Centennial Simulation Of Wind Power Through Soft Computing Algorithms

While there exists a wide knowledge on either the short-term variability or the climatological conditions (atlas) of wind and its corresponding wind power output, the study of the mid to long-term variations at a given location is still in its early stages. However, significant variations in the multidecadal variability timescales can substantially modify the performance of a wind power plant between the first layout and the end of its lifetime or eventual repowering. In this context, this manual presents a novel methodology to obtain wind and wind power centennial series with daily resolution for a specific site, contributing to enhance the knowledge on its long-term wind variability and its wind power output performance. To achieve this commitment, different tools for the statistical downscaling of wind are designed, by employing several Soft-Computing techniques as Greedy and Evolutionary Algorithms. Additionally, this allows local wind regimes to be objectively defined, a keytool to better understand the circulation mechanisms governing surface wind variability. In this manual specific results are presented for a collection of different wind farms in Spain.