摘要
风暴潮灾害是制约我国海洋开发和海洋经济发展的重要因素,也是海洋预防的重点难点问题。在风暴潮灾害损失评估这一领域,对组合评估模型的研究较少。以浙江省为例,整理收集了1990—2020年间记录完整的29个风暴潮历史灾情数据,利用麻雀搜索算法(Sparrow Search Algorithm,SSA)分别对支持向量回归(Support Vector Regression,SVR)和广义回归神经网络(Generalized Regression Neural Network,GRNN)进行模型优化改进,得到的直接经济损失拟合结果R^(2)值分别为0.8473和0.8828,将改进后的SVR和GRNN建立组合模型,用于风暴潮灾害损失评估,R^(2)提高为0.9190。结果表明:组合模型的评估精度优于单一模型,是一种适用于风暴潮灾害损失评估的高精度、稳定性方法。
Storm surge disaster is an important factor restricting the development of my country’s oceans and ocean economy,and it is also a key and difficult problem of ocean prevention.In the field of storm surge disaster loss assessment,there are few studies on the combined assessment model.Taking Zhejiang province as an example,this paper collected 29 complete storm surge historical disaster data from 1990 to 2020,and used the Sparrow Search Algorithm(Sparrow Search Algorithm,SSA)to model the Support Vector Regression(Support Vector Regression,SVR)and Generalized Regression Neural Network(Generalized Regression Neural Network,GRNN)respectively.Optimized and improved,the obtained direct economic loss fitting results were 0.8473 and 0.8828,respectively.The improved SVR and GRNN were used to establish a combined model for storm surge disaster loss assessment,which was increased to 0.9190.The results showed that the assessment accuracy of the combined model was better than that of the single model,and it was a high-precision and stable method for storm surge disaster loss assessment.
作者
贾丙宏
杨帅
冉姝
JIA Binghong;YANG Shuai;RAN Shu(College of Geodesy and Geomatics,Shandong University of Science and Technology,Qingdao Shandong 266590,China;China Jinan Research Institute of Surveying and Mapping,Jinan Shandong 250000,China;Publicity Department of Juye County Party Committee,Heze Shandong 274900,China)
出处
《北京测绘》
2022年第1期1-6,共6页
Beijing Surveying and Mapping
基金
国家重点研发计划(2017YFC1405005)。
