Due to the great increase of wind power production during the last decade, an issue with top priority is the short-term prediction of wind power. Our efforts in this paper are directed to develop a method which combines a downscaling model, that provide us with a relation between local observations of wind speed and wind power with good resolution outputs from a mesoscale atmospheric model, MM5, (better than those that we obtain with a large scale GCM) and a hybrid autoregressive-neural network model, which let us understand the dynamics immersed into the time series recorded in a wind farm in Navarra, Spain.
Tweets by SantanderMeteo