基于遗传算法的SWMM模型参数率定研究  

Parameter Calibration of SWMM Model Based on GA

作　　者：马彦斌[1] 盛旺[1] 李江云[1] 代文江 MA Yan-bin;SHENG Wang;LI Jiang-yun;DAI Wen-jiang(Wuhan University, Wuhan 430072, China)

机构地区：[1]武汉大学,武汉430000

出　　处：《中国农村水利水电》2020年第7期46-49,53,共5页China Rural Water and Hydropower

基　　金：广东省水利科技创新项目(2017-33)。

摘　　要：针对SWMM模型参数繁多,率定困难的问题,采用Morris法进行全局敏感性分析,选取敏感参数,基于MATLAB遗传算法对敏感参数进行率定。案例研究结果表明,遗传算法能有效地完成率定过程,率定后各参数的取值分别为最大渗透速率50.25,最小渗透速率12.61,管道曼宁系数0.0182,不透水区曼宁系数0.0087,该场次降雨下目标函数Nash-Sutcliffe效率系数值为0.7261。该方案可以应用于实际工程的参数率定过程,并得出与实测结果最接近的参数组合,提高模拟的精度及稳定性。To solve the problem of SWMM model with many parameters and hard to calibration,Morris method is promoted to deal with global sensitivity analysis,select sensitive parameters and calibrate sensitive parameters based on MATLAB genetic algorithm.The results of the case study show that the genetic algorithm can effectively complete the process of rate setting.After rate setting,the values of each parameter are the maximum permeability rate 50.25,the minimum permeability rate 12.61,the manning coefficient 0.0182 in the pipeline,and the manning coefficient 0.0087 in the impermeable area.The method can be applied to the process of parameter calibration and can give a set of parameters with the field measure results which can improve the accuracy and stability of simulation.

关 键 词：SWMM 遗传算法 参数率定 敏感性分析 

分 类 号：TU99[建筑科学—市政工程]