power Engineering Conference (UPEC), 2009 Proceedings of the 44th International IEEE.
Hill, D. ; Mcmillan, D. ; Bell, K. ; Infield, D. ; Ault, G. W. / Application of statistical wind models for system impacts. Universities Power Engineering Conference (UPEC), 2009 Proceedings of the 44th International . IEEE, 2010.
@inproceedings{1397e132033f4f8abb7be8b3807ae63e,
title = "Application of statistical wind models for system impacts",
abstract = "The UK government has provided an incentive mechanism for renewable electricity that is delivering a significantly increasing penetration of wind power in the electricity supply mix, and this growth is likely to continue in thenear to medium term. There is a real and pressing need to assess the impacts of increasing amounts of wind power on the UK power system. Statistical models are presented that characterize the temporal and spatial nature of windspeeds across the UK in a more comprehensive way than hitherto expressed. AutoRegressive Moving Average models (ARMA), often used for predictive purposes on shorter time-scales, are developed to characterize the windspeed field. A detrending method to allow for non-stationarity of the data is presented, developed specifically to model annual trends and a seasonally dependentdiurnal effect, noted to be present across sites studied. Vector auto-regressive (VAR) models extend the work by incorporating spatiotemporal correlations between the different sites. Results are presented demonstrating the effectiveness of the proposed approach to wind modelling and synthesis. In future work, these wind synthesis procedures will be used as input to wind andpower system time domain modeling with a view to an improved understanding of how a substantial UK wind penetration will impact on grid operation, thus providing a powerful tool for operational and planning purposes.",
keywords = "vector auto-regression, wind power, ARMA modelling, windspeed prediction, atmospheric modeling , autocorrelation , autoregressive processes, government , large scale integration , power system modeling , wind forecasting, wind energy, predictive models, power system planning",
author = "D. Hill and D. Mcmillan and K. Bell and D. Infield and Ault, {G. W.}",
year = "2010",
month = "3",
day = "11",
language = "English",
isbn = "978-1-4244-6823-2",
booktitle = "Universities Power Engineering Conference (UPEC), 2009 Proceedings of the 44th International",
publisher = "IEEE",

}

Hill, D, Mcmillan, D, Bell, K, Infield, D & Ault, GW 2010, Application of statistical wind models for system impacts. in Universities Power Engineering Conference (UPEC), 2009 Proceedings of the 44th International . IEEE, The 44th International Universities' Power Engineering Conference, Glasgow, United Kingdom, 1/09/09.

Application of statistical wind models for system impacts. / Hill, D.; Mcmillan, D.; Bell, K.; Infield, D.; Ault, G. W.

Universities Power Engineering Conference (UPEC), 2009 Proceedings of the 44th International . IEEE, 2010.

Research output: Chapter in Book/Report/Conference proceedingConference contribution book

TY - GEN

T1 - Application of statistical wind models for system impacts

AU - Hill, D.

AU - Mcmillan, D.

AU - Bell, K.

AU - Infield, D.

AU - Ault, G. W.

PY - 2010/3/11

Y1 - 2010/3/11

N2 - The UK government has provided an incentive mechanism for renewable electricity that is delivering a significantly increasing penetration of wind power in the electricity supply mix, and this growth is likely to continue in thenear to medium term. There is a real and pressing need to assess the impacts of increasing amounts of wind power on the UK power system. Statistical models are presented that characterize the temporal and spatial nature of windspeeds across the UK in a more comprehensive way than hitherto expressed. AutoRegressive Moving Average models (ARMA), often used for predictive purposes on shorter time-scales, are developed to characterize the windspeed field. A detrending method to allow for non-stationarity of the data is presented, developed specifically to model annual trends and a seasonally dependentdiurnal effect, noted to be present across sites studied. Vector auto-regressive (VAR) models extend the work by incorporating spatiotemporal correlations between the different sites. Results are presented demonstrating the effectiveness of the proposed approach to wind modelling and synthesis. In future work, these wind synthesis procedures will be used as input to wind andpower system time domain modeling with a view to an improved understanding of how a substantial UK wind penetration will impact on grid operation, thus providing a powerful tool for operational and planning purposes.

AB - The UK government has provided an incentive mechanism for renewable electricity that is delivering a significantly increasing penetration of wind power in the electricity supply mix, and this growth is likely to continue in thenear to medium term. There is a real and pressing need to assess the impacts of increasing amounts of wind power on the UK power system. Statistical models are presented that characterize the temporal and spatial nature of windspeeds across the UK in a more comprehensive way than hitherto expressed. AutoRegressive Moving Average models (ARMA), often used for predictive purposes on shorter time-scales, are developed to characterize the windspeed field. A detrending method to allow for non-stationarity of the data is presented, developed specifically to model annual trends and a seasonally dependentdiurnal effect, noted to be present across sites studied. Vector auto-regressive (VAR) models extend the work by incorporating spatiotemporal correlations between the different sites. Results are presented demonstrating the effectiveness of the proposed approach to wind modelling and synthesis. In future work, these wind synthesis procedures will be used as input to wind andpower system time domain modeling with a view to an improved understanding of how a substantial UK wind penetration will impact on grid operation, thus providing a powerful tool for operational and planning purposes.

KW - vector auto-regression

KW - wind power

KW - ARMA modelling

KW - windspeed prediction

KW - atmospheric modeling

KW - autocorrelation

KW - autoregressive processes

KW - government

KW - large scale integration

KW - power system modeling

KW - wind forecasting

KW - wind energy

KW - predictive models

KW - power system planning

UR - http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5429391

UR - http://www.upec2009.org/content/node/237

M3 - Conference contribution book

SN - 978-1-4244-6823-2

BT - Universities Power Engineering Conference (UPEC), 2009 Proceedings of the 44th International

PB - IEEE

ER -

Hill D, Mcmillan D, Bell K, Infield D, Ault GW. Application of statistical wind models for system impacts. In Universities Power Engineering Conference (UPEC), 2009 Proceedings of the 44th International . IEEE. 2010