Symbolic regression-based improved method for wind speed extrapolation from lower to higher altitudes for wind energy applications

Prolonged wind speed assessment at higher altitudes is essential for wind energy estimation and planning. However, the erection and maintenance of tall wind measuring masts for this purpose cause many practical inconveniences from the engineering perspective. A rather simplified method often used for this task is to measure the wind profile at relatively lower altitudes and extrapolate the same to the required higher heights by empirical equations framed using hypothetical and experiential research. Such models often show errors due to the uncertainties caused by the complex nature of turbulent flows and the terrain. In this paper, we propose a new method of applying symbolic regression to the wind speed data over a short duration measured at a reference location to obtain a symbolic function capable of estimating wind speeds at higher altitudes using wind speed data at lower altitudes at different locations. Compared to the traditional power law method, the new method performs more accurately in different seasons at the reference as well as far away locations, achieving a maximum of 61.04% reduction in daily RMSE when analyzed with wind speeds averaged over 10-min intervals in this study. The new method opens up the possibility of wind resource assessment at higher altitudes at different locations by employing engineering-friendly shorter wind measuring masts.