Heuristic correction of wind speed mesoscale models simulations for wind farms prospecting and micrositing

Abstract

The distribution of surface-level wind speeds over a given area is important information that is related to several processes in wind farm prospecting, design and micrositing. This information is often obtained from simulations using mesoscale models that take variables from global models as starting points. Improved outputs from mesoscale models can lead to reduced error compared to real wind speeds in the study area if in situ wind speed measurements are available. In this paper, we present several techniques to correct surface wind speed simulations from mesoscale models using data from measuring stations in wind farms. Specifically, we propose different heuristic corrections of the outputs from mesoscale models by means of surface fitting between the Weibull parameters of the wind speed series (from the mesoscale model) and those from the measuring stations (real wind speed) in the wind farm. The proposed methodology has direct applications in wind farm design, site prospection and micrositing. The good performance of our method is evident in the more accurate surface wind speeds obtained from mesoscale models in two wind farm prospection sites in Spain, where several measuring towers are installed.