A boundary layer scaling (BLS) method for predicting long-term average near-surface wind speeds and power densities was developed in this work. The method was based on the scaling of reference climatological data either from long-term average wind maps or from hourly wind speeds obtained from high-resolution Numerical Weather Prediction (NWP) models, with case study applications from Great Britain. It incorporated a more detailed parameterisation of surface aerodynamics than previous studies and the predicted wind speeds and power densities were validated against observational wind speeds from 124 sites across Great Britain. The BLS model could offer long-term average wind speed predictions using wind map data derived from long-term observational data, with a mean percentage error of 1.5% which provided an improvement on the commonly used NOABL (Numerical Objective Analysis of Boundary Layer) wind map. The boundary layer scaling of NWP data was not, however, able to improve upon the use of raw NWP data for near surface wind speed predictions. However, the use of NWP data scaled by the BLS model could offer improved power density predictions compared to the use of the reference data sets. Using a vertical scaling of the shape factor of a Weibull distribution fitted to the BLS NWP data, power density predictions with a 1% mean percentage error were achieved. This provided a significant improvement on the use of a fixed shape factor which must be utilised when only long-term average wind speeds are available from reference wind maps. The work therefore highlights the advantages that use of a BLS model for wind speed and NWP data for power density predictions can offer for small to medium scale wind energy resource assessments, potentially facilitating more robust annual energy production and financial assessments of prospective small and medium scale wind turbine installations.

在这项工作中，开发了一种用于预测长期平均近地表风速和功率密度的边界层缩放（BLS）方法。该方法基于从长期平均风图或从高分辨率数值天气预报（NWP）模型获得的每小时风速参考气候数据的缩放比例，并具有英国的案例研究应用。与以前的研究相比，它结合了更详细的表面空气动力学参数设置，并且根据来自英国124个站点的观测风速对预测的风速和功率密度进行了验证。BLS模型可以使用源自长期观测数据的风图数据提供长期平均风速预测，平均百分比误差为1。5％，这对常用的NOABL（边界层数值目标分析）风图进行了改进。但是，将原始NWP数据用于近地表风速预测后，NWP数据的边界层缩放无法改善。但是，与参考数据集相比，使用由BLS模型缩放的NWP数据可以提供改进的功率密度预测。使用拟合到BLS NWP数据的Weibull分布的形状因子的垂直比例，可以实现平均误差为1％的功率密度预测。这极大地改善了固定形状因数的使用，当从参考风图仅获得长期平均风速时必须使用固定形状因数。