You need JavaScript to view this

Relative performance of different numerical weather prediction models for short term predition of wind wnergy

Abstract

In several approaches presented in other papers in this conference, short term forecasting of wind power for a time horizon covering the next two days is done on the basis of Numerical Weather Prediction (NWP) models. This paper explores the relative merits of HIRLAM, which is the model used by the Danish Meteorological Institute, the Deutschlandmodell from the German Weather Service and the Nested Grid Model used in the US. The performance comparison will be mainly done for a site in Germany which is in the forecasting area of both the Deutschlandmodell and HIRLAM. In addition, a comparison of measured data with the forecasts made for one site in Iowa will be included, which allows conclusions on the merits of all three models. Differences in the relative performances could be due to a better tailoring of one model to its country, or to a tighter grid, or could be a function of the distance between the grid points and the measuring site. Also the amount, in which the performance can be enhanced by the use of model output statistics (topic of other papers in this conference) could give insights into the performance of the models. (au)
Authors:
Giebel, G; Landberg, L; [1]  Moennich, K; Waldl, H P [2] 
  1. Risoe National Lab., Wind Energy and Atmospheric Physics Dept., Roskilde (Denmark)
  2. Carl con Ossietzky Univ., Faculty of Physics, Dept. of Energy and Semiconductor, Oldenburg (Germany)
Publication Date:
Mar 01, 1999
Product Type:
Conference
Report Number:
RISO-R-1114(EN); CONF-990314-
Reference Number:
SCA: 170100; PA: DK-99:001413; EDB-99:087286; SN: 99002098319
Resource Relation:
Conference: EWEC`99. European wind energy conference, Nice (France), 1-5 Mar 1999; Other Information: PBD: Mar 1999; Related Information: Is Part Of Contributions from the Department of Wind Energy and Atmospheric Physics to EWEC `99 in Nice, France; Larsen, Gunnar C.; Westermann, Kirsten; Noergaard, Per [eds.]; PB: 256 p.
Subject:
17 WIND ENERGY; CLIMATE MODELS; WIND POWER; FORECASTING; PERFORMANCE
OSTI ID:
679631
Research Organizations:
Risoe National Lab., Roskilde (Denmark). Wind Energy and Atmospheric Physics Dept.
Country of Origin:
Denmark
Language:
English
Other Identifying Numbers:
Other: ON: DE99747779; CNN: Contract JOR3-CT95-0008; JOR3-CT98-0272; ISBN 87-550-2542-0; TRN: DK9901413
Availability:
OSTI as DE99747779
Submitting Site:
DK
Size:
pp. 157-160
Announcement Date:

Citation Formats

Giebel, G, Landberg, L, Moennich, K, and Waldl, H P. Relative performance of different numerical weather prediction models for short term predition of wind wnergy. Denmark: N. p., 1999. Web.
Giebel, G, Landberg, L, Moennich, K, & Waldl, H P. Relative performance of different numerical weather prediction models for short term predition of wind wnergy. Denmark.
Giebel, G, Landberg, L, Moennich, K, and Waldl, H P. 1999. "Relative performance of different numerical weather prediction models for short term predition of wind wnergy." Denmark.
@misc{etde_679631,
title = {Relative performance of different numerical weather prediction models for short term predition of wind wnergy}
author = {Giebel, G, Landberg, L, Moennich, K, and Waldl, H P}
abstractNote = {In several approaches presented in other papers in this conference, short term forecasting of wind power for a time horizon covering the next two days is done on the basis of Numerical Weather Prediction (NWP) models. This paper explores the relative merits of HIRLAM, which is the model used by the Danish Meteorological Institute, the Deutschlandmodell from the German Weather Service and the Nested Grid Model used in the US. The performance comparison will be mainly done for a site in Germany which is in the forecasting area of both the Deutschlandmodell and HIRLAM. In addition, a comparison of measured data with the forecasts made for one site in Iowa will be included, which allows conclusions on the merits of all three models. Differences in the relative performances could be due to a better tailoring of one model to its country, or to a tighter grid, or could be a function of the distance between the grid points and the measuring site. Also the amount, in which the performance can be enhanced by the use of model output statistics (topic of other papers in this conference) could give insights into the performance of the models. (au)}
place = {Denmark}
year = {1999}
month = {Mar}
}