| Autor: |
Kratzenberg, Manfred Georg, Zürn, Hans Helmut, Revheim, Pål Preede, Beyer, Hans Georg |
| Předmět: |
|
| Zdroj: |
Proceedings of the 13th International Workshop on Large - Scale Integration of Wind Power into Power Systems as well as on Transmission Networks for Offshore Wind Plants; 2014, p825-829, 5p |
| Abstrakt: |
This paper discusses options for improvement of short-term wind power forecast stemming from wind speed forecasts based on numerical weather prediction models. By use of different post-processing methods the improvement of the forecast performance is verified. These methods comprise a linear regression model and a wavelet neural network model. The respective schemes are applied to single site wind power forecasts for two different sites in Norway and are compared with regard to the model uncertainty. As for the sites, only information on measured wind speed and wind speed forecasts are available, wind power data are modeled by use of a generic wind turbine model and an extrapolation model which extrapolates the measured and predicted wind speed from ten meters above ground level to the turbines axis height. As auxiliary information, forecasts from a Numeric Weather Prediction model and past measurements from nearby sites are included as predictors or input variables into the post-processing models. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
