THE CASANDRA PROJECT: FIRST RESULTS OF A NEW WIND POWER OPERATIONAL FORECASTING SYSTEM IN SPAIN

摘要

A new wind power prediction system named Casandra is described. An important characteristic of this system is that it includes within the operational chain a numerical mesoscale meteorological model, instead of retrieving the mesoscale forecast data from an external source. After applying a statistical downscaling method and a wind farm model, the system provides a probabilistic forecast of the wind power production. Results are shown for two wind farms over moderately to highly complex terrain in Spain, where the system is being run operationally. An analysis of errors related to the uncertainty of meteorological global predictions is performed, as a way to provide an estimate of the maximum improvement that could be achieved by ensemble prediction techniques.