A comparison of four microscale wind flow models in predicting the real-world performance of a large-scale peri-urban wind turbine, using onsite LiDAR wind measurements

Abstract As wind energy continues to expand, new markets for distributed wind applications are expected, including behind-the-meter deployment at industrial consumer sites in peri-urban areas. The presence of buildings in these areas can give rise to complex wind regimes. This study compares the measured annual energy output of an operating Vestas V52 wind turbine, in a peri-urban environment, with predictions from four microscale wind flow models. The models upscale and downscale onsite LiDAR wind measurements at six heights from 10 m to 200 m. The models consist of a linear shelter model and three RANS CFD approaches, implemented in two widely used micro-siting tools, namely WAsP and WindSim. Variations from 1.4% to 64% between model predicted and measured electrical energy output are observed. For low-rise buildings with heights up to 20% of the turbine hub height, the models are most accurate using wind measurements at ~ 3 times the height of the buildings. In the case of a tall narrow building, ~ 80% of the hub height, best model predictions are from twice the height of the building. Wind shear measurements suggest that obstacles down to 20% of the hub height, up to distances of 100 obstacles heights away, are significant.