首页 | 官方网站   微博 | 高级检索  
     

基于投影寻踪回归的规范指标的气象灾情评估
引用本文:李祚泳,徐源蔚,汪嘉杨,刘韵.基于投影寻踪回归的规范指标的气象灾情评估[J].应用气象学报,2016,27(4):480-487.
作者姓名:李祚泳  徐源蔚  汪嘉杨  刘韵
作者单位:成都信息工程大学资源环境学院, 成都 610225
摘    要:科学合理地评估气象灾害造成的损失, 对减灾、防灾决策具有重要意义。为了建立不同气象灾害系统均能适用的投影寻踪回归 (projection pursuit regression, PPR) 矩阵表示灾情评估模型,在对气象灾情指标值进行规范变换基础上,规范变换后的各指标皆等效于同一个规范指标,因而只需构建对任意2个规范指标值适用的投影寻踪回归 (projection pursuit regression based on normalized index value, NV-PPR) NV-PPR (2) 模型和3个规范指标值适用的投影寻踪回归NV-PPR (3) 模型;对3个以上指标的气象灾害系统NV-PPR建模,只需将其分解为若干个NV-PPR (2) 模型和 (或) NV-PPR (3) 模型的组合即可。模型用于广东省台风和重庆市两起雷电灾情评估,并与其他方法的评估结果进行比较。结果表明:基于NV-PPR的气象灾情评估模型不受指标数量的限制,具有简单、实用的特点,该模型还可推广为适用于其他灾情评估。

关 键 词:规范变换    指标规范值    投影寻踪回归    气象灾害    灾情评估
收稿时间:2016-01-15

Evaluation Model of Meteorological Disaster Loss with Normalized Indices Based on Projection Pursuit Regression
Li Zuoyong,Xu Yuanwei,Wang Jiayang and Liu Yun.Evaluation Model of Meteorological Disaster Loss with Normalized Indices Based on Projection Pursuit Regression[J].Quarterly Journal of Applied Meteorology,2016,27(4):480-487.
Authors:Li Zuoyong  Xu Yuanwei  Wang Jiayang and Liu Yun
Affiliation:College of Resources and Environment, Chengdu University of Information Technology, Chengdu 610225
Abstract:Scientific and reasonable evaluation of meteorological disaster loss has important significance for decisions of disaster reduction and relief. A universal and general model of projection pursuit regression (PPR) is proposed with the matrix for the evaluations of different meteorological disaster systems, on the basis of gauge transformation for disaster loss indexes. Because of the "equivalence" of each normalized index, only models of NV-PPR (2) and NV-PPR (3) are necessary for the normalized index values (NV) of 2 indices and 3 indices of meteorological disaster system. Furthermore, the NV-PPR modeling for over 3 indexes of meteorological disaster system could be represented by the combinations of some NV-PPR (2) and (or) NV-PPR (3) models. Models are applied to the evaluations of typhoon disaster loss in Guangdong and 2 lightning disasters loss in Chongqing, and evaluation results of this method are compared with those of other methods. It shows that the evaluation model (NV-PPR) of meteorological disaster loss with gauged transformation based on projection pursuit regression is independent of index numbers, with features of simplicity and utility. The model can also be extended and apply to other disaster loss assessment systems.
Keywords:gauged transformation  normalized index value  projection pursuit regression  meteorological disaster  evaluation of disaster loss
本文献已被 CNKI 等数据库收录!
点击此处可从《应用气象学报》浏览原始摘要信息
点击此处可从《应用气象学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号