In this section
@inproceedings{Giebel:2007,
author = {Giebel, G and Badger, J and Landberg, L and Nielsen, HA and Nielsen, TS and Madsen, H and Pinson, P and Sattler, K and Feddersen, H and Vedel, H},
pages = {838--849},
title = {Ensemble predictions: Understanding uncertainties},
year = {2007}
}
TY - CPAPER
AB - Short-term Prediction of wind power is used operationally by utilities with high wind power penetration for scheduling, trading or maintenance planning. The uncertainty of the forecast so far was produced from an analysis of historical errors as a situation-independent average value. Now, new meteorological tools are available to take the different predictability of the meteorological situation into account: Ensembles. The idea is that an ensemble of multiple forecasts, based on different initial conditions / numerical models / model runs, should give a measure of the forecast uncertainty. If the different members go on different tracks, they open up alternative scenarios for future weather. We have developed methods for (i) transforming the meteorological ensemble forecast into a wind power ensemble, and (ii) ensuring that the probabilistic properties of the derived forecasted wind power probability quantiles are correct. Also other ensemble types have been investigated. This paper summarises the findings of a Danish three year project, carried out by Ris0, DTU and DMI in conjunction with the major Danish utilities.
AU - Giebel,G
AU - Badger,J
AU - Landberg,L
AU - Nielsen,HA
AU - Nielsen,TS
AU - Madsen,H
AU - Pinson,P
AU - Sattler,K
AU - Feddersen,H
AU - Vedel,H
EP - 849
PY - 2007///
SP - 838
TI - Ensemble predictions: Understanding uncertainties
ER -
Dyson School of Design Engineering
Imperial College London
25 Exhibition Road
South Kensington
London
SW7 2DB
design.engineering@imperial.ac.uk
Tel: +44 (0) 20 7594 8888