Leveraging stochastic differential equations for probabilistic forecasting of wind power using a dynamic power curve

Jan Emil Banning Iversen, Juan Miguel Morales González, Jan Kloppenborg Møller, Pierre-Julien Trombe, Henrik Madsen

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

Short-term (hours to days) probabilistic forecasts of wind power generation provide useful information about the associated uncertainty of these forecasts. Standard probabilistic forecasts are usually issued on a per-horizon-basis, meaning that they lack information about the development of the uncertainty over time or the inter-temporal correlation of forecast errors for different horizons. This information is very important for forecast end-users optimizing time-dependent variables or dealing with multi-period decision-making problems, such as the management and operation of power systems with a high penetration of renewable generation. This paper provides input to these problems by proposing a model based on stochastic differential equations that allows generating predictive densities as well as scenarios for wind power. We build upon a probabilistic model for wind speed and introduce a dynamic power curve. The model thus decomposes the dynamics of wind power prediction errors into wind speed forecast errors and errors related to the conversion from wind speed to wind power. We test the proposed model on an out-of-sample period of 1 year for a wind farm with a rated capacity of 21 MW. The model outperforms simple as well as advanced benchmarks on horizons ranging from 1 to 24 h.
Original languageEnglish
JournalWind Energy
Volume20
Issue number1
Pages (from-to)33–44
ISSN1095-4244
DOIs
Publication statusPublished - 2017

Cite this

Iversen, Jan Emil Banning ; Morales González, Juan Miguel ; Møller, Jan Kloppenborg ; Trombe, Pierre-Julien ; Madsen, Henrik. / Leveraging stochastic differential equations for probabilistic forecasting of wind power using a dynamic power curve. In: Wind Energy. 2017 ; Vol. 20, No. 1. pp. 33–44.
@article{b457f45707484b9fb771fe3365b23933,
title = "Leveraging stochastic differential equations for probabilistic forecasting of wind power using a dynamic power curve",
abstract = "Short-term (hours to days) probabilistic forecasts of wind power generation provide useful information about the associated uncertainty of these forecasts. Standard probabilistic forecasts are usually issued on a per-horizon-basis, meaning that they lack information about the development of the uncertainty over time or the inter-temporal correlation of forecast errors for different horizons. This information is very important for forecast end-users optimizing time-dependent variables or dealing with multi-period decision-making problems, such as the management and operation of power systems with a high penetration of renewable generation. This paper provides input to these problems by proposing a model based on stochastic differential equations that allows generating predictive densities as well as scenarios for wind power. We build upon a probabilistic model for wind speed and introduce a dynamic power curve. The model thus decomposes the dynamics of wind power prediction errors into wind speed forecast errors and errors related to the conversion from wind speed to wind power. We test the proposed model on an out-of-sample period of 1 year for a wind farm with a rated capacity of 21 MW. The model outperforms simple as well as advanced benchmarks on horizons ranging from 1 to 24 h.",
author = "Iversen, {Jan Emil Banning} and {Morales Gonz{\'a}lez}, {Juan Miguel} and M{\o}ller, {Jan Kloppenborg} and Pierre-Julien Trombe and Henrik Madsen",
year = "2017",
doi = "10.1002/we.1988",
language = "English",
volume = "20",
pages = "33–44",
journal = "Wind Energy",
issn = "1095-4244",
publisher = "JohnWiley & Sons Ltd.",
number = "1",

}

Leveraging stochastic differential equations for probabilistic forecasting of wind power using a dynamic power curve. / Iversen, Jan Emil Banning; Morales González, Juan Miguel; Møller, Jan Kloppenborg; Trombe, Pierre-Julien ; Madsen, Henrik.

In: Wind Energy, Vol. 20, No. 1, 2017, p. 33–44.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Leveraging stochastic differential equations for probabilistic forecasting of wind power using a dynamic power curve

AU - Iversen, Jan Emil Banning

AU - Morales González, Juan Miguel

AU - Møller, Jan Kloppenborg

AU - Trombe, Pierre-Julien

AU - Madsen, Henrik

PY - 2017

Y1 - 2017

N2 - Short-term (hours to days) probabilistic forecasts of wind power generation provide useful information about the associated uncertainty of these forecasts. Standard probabilistic forecasts are usually issued on a per-horizon-basis, meaning that they lack information about the development of the uncertainty over time or the inter-temporal correlation of forecast errors for different horizons. This information is very important for forecast end-users optimizing time-dependent variables or dealing with multi-period decision-making problems, such as the management and operation of power systems with a high penetration of renewable generation. This paper provides input to these problems by proposing a model based on stochastic differential equations that allows generating predictive densities as well as scenarios for wind power. We build upon a probabilistic model for wind speed and introduce a dynamic power curve. The model thus decomposes the dynamics of wind power prediction errors into wind speed forecast errors and errors related to the conversion from wind speed to wind power. We test the proposed model on an out-of-sample period of 1 year for a wind farm with a rated capacity of 21 MW. The model outperforms simple as well as advanced benchmarks on horizons ranging from 1 to 24 h.

AB - Short-term (hours to days) probabilistic forecasts of wind power generation provide useful information about the associated uncertainty of these forecasts. Standard probabilistic forecasts are usually issued on a per-horizon-basis, meaning that they lack information about the development of the uncertainty over time or the inter-temporal correlation of forecast errors for different horizons. This information is very important for forecast end-users optimizing time-dependent variables or dealing with multi-period decision-making problems, such as the management and operation of power systems with a high penetration of renewable generation. This paper provides input to these problems by proposing a model based on stochastic differential equations that allows generating predictive densities as well as scenarios for wind power. We build upon a probabilistic model for wind speed and introduce a dynamic power curve. The model thus decomposes the dynamics of wind power prediction errors into wind speed forecast errors and errors related to the conversion from wind speed to wind power. We test the proposed model on an out-of-sample period of 1 year for a wind farm with a rated capacity of 21 MW. The model outperforms simple as well as advanced benchmarks on horizons ranging from 1 to 24 h.

U2 - 10.1002/we.1988

DO - 10.1002/we.1988

M3 - Journal article

VL - 20

SP - 33

EP - 44

JO - Wind Energy

JF - Wind Energy

SN - 1095-4244

IS - 1

ER -