暴雨灾害评估方法研究进展

Advances in the Research of Torrential Rain Disaster Assessment

在系统论述暴雨灾害评估体系结构的基础上,比较和分析了目前国内研究和应用较多的几类暴雨灾害评估理论和方法,如灰色关联法、概率统计法和主成分分析法,并以实际应用为例总结出它们各自的主要评估原理和优缺点。结果表明,灰色关联法对数据要求较低且计算量小,是比较简便易行的灾情评估方法;概率统计法以历史重现率来描述暴雨灾害等级是比较合理的方法;主成分分析法可以将多种致灾因子按权重简化成一个综合指标来描述暴雨灾害强度,比较方便直观。

On the basis of the systematic analysis of the torrential rain disaster assessment system structure,this paper compared and discussed several types of torrential rain disaster assessment methods（gray correlation method,probability and statistics method and principal component analysis）,and summarized their advances and weaknesses of each methods by taking actual application as an example.The results showed that Gray correlation method which had lower requirements for data and calculation was a more user-friendly disaster assessment method;Probability and Statistics was a more reasonable approach to describe the history reproduce rate of heavy rainfall disaster level;Principal component analysis can simplify several disaster-causing factors to a composite indicator to describe the intensity of torrential rain disasters,and it is more convenient and intuitive.