基于LH-OAT方法的VG模型参数敏感性分析  

Sensitivity Analysis of VG Model Parameter Based on LH-OAT Method

作　　者：秦萍[1,4] 王正 孙兆军 禹昭[2,3] QIN Ping;WANG Zheng;SUN Zhao-jun;YU Zhao(Institute of Environmental Engineering,Ningxia University,Yinchuan 750021,China;School of Civil Engineering and Hydraulic Engineering,Ningxia University,Yinchuan 750021,China;Xinhua College,Ningxia University,Yinchuan 750021,China;China Ningxia(China-Arab)Key Laboratory of Resources Assessment and Environmental Regulation in Arid Regions,Yinchuan 750021,China)

机构地区：[1]宁夏大学环境工程研究院,银川750021 [2]宁夏大学土木与水利工程学院,银川750021 [3]宁夏大学新华学院,银川750021 [4]宁夏(中阿)旱区资源评价与环境调控重点实验室,银川750021

出　　处：《节水灌溉》2019年第10期97-102,共6页Water Saving Irrigation

基　　金：宁夏回族自治区重点研发计划重大(重点)项目(2018BFG02016);国家自然科学基金项目(31460220)

摘　　要：通过研究VG模型中参数变化对输出结果的影响,不仅可以提高率定模型参数的效率,而且可以为获取更加可信及稳定的数学模型提供支持。以宁夏中部平原引黄灌区2017年林地田间试验为验证实例,选取了VG模型中11参数进行了全局敏感性分析。对每个参数都进行了全局敏感度的定量计算,依照敏感度大小将它们进行敏感性的分级。结果表明:以Se为输出变量时,参数θsa、αa和na表现极为敏感,而αb、θra、nb和θsb表现较为敏感,但Ksa、Ksb和l的表现却为一般敏感或不敏感;而以K(Sa)为输变量出时,参数θsa表现最为敏感,而αa、na、Ksa表现较高的敏感,但其他参数敏感表现为较弱;对表现比较敏感的参数进行了修正,并将修正后的参数应用于VG模型中,取得了良好的效果,而且所构建的全局敏感性分析方法可有效地提高VG模型参数率定效率,田间普适性较强。By studying the influence of parameter changes in VG model on output results,not only the efficiency of calibration model parameters can be improved,but also a more reliable and stable mathematical model can be obtained.In this study,the forest field test in the yellow irrigation area of the central Ningxia plain in 2017 was taken as a verification example,and 11 parameters of VG model were selected for global sensitivity analysis.The global sensitivity of each parameter is quantitatively calculated and graded according to the sensitivity.The results show that:when taking Se as the output variable,the performance of parameters,θsa,αa and na,is extremely sensitive,while αb,θra,nb and θsb are relatively sensitive,but the performance of Ksa,Ksb and l is generally sensitive or insensitive;when the output variable is K(Se),the parameter θsa is the most sensitive,while αa,na and Ksa are relatively sensitive,but other parameters are relatively weak.The more sensitive parameters are modified and applied to VG model,and good results have been achieved.Moreover,the global sensitivity analysis method can effectively improve the efficiency of parameter calibration of VG model,and the field universality is strong.

关 键 词：拉丁超立方 VG模型 参数敏感性 土壤含水率 

分 类 号：S152.71[农业科学—土壤学]