A sparsified vector autoregressive model for short-term wind farm power forecasting

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

3 Citations (Scopus)

Abstract

Short-term wind farm power forecasting is studied by exploiting the spatio-temporal correlation between individual turbine's power output. A multivariate time series model for wind farm power generation is developed by using vector autoregression (VAR). In order to avoid the possible over-fitting issues caused by a large number of autoregressive coefficients and the impact on the forecasting performance of VAR models, a sparsified autoregressive coefficient matrix is constructed by utilizing the information on wind direction, wind speed and wind farm's layout. Then, the VAR model parameters are obtained through maximum likelihood estimation of real-time measurement data, by taking into account the sparse structure of the autoregressive coefficient matrix. The proposed approach is compared with univariate autoregressive models through numerical experiments, resulting in significant improvement, which is attributed to the turbine-level correlation captured by the developed VAR model.

Original languageEnglish (US)
Title of host publicationIEEE Power and Energy Society General Meeting
PublisherIEEE Computer Society
Volume2015-September
ISBN (Print)9781467380409
DOIs
StatePublished - Sep 30 2015
EventIEEE Power and Energy Society General Meeting, PESGM 2015 - Denver, United States
Duration: Jul 26 2015Jul 30 2015

Other

OtherIEEE Power and Energy Society General Meeting, PESGM 2015
CountryUnited States
CityDenver
Period7/26/157/30/15

Fingerprint

Farms
Turbines
Maximum likelihood estimation
Time measurement
Power generation
Time series
Experiments

Keywords

  • Multivariate time series analysis
  • short-term wind power forecasting
  • vector autoregression
  • wind farm

ASJC Scopus subject areas

  • Energy Engineering and Power Technology
  • Nuclear Energy and Engineering
  • Renewable Energy, Sustainability and the Environment
  • Electrical and Electronic Engineering

Cite this

He, M., Vittal, V., & Zhang, J. (2015). A sparsified vector autoregressive model for short-term wind farm power forecasting. In IEEE Power and Energy Society General Meeting (Vol. 2015-September). [7285972] IEEE Computer Society. https://doi.org/10.1109/PESGM.2015.7285972

A sparsified vector autoregressive model for short-term wind farm power forecasting. / He, Miao; Vittal, Vijay; Zhang, Junshan.

IEEE Power and Energy Society General Meeting. Vol. 2015-September IEEE Computer Society, 2015. 7285972.

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

He, M, Vittal, V & Zhang, J 2015, A sparsified vector autoregressive model for short-term wind farm power forecasting. in IEEE Power and Energy Society General Meeting. vol. 2015-September, 7285972, IEEE Computer Society, IEEE Power and Energy Society General Meeting, PESGM 2015, Denver, United States, 7/26/15. https://doi.org/10.1109/PESGM.2015.7285972
He M, Vittal V, Zhang J. A sparsified vector autoregressive model for short-term wind farm power forecasting. In IEEE Power and Energy Society General Meeting. Vol. 2015-September. IEEE Computer Society. 2015. 7285972 https://doi.org/10.1109/PESGM.2015.7285972
He, Miao ; Vittal, Vijay ; Zhang, Junshan. / A sparsified vector autoregressive model for short-term wind farm power forecasting. IEEE Power and Energy Society General Meeting. Vol. 2015-September IEEE Computer Society, 2015.
@inproceedings{0e925a336d114ae192b4a6dd0fb67ba8,
title = "A sparsified vector autoregressive model for short-term wind farm power forecasting",
abstract = "Short-term wind farm power forecasting is studied by exploiting the spatio-temporal correlation between individual turbine's power output. A multivariate time series model for wind farm power generation is developed by using vector autoregression (VAR). In order to avoid the possible over-fitting issues caused by a large number of autoregressive coefficients and the impact on the forecasting performance of VAR models, a sparsified autoregressive coefficient matrix is constructed by utilizing the information on wind direction, wind speed and wind farm's layout. Then, the VAR model parameters are obtained through maximum likelihood estimation of real-time measurement data, by taking into account the sparse structure of the autoregressive coefficient matrix. The proposed approach is compared with univariate autoregressive models through numerical experiments, resulting in significant improvement, which is attributed to the turbine-level correlation captured by the developed VAR model.",
keywords = "Multivariate time series analysis, short-term wind power forecasting, vector autoregression, wind farm",
author = "Miao He and Vijay Vittal and Junshan Zhang",
year = "2015",
month = "9",
day = "30",
doi = "10.1109/PESGM.2015.7285972",
language = "English (US)",
isbn = "9781467380409",
volume = "2015-September",
booktitle = "IEEE Power and Energy Society General Meeting",
publisher = "IEEE Computer Society",

}

TY - GEN

T1 - A sparsified vector autoregressive model for short-term wind farm power forecasting

AU - He, Miao

AU - Vittal, Vijay

AU - Zhang, Junshan

PY - 2015/9/30

Y1 - 2015/9/30

N2 - Short-term wind farm power forecasting is studied by exploiting the spatio-temporal correlation between individual turbine's power output. A multivariate time series model for wind farm power generation is developed by using vector autoregression (VAR). In order to avoid the possible over-fitting issues caused by a large number of autoregressive coefficients and the impact on the forecasting performance of VAR models, a sparsified autoregressive coefficient matrix is constructed by utilizing the information on wind direction, wind speed and wind farm's layout. Then, the VAR model parameters are obtained through maximum likelihood estimation of real-time measurement data, by taking into account the sparse structure of the autoregressive coefficient matrix. The proposed approach is compared with univariate autoregressive models through numerical experiments, resulting in significant improvement, which is attributed to the turbine-level correlation captured by the developed VAR model.

AB - Short-term wind farm power forecasting is studied by exploiting the spatio-temporal correlation between individual turbine's power output. A multivariate time series model for wind farm power generation is developed by using vector autoregression (VAR). In order to avoid the possible over-fitting issues caused by a large number of autoregressive coefficients and the impact on the forecasting performance of VAR models, a sparsified autoregressive coefficient matrix is constructed by utilizing the information on wind direction, wind speed and wind farm's layout. Then, the VAR model parameters are obtained through maximum likelihood estimation of real-time measurement data, by taking into account the sparse structure of the autoregressive coefficient matrix. The proposed approach is compared with univariate autoregressive models through numerical experiments, resulting in significant improvement, which is attributed to the turbine-level correlation captured by the developed VAR model.

KW - Multivariate time series analysis

KW - short-term wind power forecasting

KW - vector autoregression

KW - wind farm

UR - http://www.scopus.com/inward/record.url?scp=84956855138&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84956855138&partnerID=8YFLogxK

U2 - 10.1109/PESGM.2015.7285972

DO - 10.1109/PESGM.2015.7285972

M3 - Conference contribution

SN - 9781467380409

VL - 2015-September

BT - IEEE Power and Energy Society General Meeting

PB - IEEE Computer Society

ER -