摘要
应用EC细网格资料和站点实况资料,结合风速高度订正和分风向误差风速分析,建立考虑地形的PP法模型。释用结果表明:考虑地形的PP法模型能有效提高EC细网格产品大风预报精度,平均绝对误差低至2.57 m/s;应用卡尔曼滤波法实时吸收最新资料,能实现释用模型回归系数的滚动更新,预报风速平均绝对误差由2.91 m/s降至2.66 m/s。
A PP method model considering topography is established in this paper using the EC fine grid data and site observations combined with wind speed height correction and wind direction error analysis,The interpretation results show that the accuracy of the EC fine-grid gale products can be effectively improved with the mean absolute error as low as 2.57 m/s by using the PP model considering the topography.The latest data can be absorbed in real time by using the Kalman filter method,which realizes the rolling update of the regression coefficients of the interpretation model,and the mean absolute error of forecast wind speed is reduced from 2.91 m/s to 2.66 m/s.
作者
於敏佳
刘菡
王倩
胡潇杰
YU Min-jia;LIU Han;WANG Qian;HU Xiao-jie(Zhoushan Meteorological Bureau,Zhoushan 316000 China)
出处
《海洋预报》
CSCD
北大核心
2021年第2期55-61,共7页
Marine Forecasts
基金
舟山市公益类科技项目(2019C31055)。
关键词
EC细网格
风速误差
高度订正
PP法
地形
卡尔曼滤波法
EC fine grid
wind speed error
altitude correction
PP method
terrain
Kalman filter method
