Multiple linear regression MOS for short-term wind power forecast
Versandkostenfrei!
Versandfertig in 6-10 Tagen
24,99 €
inkl. MwSt.
PAYBACK Punkte
12 °P sammeln!
This study was carried out at Risø DTU (Denmark) during summer 2010 and was presented at the University of Barcelona (Spain) in July 2011 as the thesis of the Master in Meteorology. The main objective of this research work was to develop a statistical treatment to post-process Numerical Weather Predictions (NWP) outputs in order to improve short-term wind power forecasts. The proposed method is a Model Output Statistics (MOS) based on a multiple linear regression that screens the most relevant NWP forecast variables and selects best predictors, utilizing wind farm power output measurements as...
This study was carried out at Risø DTU (Denmark) during summer 2010 and was presented at the University of Barcelona (Spain) in July 2011 as the thesis of the Master in Meteorology. The main objective of this research work was to develop a statistical treatment to post-process Numerical Weather Predictions (NWP) outputs in order to improve short-term wind power forecasts. The proposed method is a Model Output Statistics (MOS) based on a multiple linear regression that screens the most relevant NWP forecast variables and selects best predictors, utilizing wind farm power output measurements as input. The proposed MOS performed well in the 2 studied wind farms. Its forecasts compare positively with actual operative model in use at Risø DTU and other MOS types, showing minimum BIAS and improving NWP power forecast of around 15% in terms of root mean square error.
Andere Kunden interessierten sich für
