Presentations

Abstract

Characterization of the wind resource is essential for the design and operation of wind farms. However, deploying measurement equipment can be difficult, expensive, and time-consuming. Instead, the wind energy industry relies on numerically modelled historical datasets for i) initial prospecting of new wind farm locations, ii) correction of short-term measurement data to account for the long term interannual variability and iii) production planning of future power generation. There are many commercial and public datasets available, and it is essential that the optimum dataset is selected for each use case. However, the wind energy industry has a poor understanding of how these different datasets compare, and their respective strengths and weaknesses. RWE address all of the above limitations by conducting the most extensive benchmarking of model data to wind measurements to date. We leverage the benefits of a multi-year big-data project to digitalise, automatically clean and harmonise every wind measurement deployed by RWE globally. With this database, we have access to 150 sites with over 350 independent wind measurement. Each site contains at least one wind measurement with over 1 year of recorded data. We first demonstrate the benefits of utilizing models with higher horizontal resolution for onshore sites. Then we show the impact of different boundary data on a commercial WRF based product. We consider the strengths and weaknesses of the different models across different geographic regions, focusing on key wind energy markets. Finally, we discuss the strengths and weaknesses global reanalysis versus regional (mesoscale) versus LES models.