Using conditional kernel density estimation for wind power density forecasting

Jooyong Jeon, James W. Taylor

Research output: Contribution to journalArticle

  • 81 Citations

Abstract

Of the various renewable energy resources, wind power is widely recognized as one of the most promising. The management of wind farms and electricity systems can benefit greatly from the availability of estimates of the probability distribution of wind power generation. However, most research has focused on point forecasting of wind power. In this article, we develop an approach to producing density forecasts for the wind power generated at individual wind farms. Our interest is in intraday data and prediction from 1 to 72 hours ahead. We model wind power in terms of wind speed and wind direction. In this framework, there are two key uncertainties. First, there is the inherent uncertainty in wind speed and direction, and we model this using a bivariate vector autoregressive moving average-generalized autoregressive conditional heteroscedastic (VARMA-GARCH) model, with a Student t error distribution, in the Cartesian space of wind speed and direction. Second, there is the stochastic nature of the relationship of wind power to wind speed (described by the power curve), and to wind direction. We model this using conditional kernel density (CKD) estimation, which enables a nonparametric modeling of the conditional density of wind power. Using Monte Carlo simulation of the VARMA-GARCH model and CKD estimation, density forecasts of wind speed and direction are converted to wind power density forecasts. Our work is novel in several respects: previous wind power studies have not modeled a stochastic power curve; to accommodate time evolution in the power curve, we incorporate a time decay factor within the CKD method; and the CKD method is conditional on a density, rather than a single value. The new approach is evaluated using datasets from four Greek wind farms.
LanguageEnglish
Pages66-79
Number of pages14
JournalJournal of the American Statistical Association
Volume107
Issue number497
Early online date11 Jun 2012
DOIs
StatusPublished - 2012

Fingerprint

Wind Power
Kernel Density Estimation
Conditional Density
Forecasting
Wind Speed
Kernel Density
Heteroscedastic Model
Forecast
Conditional Model
Autoregressive Moving Average
Curve
Kernel density estimation
Wind power
Density forecasting
Uncertainty
Renewable Energy
Electricity
Cartesian
Probability Distribution
Availability

Keywords

  • Sustainability
  • Bivariate GARCH
  • Kernel estimation
  • Predictive distribution
  • Wind direction
  • Wind energy
  • Wind speed

Cite this

Using conditional kernel density estimation for wind power density forecasting. / Jeon, Jooyong; Taylor, James W.

In: Journal of the American Statistical Association, Vol. 107, No. 497, 2012, p. 66-79.

Research output: Contribution to journalArticle

@article{559b08af8b5847eda536d9871b7e2c26,
title = "Using conditional kernel density estimation for wind power density forecasting",
abstract = "Of the various renewable energy resources, wind power is widely recognized as one of the most promising. The management of wind farms and electricity systems can benefit greatly from the availability of estimates of the probability distribution of wind power generation. However, most research has focused on point forecasting of wind power. In this article, we develop an approach to producing density forecasts for the wind power generated at individual wind farms. Our interest is in intraday data and prediction from 1 to 72 hours ahead. We model wind power in terms of wind speed and wind direction. In this framework, there are two key uncertainties. First, there is the inherent uncertainty in wind speed and direction, and we model this using a bivariate vector autoregressive moving average-generalized autoregressive conditional heteroscedastic (VARMA-GARCH) model, with a Student t error distribution, in the Cartesian space of wind speed and direction. Second, there is the stochastic nature of the relationship of wind power to wind speed (described by the power curve), and to wind direction. We model this using conditional kernel density (CKD) estimation, which enables a nonparametric modeling of the conditional density of wind power. Using Monte Carlo simulation of the VARMA-GARCH model and CKD estimation, density forecasts of wind speed and direction are converted to wind power density forecasts. Our work is novel in several respects: previous wind power studies have not modeled a stochastic power curve; to accommodate time evolution in the power curve, we incorporate a time decay factor within the CKD method; and the CKD method is conditional on a density, rather than a single value. The new approach is evaluated using datasets from four Greek wind farms.",
keywords = "Sustainability, Bivariate GARCH, Kernel estimation, Predictive distribution, Wind direction, Wind energy, Wind speed",
author = "Jooyong Jeon and Taylor, {James W.}",
year = "2012",
doi = "10.1080/01621459.2011.643745",
language = "English",
volume = "107",
pages = "66--79",
journal = "Journal of the American Statistical Association",
issn = "0162-1459",
publisher = "Taylor & Francis",
number = "497",

}

TY - JOUR

T1 - Using conditional kernel density estimation for wind power density forecasting

AU - Jeon, Jooyong

AU - Taylor, James W.

PY - 2012

Y1 - 2012

N2 - Of the various renewable energy resources, wind power is widely recognized as one of the most promising. The management of wind farms and electricity systems can benefit greatly from the availability of estimates of the probability distribution of wind power generation. However, most research has focused on point forecasting of wind power. In this article, we develop an approach to producing density forecasts for the wind power generated at individual wind farms. Our interest is in intraday data and prediction from 1 to 72 hours ahead. We model wind power in terms of wind speed and wind direction. In this framework, there are two key uncertainties. First, there is the inherent uncertainty in wind speed and direction, and we model this using a bivariate vector autoregressive moving average-generalized autoregressive conditional heteroscedastic (VARMA-GARCH) model, with a Student t error distribution, in the Cartesian space of wind speed and direction. Second, there is the stochastic nature of the relationship of wind power to wind speed (described by the power curve), and to wind direction. We model this using conditional kernel density (CKD) estimation, which enables a nonparametric modeling of the conditional density of wind power. Using Monte Carlo simulation of the VARMA-GARCH model and CKD estimation, density forecasts of wind speed and direction are converted to wind power density forecasts. Our work is novel in several respects: previous wind power studies have not modeled a stochastic power curve; to accommodate time evolution in the power curve, we incorporate a time decay factor within the CKD method; and the CKD method is conditional on a density, rather than a single value. The new approach is evaluated using datasets from four Greek wind farms.

AB - Of the various renewable energy resources, wind power is widely recognized as one of the most promising. The management of wind farms and electricity systems can benefit greatly from the availability of estimates of the probability distribution of wind power generation. However, most research has focused on point forecasting of wind power. In this article, we develop an approach to producing density forecasts for the wind power generated at individual wind farms. Our interest is in intraday data and prediction from 1 to 72 hours ahead. We model wind power in terms of wind speed and wind direction. In this framework, there are two key uncertainties. First, there is the inherent uncertainty in wind speed and direction, and we model this using a bivariate vector autoregressive moving average-generalized autoregressive conditional heteroscedastic (VARMA-GARCH) model, with a Student t error distribution, in the Cartesian space of wind speed and direction. Second, there is the stochastic nature of the relationship of wind power to wind speed (described by the power curve), and to wind direction. We model this using conditional kernel density (CKD) estimation, which enables a nonparametric modeling of the conditional density of wind power. Using Monte Carlo simulation of the VARMA-GARCH model and CKD estimation, density forecasts of wind speed and direction are converted to wind power density forecasts. Our work is novel in several respects: previous wind power studies have not modeled a stochastic power curve; to accommodate time evolution in the power curve, we incorporate a time decay factor within the CKD method; and the CKD method is conditional on a density, rather than a single value. The new approach is evaluated using datasets from four Greek wind farms.

KW - Sustainability

KW - Bivariate GARCH

KW - Kernel estimation

KW - Predictive distribution

KW - Wind direction

KW - Wind energy

KW - Wind speed

UR - http://www.scopus.com/inward/record.url?scp=84859596365&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1080/01621459.2011.643745

U2 - 10.1080/01621459.2011.643745

DO - 10.1080/01621459.2011.643745

M3 - Article

VL - 107

SP - 66

EP - 79

JO - Journal of the American Statistical Association

T2 - Journal of the American Statistical Association

JF - Journal of the American Statistical Association

SN - 0162-1459

IS - 497

ER -