结合风向分布的新疆西北地区极值风速变化研究——以博尔塔拉蒙古自治州富蕴县为例  

Study on Extreme Wind Speed Changes in Northwest Xinjiang Combined with Wind Direction Distribution:In Case of Fuyun County in Bortala Mongolian Autonomous Prefecture

作　　者：安晓燕 闫在在[1] AN Xiao-yan;YAN Zai-zai(School of Science,Inner Mongolia University of Technology,Hohhot 010051,China)

机构地区：[1]内蒙古工业大学理学院,内蒙古呼和浩特010051

出　　处：《数理统计与管理》2023年第4期571-580,共10页Journal of Applied Statistics and Management

基　　金：国家自然科学基金项目(11861049)。

摘　　要：本文基于Copula函数建立联合分布的方法,以新疆西北地区富蕴县的数据为例,研究考虑风向的极值风速概率分布。首先,利用非参数核密度估计模型对极值风速概率密度进行研究,选择高斯核密度和Scott无偏交叉验证法建立的模型对该地风速分布拟合效果优于传统威布尔分布和广义极值分布;其次,运用von Mises分布对风向变化规律进行拟合,4阶的von Mises分布对该地风向分布的拟合效果优于2阶、3阶模型;最后,建立基于Frank-Copula函数的风速风向联合概率密度函数,分析表明研究风速分布时考虑风向的作用是十分必要的。另外,利用模型对富蕴县50年重现期的极值风速和风能密度预测,结果发现未来南部方向的风速较小,达到12.051m/s,对应的风能密度较小,北部偏东和偏西两个方向的风速较大,最大风速达到32.548m/s,对应的风能密度也较大。如果不考虑风向的影响会导致高估该地区风速,对应风能密度的预测也存在较大差异。与传统的风速模型相比,本文提出的Frank-Copula模型的研究结果对预测富蕴县风速有一定的参考价值,同时也对研究其他地区风速分布提供了新思路。This paper studies the probability distribution of extreme wind speed considering wind di-rection based on the Copula function,taking the data of Fuyun County in northwestern Xinjiang as an example.Firstly,the non-parametric kernel density estimation model,which is building by Gaussian kernel density and Scott's unbiased cross-validation method,is used to study the wind speed distribution and the model is better than the traditional Weibull and generalized extreme value distributions.Sec-ondly,we model the wind direction distribution,called the von Mises distribution,and the fitting effect of the 4-order von Mises model is better than the 2 and 3-order models.Finally,the joint dist ribution based on the Frank-Copula function is established.The analysis result shows that it is necessary to consider the role of wind direction.In addition,during the 50-year return period,it is found that the wind speed in the northeast and northwest directions will be higher,the maximum wind speed will reach 32.548m/s and the corresponding wind power density is also higher.The opposite direction is smaller.If the infuence of wind direction is not considered,the result is the opposite.The Frank-Copula model proposed in the paper fits the wind speed distribution more reasonably,which has certain value for the related departments.

关 键 词：核密度估计模型 von Mises分布 COPULA函数 风速风向联合概率分布 

分 类 号：O212[理学—概率论与数理统计] X8[理学—数学]