|
|||||
|
|
| 用支持向量机方法做登陆热带气旋站点大风预报 | |
| 引用本文： | 钱燕珍,孙军波,余 晖,陈佩燕.用支持向量机方法做登陆热带气旋站点大风预报[J].气象,2012,38(3):300-306. |
| 作者姓名： | 钱燕珍 孙军波 余 晖 陈佩燕 |
| 作者单位： | 1. 浙江省宁波市气象台,宁波,315012 2. 浙江省慈溪市气象局,慈溪,315300 3. 上海台风研究所,上海,200030 |
| 基金项目： | 国家科技部科研院所社会公益研究专项(2005DIB3J104)资助 |
| 摘 要： | 将支持向量机（SVM）回归方法应用于在登陆热带气旋影响下,每天00、06、12、18 UTC 4时次2分钟平均的站点风速预报。从2002-2007年热带气旋本身强度、站点地形情况和站点附近高低空环境场要素,设计相关因子,建立了4种预报模式,其中模式4的风速拟合误差的标准差为1.591 m·s~（-1）。用2008年8个登录热带气旋做独立样本检验,预报风速与实际风速的平均绝对值误差为1.750 m·s~（-1）,标准差为2.367 m·s~（-1）。结果表明,在适当的样本截取和预报因子选取后,SVM方法建模的风速预报48小时内效果较好。 |
| 关 键 词： | 支持向量机 登陆热带气旋 站点大风预报 地形 |
| 收稿时间： | 2011/2/22 0:00:00 |
| 修稿时间： | 2011/9/19 0:00:00 |
Application of SVM Method to the Station Strong Wind Forecast in Landfalling Tropical Cyclones |
|
| Qian Yanzhen,Sun Junbo,Yu Hui and Chen Peiyan.Application of SVM Method to the Station Strong Wind Forecast in Landfalling Tropical Cyclones[J].Meteorological Monthly,2012,38(3):300-306. | |
| Authors: | Qian Yanzhen Sun Junbo Yu Hui and Chen Peiyan |
| Institution: | 1 Ningbo Meteorological Observatory of Zhejiang Province,Ningbo 315012 2 Cixi Meteorological Office of Zhejiang Province,Cixi 315300 3 Shanghai Typhoon Institute/CMA,Shanghai 200030 |
| Abstract: | In this paper we choose the SVM（support vector machine） method for forecasting 2-min average wind speed four times daily（00,06,12 and 18 UTC） when it is affected by landfalling tropical cyclones. First we select related factors based on the intensity of the tropical cyclones during 2002-2007,the landform and the environment element variables at low-level and upper-level around the station.And then we establish numerical weather prediction models.The standard deviation of the wind speed fitting error in Model 4 is 1.591 m?s-1.By testing with 8 landfalling tropical cyclones in 2008 as independent samples, the difference of the actual average absolute wind speed from the forecast one is 1.75m?s-1,and the standard deviation is 2.367 m?s-1.The precision of wind speed forecast in 48 h can be better when the SVM method is used under conditions of selecting appropriate forecast factors and sample truncation. |
| Keywords: | support vector machine (SVM) landfalling tropical cyclone station wind speed forecast landform |
| 本文献已被 CNKI 万方数据 等数据库收录！ | |
| 点击此处可从《气象》浏览原始摘要信息 | |
| 点击此处可从《气象》下载免费的PDF全文 | |