Posters

Abstract

A representative wind shear coefficient is crucial for the accurate vertical extrapolation of wind speed measurements. Although the industry standard is to extend measurement campaigns to a full year to avoid seasonal biases, this comes at a time and cost expense. In this light, the goal of this work is to investigate the representativeness of very short-term campaigns and to propose a methodology to correct possible seasonal biases on the wind shear coefficient. Five correction methods for the estimation of an annual average shear are compared on a validation dataset of 50 measurement campaigns across Europe. The datasets are subdivided into 120 very short-term measurement datasets to simulate the effect of shortening the measurement period. The investigated correction approaches are variations of proposed methods in the literature, consisting in subdividing the very-short-term data into bins with increasing number of dimensions, including the shear coefficient, Monin-Obukhov length, atmospheric boundary layer thickness and wind direction. In each bin, ERA5 wind shear is compared to the long-term average to calculate a long-term correction factor. Shear correction factors were validated against the annual average measured shear, assumed representative of the long term. The directional method incorporating the shear distribution, the Monin-Obukhov length and the surface layer height yielded the most accurate results. The correction method reduces the absolute bias on the shear of a summer 3-months campaign from 11.6% to 0.1% while decreasing the related uncertainty from 7.2% to 5.4%. However, the error and uncertainty on the corrected shear value highly depend on the measurement period, the duration, and location of the campaign. Notably, applying the correction method on shoulder seasons tends to over-correct the shear. This emphasizes the need for wind farm developers both to consider the timing of their campaign and to make a deliberate choice on whether to apply the correction or not. While shear correction still significantly improved the accuracy, results on measurement campaigns outside of Europe were significantly worse. Including directionality did not significantly improve the correction accuracy. Additionally, a sensitivity analysis reveals that the binning strategy has a significant impact on the correction accuracy. Further work is required to better understand and optimise the choice of the binning strategy, and to introduce the consideration of different stratification scenarios depending on the hour of the day and the month. Finally, integrating the measured turbulence intensity could significantly improve the results as this parameter is proven to be strongly related to wind shear.