一种基于径向基函数插值的风资源分布估计方法  

A method for estimating wind resource distribution based on interpolation by radial basis functions

作　　者：刘尉 夏雪 李舒月 王彧 许子诺 LIU Wei;XIA Xue;LI Shuyue;WANG Yu;XU Zinuo(School of Mathematics,Hohai University,Nanjing 210098,China)

机构地区：[1]河海大学数学学院,江苏南京210098

出　　处：《甘肃科学学报》2025年第1期25-32,共8页Journal of Gansu Sciences

摘　　要：风资源分布信息在风力发电、可持续发展等领域具有重要价值。以黑龙江西北部23个气象站点在1989—2018年30年间的逐日平均风速数据为基础,依据相关系数与推土机距离(EMD)为聚类指标进行站点分类,并通过F检验确定最佳聚类。对待测点插值时选用适当的类别中的站点,与径向基函数(RBF)插值法相结合,构建基于风速相关性的风速插值模型,并使用均方误差(MSE)和决定系数(R2),与不同插值模型的插值精度进行评估比较。结果表明:在一定距离范围内,不同站点风速之间的相关性整体随着距离的增大而减小,将构建出的风速插值模型与其他插值模型以及未分组的直接插值方法对比,得出本文所提方法的插值效果较好,插值精度得到提升,平均MSE降至0.157 6 m/s,R^(2)提升至0.903 9。研究对探索区域风资源分布具有理论和实践意义。The distribution information of wind resources holds significant value in fields such as wind power generation and sustainable development.Based on the daily average wind speed data from 23 meteorological stations in northwest of Heilongjiang Province over a 30-year period(1989 to 2018),station classification was performed using the correlation coefficient and EMD as clustering indicators.The optimal clustering was determined through an F-test.For interpolating measurement points,appropriate stations from different categories were selected and combined with radial basis function(RBF)interpolation to construct a wind speed interpolation model based on wind speed correlation.The interpolation accuracy was evaluated and compared using mean square error(MSE)and the coefficient of determination(R 2)with different interpolation models.The results show that within a certain distance range,the correlation between wind speeds at different stations generally decreases as the distance increases.Compared with other interpolation models and ungrouped direct interpolation methods,the constructed wind speed interpolation model demonstrates better interpolation effect and improved accuracy.The average MSE decreases to 0.1576 m/s and R^(2) increases to 0.9039.

关 键 词：空间插值 径向基函数插值 推土机距离 风场 

分 类 号：O241.3[理学—计算数学]