基于Weibull分布的最大风速变点估计  

Change Point Estimation of Annual Maximum Wind Speed Based on Weibull Distribution

作　　者：戴迪昊 钱泽平 丁明月 DAI Di-hao;QIAN Ze-ping;DING Ming-yue(School of Mathematics,Hefei University of Technology,Hefei 230009,China)

机构地区：[1]合肥工业大学数学学院,合肥230009

出　　处：《大学数学》2018年第5期12-18,共7页College Mathematics

基　　金：国家社科基金年度项目(16BTJ023)

摘　　要：利用三参数的Weibull分布分析安徽砀山气象站年最大风速数据,建立似然比变点模型和回归变点模型,并对年最大风速序列的变点进行检验和估计.考虑非气象因素对年最大风速序列的影响,选取A气象站附近数个气象站的年最大风速序列作为参考序列,通过比值法,消除气象因素的影响.计算结果表明,由似然比变点模型可以估计出A气象站最大风速变点出现在2003年,由回归变点模型估计出的变点出现在2009年和2003年.结合安徽砀山气象站实际迁移和海拔变化的历史沿革,可以发现,基于Weibull分布的变点模型成功检测出了2009年台站迁移和2003年前后台站迁移且海拔显著变化的情况.Three-parameter Weibull distribution function will be used to fit the annual maximum wind speed data of A station.On this basis,a likelihood ratio model and a regression model will be established to estimate the change point of the series.In order to study the change point caused by non-meteorological factors of the maximum wind speed sequence,the annual maximum wind speed information of several stations near A station will be selected as the reference sequence,and compared with the original sequence of A station to eliminate the influence of meteorological factors.The results show that the change point of maximum wind speed series of A station can be detected at 2003 by the likelihood ratio model,and the change points estimated by the regression model are at 2009 and 2003.Combined with the actual record of A station,it can be found that the change point model based on Weibull distribution successfully estimate the station relocation in 2009 and the significant change of altitude around 2003.

关 键 词：WEIBULL分布 似然比 回归分析 变点模型 风速 

分 类 号：O212.1[理学—概率论与数理统计]