Comparative examinations of wind speed and energy extrapolation methods using remotely sensed data – A case study from Hungary

Exact knowledge of wind energy potential is a fundamental issue in wind energy utilization. The vertical changes in wind speeds, that is, the wind profile, have a predominant impact on the wind energy available at a location because the kinetic energy of moving air is proportional to the square of the wind speed. Roughness describes the resistance of a 3D surface to moving air. The exponent α of the power law of Hellmann and the roughness length (z0) are two parameters that describe the effects of the roughness of the surface on the wind profile. They can be used for the vertical extrapolation of wind speeds. The exponent α can be determined using multiple height level wind speed measurement data, whereas a reliable technique for the calculation of the roughness length requires detailed knowledge of the 3D geometry of the measurement site. In the present study, the exponent α was calculated based on SODAR wind speed measurements, while (z0) was determined using a combination of GIS and UAS-based aerial survey methods. Wind speeds measured at 50 m were extrapolated for height levels of 80, 90, 100, 110, and 120 m using dynamic power law exponent values. Wind power was determined using the power law (method V1), roughness length (method V2), frequency distribution (method W-RF), and gamma distribution (method W-G), and Windographer software was compared to the values calculated from the empirical (measured) wind speeds. A comparative statistical analysis of the datasets of the power law and roughness length methods on monthly/diurnal, annual/diurnal, and month/direction contexts showed no significant differences at all height levels. Differences can be detected in the distribution of the signs of the differences at heights of 80 and 120 m for the entire dataset. Underestimation was dominant with a significant frequency (over 70 %) in the case of both methods and heights. There were no significant differences between the wind power estimations provided by the different methods, and all the methods involved in the study underestimated the wind speeds and wind energy potential for each height level. Methods V1 and V2 can be used alternatively, depending on the input data available for analysis. The major advantage of method V2 is that it provides the same accuracy as V1, which requires a UAS-based aerial survey at the beginning, but continuous wind measurements must be performed at a lower height only. This means that there is no need for a high measurement tower, which makes the measurements simpler, more cost-effective, and causes much less disturbance to the environment. Another important advantage of the methods presented here is that they use a dynamic approach of power law exponent values that provide a more realistic estimation of wind speed and energy on a diurnal scale.