首页|基于SVR台风距离与最佳风暴潮增水影响因素组合的关系研究

基于SVR台风距离与最佳风暴潮增水影响因素组合的关系研究

扫码查看
风暴潮增水通常会造成巨大灾害和损失,对其预测准确性的研究对防灾减灾部门的灾害防御工作具有重大现实意义.文章基于永兴岛验潮站和杨柳等三个台风,收集了上一时刻风暴潮增水值、验潮站与台风直线距离等8项数据,使用支持向量回归方法(SVR),建立支持向量回归风暴潮增水预测模型,通过不断改变模型的输入,以获得最佳风暴潮增水影响因素组合.试验结果表明预测模型的最佳影响因素组合在一定程度上受到验潮站-台风直线距离的影响,当验潮站位于台风风圈半径边缘及外侧时,台风位置对预测模型的影响较大,且进行距离细化后的预测模型准确度可提高32.76%.
A study on the relationship between typhoon distance and optimum storm surge factors based on SVR
Storm surge usually causes huge disasters and losses,and it is great practical significant for disaster mitigation departments to prevent disaster.Eight kinds of data,such as last hour storm surge,the straight line distance between the tide stations and three typhoons,were collected about Yongxing island tide station,and a support vector regression prediction model for storm surge was built.The ex-perimental results showed that the best combination of influencing factors of the prediction model was affected to a certain extent by the straight line distance between tide station and typhoon.When the tide station was located on the edge and outside of the typhoon circle radius,and the typhoon position had a great influence on the prediction model.The accuracy of the prediction model can be improved by 32.76%.

storm surgepredictionsupport vector regressionoptimal influencing factors combination

张晓琪、刘强

展开 >

中国海洋大学工程学院土木工程系,山东青岛 266100

风暴潮增水 预测 支持向量回归 最佳影响因素组合

国家自然科学基金国家自然科学基金国家科技支撑计划

41072176413714962013BAK05B04

2024

海洋湖沼通报
山东海洋湖沼学会

海洋湖沼通报

CSTPCD北大核心
影响因子:0.464
ISSN:1003-6482
年,卷(期):2024.46(2)
  • 30