Wind Profile-Based Vertical Extrapolation and Residual Learning Correction for Winter Wind Speed in Southern Henan
Wu Jingyan
Wang Li
Wang Xiao
Wei Lu
Xie Boyu
Huang Jie
Abstract：This study utilizes near-surface observational data from meteorological stations around wind farms in Henan Province from 2020 to 2024.Based on an assessment of wind field characteristics across the province,wind profile models are applied to vertically extrapolate winter wind speeds near two wind farms in southern Henan.The extrapolation results are then corrected using a residual learning method.The results are as follows.(1)The wind fields near wind farms exhibit distinct seasonal fea-tures,closely related to station elevation.Wind speeds are weakest in summer and strongest in winter.In high-altitude regions,wind speed distribution shows a wider range.Regarding wind direction,low-alti-tude areas are dominated by southerly winds in spring and summer and by northeasterly winds in autumn and winter.At altitudes above 600 m,the seasonal prevailing wind directions become more complex,typically characterized by southeasterly and northwesterly winds.(2)Wind speeds at heights of around 100 meters were extrapolated using both the power law and the logarithmic law wind profile models.The results from the two models are similar,and both outperform the 100-meter wind speeds derived from nearby ERA5 grid data.(3)The residual learning method was employed to correct the vertically extrapo-lated wind speeds,leading to a significant improvement in model accuracy.This technical approach con-tributes to addressing the lack of vertical observational data for low-altitude wind fields in the vicinity of wind farms.
Keywords：vertical wind speed extrapolationlow-level wind fieldresidual learning correctionwind farm
Publication Date：2026-05-30
Online Publishing Date：2026-09-12(First online date of this platform, not the publication date of the document)
Pages：9( 103-111 )