基于模式匹配的飓风轨迹预测方法  

Hurricane Trajectory Prediction Method with Pattern Matching

作　　者：董鑫[1] 皮德常[1] 

机构地区：[1]南京航空航天大学计算机科学与技术学院,南京210016

出　　处：《小型微型计算机系统》2014年第5期983-988,共6页Journal of Chinese Computer Systems

基　　金：国家"八六三"高技术研究发展计划项目(2007AA01Z404)资助;航空科学基金项目(20111052010)资助;江苏省"青蓝工程"项目资助;江苏省"333工程"项目资助

摘　　要：结合飓风移动轨迹及其自身的运动特点,本文提出了一种基于模式匹配的飓风轨迹预测方法.首先采用关联分析法对飓风的历史轨迹进行频繁轨迹挖掘,生成代表飓风运动模式的规则;然后将飓风当前运动的轨迹与运动模式进行匹配,实现轨迹预测;如果匹配失败,则根据飓风当前的运动趋势返回一个预测结果.采用1900-2008年美国大西洋飓风数据进行了实验和分析,取得了比较满意的实验结果:若包含模式匹配失败部分,其预测正确率为57.5%;若模式匹配全部成功,则该方法的预测正确率为65%.Combined with the hurricane movement and its own motion characteristics, this paper proposes a hurricane trajectory predic- tion method based on pattern matching. Firstly, all frequent trajectories in historical hurricane trajectory database are mined by using association analysis technology and the association rules are generated as motion patterns. Then, the current hurricane trajectories are matched with the motion patterns for predicting. If the match falls, it will return a forecasting result according to the current trend of the hurricane movement. All experiments are conducted with the Atlantic weather Hurricane/Tropical Data froml900 to 2008. The experimental results are satisfactory： the prediction accuracy rate is 57.5% if the matching failure part is contained and if all matches are successful, the accuracy rate is 65%.

关 键 词：飓风 关联规则 轨迹预测 模式匹配 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]