Wind Farm Power Forecasting for Less Than an Hour Using Multi Dimensional Models

Torben Knudsen, Thomas Bak, Tom Nørgaard Jensen

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

Resumé

The paper focus on prediction of wind farm power for horizons of 0-10 minutes and not more than one hour using statistical methods. These short term predictions
are relevant for both transmission system operators, wind farm operators and traders. Previous research indicates that for short time horizons the persistence method performs as well as more complex methods. However, these results are based on
accumulated power for an entire wind farm. The contribution in this paper is to develop multi-dimensional linear methods based on measurements of power or wind speed from individual wind turbine in a wind farm. These multi-dimensional methods are compared with the persistence method using real 1 minute average data from the Sheringham Shoal wind farm with 88 turbines. The results show that the use of measurements from individual turbines reduce the prediction errors 5-10% and also improves the prediction error variance estimate compared to the persistence method. We also present convincing examples showing that the predictions follow the wind farm power over a window of an hour.
OriginalsprogEngelsk
TitelProceedings, European Control Conference 2018
Antal sider8
ForlagIEEE
Publikationsdato29 nov. 2018
Sider3057-3064
ISBN (Trykt)978-1-5386-5303-6
ISBN (Elektronisk)978-3-9524-2698-2
DOI
StatusUdgivet - 29 nov. 2018
BegivenhedEuropean Control Conference 2018 - , Cypern
Varighed: 12 jun. 201815 jun. 2018

Konference

KonferenceEuropean Control Conference 2018
LandCypern
Periode12/06/201815/06/2018

Fingerprint

Farms
Turbines
Wind turbines
Statistical methods

Citer dette

Knudsen, T., Bak, T., & Jensen, T. N. (2018). Wind Farm Power Forecasting for Less Than an Hour Using Multi Dimensional Models. I Proceedings, European Control Conference 2018 (s. 3057-3064). IEEE. https://doi.org/10.23919/ECC.2018.8550286
Knudsen, Torben ; Bak, Thomas ; Jensen, Tom Nørgaard. / Wind Farm Power Forecasting for Less Than an Hour Using Multi Dimensional Models. Proceedings, European Control Conference 2018. IEEE, 2018. s. 3057-3064
@inproceedings{166a77b618764936ba3505894e43b194,
title = "Wind Farm Power Forecasting for Less Than an Hour Using Multi Dimensional Models",
abstract = "The paper focus on prediction of wind farm power for horizons of 0-10 minutes and not more than one hour using statistical methods. These short term predictionsare relevant for both transmission system operators, wind farm operators and traders. Previous research indicates that for short time horizons the persistence method performs as well as more complex methods. However, these results are based onaccumulated power for an entire wind farm. The contribution in this paper is to develop multi-dimensional linear methods based on measurements of power or wind speed from individual wind turbine in a wind farm. These multi-dimensional methods are compared with the persistence method using real 1 minute average data from the Sheringham Shoal wind farm with 88 turbines. The results show that the use of measurements from individual turbines reduce the prediction errors 5-10{\%} and also improves the prediction error variance estimate compared to the persistence method. We also present convincing examples showing that the predictions follow the wind farm power over a window of an hour.",
author = "Torben Knudsen and Thomas Bak and Jensen, {Tom N{\o}rgaard}",
year = "2018",
month = "11",
day = "29",
doi = "10.23919/ECC.2018.8550286",
language = "English",
isbn = "978-1-5386-5303-6",
pages = "3057--3064",
booktitle = "Proceedings, European Control Conference 2018",
publisher = "IEEE",
address = "United States",

}

Knudsen, T, Bak, T & Jensen, TN 2018, Wind Farm Power Forecasting for Less Than an Hour Using Multi Dimensional Models. i Proceedings, European Control Conference 2018. IEEE, s. 3057-3064, Cypern, 12/06/2018. https://doi.org/10.23919/ECC.2018.8550286

Wind Farm Power Forecasting for Less Than an Hour Using Multi Dimensional Models. / Knudsen, Torben; Bak, Thomas; Jensen, Tom Nørgaard.

Proceedings, European Control Conference 2018. IEEE, 2018. s. 3057-3064.

Publikation: Bidrag til bog/antologi/rapport/konference proceedingKonferenceartikel i proceedingForskningpeer review

TY - GEN

T1 - Wind Farm Power Forecasting for Less Than an Hour Using Multi Dimensional Models

AU - Knudsen, Torben

AU - Bak, Thomas

AU - Jensen, Tom Nørgaard

PY - 2018/11/29

Y1 - 2018/11/29

N2 - The paper focus on prediction of wind farm power for horizons of 0-10 minutes and not more than one hour using statistical methods. These short term predictionsare relevant for both transmission system operators, wind farm operators and traders. Previous research indicates that for short time horizons the persistence method performs as well as more complex methods. However, these results are based onaccumulated power for an entire wind farm. The contribution in this paper is to develop multi-dimensional linear methods based on measurements of power or wind speed from individual wind turbine in a wind farm. These multi-dimensional methods are compared with the persistence method using real 1 minute average data from the Sheringham Shoal wind farm with 88 turbines. The results show that the use of measurements from individual turbines reduce the prediction errors 5-10% and also improves the prediction error variance estimate compared to the persistence method. We also present convincing examples showing that the predictions follow the wind farm power over a window of an hour.

AB - The paper focus on prediction of wind farm power for horizons of 0-10 minutes and not more than one hour using statistical methods. These short term predictionsare relevant for both transmission system operators, wind farm operators and traders. Previous research indicates that for short time horizons the persistence method performs as well as more complex methods. However, these results are based onaccumulated power for an entire wind farm. The contribution in this paper is to develop multi-dimensional linear methods based on measurements of power or wind speed from individual wind turbine in a wind farm. These multi-dimensional methods are compared with the persistence method using real 1 minute average data from the Sheringham Shoal wind farm with 88 turbines. The results show that the use of measurements from individual turbines reduce the prediction errors 5-10% and also improves the prediction error variance estimate compared to the persistence method. We also present convincing examples showing that the predictions follow the wind farm power over a window of an hour.

U2 - 10.23919/ECC.2018.8550286

DO - 10.23919/ECC.2018.8550286

M3 - Article in proceeding

SN - 978-1-5386-5303-6

SP - 3057

EP - 3064

BT - Proceedings, European Control Conference 2018

PB - IEEE

ER -