沿海地区风暴潮灾害风险评估与区划——基于混合算法优化的PPDC模型

以沿海11省市的风暴潮灾害风险为研究对象,采用遗传与粒子群混合算法对投影寻踪动态聚类(PPDC)模型进行优化,将粗糙集理论(RST)与修正的PPDC模型组合运用,对中国沿海地区风暴潮灾害的风险进行综合评估与区域等级划分。实证结果表明:广东和福建两省是中国风暴潮灾害的高风险区,风险评估值超过2.5,山东、浙江、海南和广西属于风暴潮灾害的中风险区,风险评估值处于1.8,2.2]之间,江苏、天津、辽宁、河北和上海属于风暴潮灾害的低风险区,风险评估值低于1.5。研究结论为国家实施差异化的灾害风险管理战略提供了思路与参考。

Risk Assessment and Regional Classification of Storm Surge Disasters of Coastal Areas:Based on PPDC Model with Hybrid Optimization Algorithm

This paper combines Rough Set Theory (RST ) with Projection Pursuit Dynamic Cluster (PPDC) model optimized by Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) to assess the comprehensive risk and make regional classification of storm surge disasters of eleven provinces and cities of Coastal Areas .The empirical results are as follows :Guangdong and Fujian are the high risk area of storm surge disaster in China , w hose risk assessment scores are both more than 2 .5 . Shandong , Hainan ,Guangxi and Zhejiang belong to the medium risk area of storm surge disaster and their risk assessment scores are between 1 .8 and 2 .2 .Tianjin ,Liaoning ,Hebei ,Shanghai and Jiangsu belong to the low risk area of storm surge disaster w hose risk assessment scores are all less than 1 .5 .T he conclusion of this thesis is to provide a reference for the implementation of the risk management strategy of China .