洪水灾情评价的遗传程序设计方法  

Genetic programming method for evaluation of flood disaster loss

作　　者：周玉良[1] 金菊良[1] 魏一鸣[2] 汪明武[1] 

机构地区：[1]合肥工业大学土木建筑工程学院,安徽合肥230009 [2]中国科学院科技政策与管理科学研究所,北京100080

出　　处：《农业系统科学与综合研究》2004年第4期241-245,共5页System Sciemces and Comprehensive Studies In Agriculture

基　　金：国家自然科学基金重大项目(50099620);教育部优秀青年教师资助计划(教人司[2002]350);四川大学高速水力学国家重点实验室开放基金(0201);安徽省优秀青年科技基金;安徽省自然科学基金(01045102;01045409)

摘　　要：洪水灾情评价在洪水灾害分类管理决策中具有重要的指导意义,其关键问题就是在评价模型的结构未确知时,如何建立合适的模型来描述以灾情评价指标值为系统输入与以灾情评价等级值为系统输出之间所隐含的复杂关系。为此,提出了用FORTRAN语言实现洪水灾情评价的遗传程序设计新方法(GP-EFDL)。研究结果说明,GP-EFDL可在较大的程序搜索空间中获取一个最佳函数表达式,来反映洪水灾情评价系统的输入与输出样本数据系列所隐含的关系,在自然灾害系统建模、预测和控制等计算机程序自动设计中具有应用价值。表4,参13。Evaluation of flood disaster loss is very important to guiding decision-making of classified management of flood disaster, which key problem is how to establish a right model for describe the complex relation hidden in input and output data series of flood disaster loss system when the model structure of the complex flood disaster loss system has been unknown. For this reason, a new method, named GP-EFDL for short, is presented in this paper for evaluation of flood disaster loss based on genetic programming by using FORTRAN programming language. The result of case study shows that a good function expression can be obtained within big programming search space with using GP-EFDL method, which expression can reflect the complex relation hidden in system input and output data series of flood disaster loss system, and that the method can be widely used in automation of computer programming of modeling, prediction and control of natural disaster systems.

关 键 词：洪水灾情评价 遗传程序设计 遗传算法 自然灾害管理 

分 类 号：S422[农业科学—植物保护]