We present a method for generalizing wind climatologies generated from mesoscale model output (e.g. the Weather, Research and Forecasting (WRF) model.) The generalization procedure is based on Wind Atlas framework of WAsP and KAMM/WAsP, and been extensively in wind resources assessment in DTU Wind Energy for many years (Mortensen et al 2005, Badger et al 2010). The method is verified by two distinct sets of WRF simulations: A single WRF wind atlas over Denmark, and a series of WRF simulations with the Vortex MAPS system with 29 sites in many wind climate regimes. Comparison is made between generalized wind climatologies estimated by the microscale model WAsP and the methodology presented here. For the Danish wind measurements the mean absolute error in the ‘raw’ wind speeds is 9.2%, while the mean absolute error in the generalized wind speeds is 4.1%. The generalization procedure has been installed and automated in the Vortex downscaling chain for the MAPS calculation, which are computed at 3 km horizontal resolution grid. For the MAPS-derived wind speeds the mean absolute error in the ‘raw’ wind speeds is 17.3%, while the mean absolute error in the generalized winds is 10.7%.
Main Research Area:
European Wind Energy Conference & Exhibition 2013