无锡梁溪河水质模型及其不确定性分析

发布时间：2018-01-21 02:07

本文关键词： 水质模型参数率定灵敏度分析不确定性分析　出处：《南京大学》2017年硕士论文　论文类型：学位论文

【摘要】：以梁溪河和京杭运河无锡段的网状水系为研究区,根据区域水动力路径以及水流控制条件对河网水系进行概化,基于MIKE11 AD模型建立梁溪河水质模型,使用2013-2014年水文实测资料和水质监测数据对水量-水质模型进行率定与验证。在此基础上,采用Morris法对梁溪河水质模型的率定参数进行了灵敏度分析,使用GLUE法对建立的梁溪河水质模型进行输入数据与参数的不确定性综合研究。研究结果表明:(1)MIKE11模型在研究区域具备很好的适用性。(2)河床糙率系数为第Ⅲ类敏感性系数,而扩散系数属第Ⅰ类敏感性系数。河床糙率取值越大,模拟水位越高。(3)COD的不确定性主要来源于参数n和D;TP的不确定性主要来源于扩散系数D。与COD相比,TP更容易受输入数据的不确定性的影响。无锡梁溪河的水质模型的不确定性主要来源于参数的不确定性。
[Abstract]:Taking the network water system in Wuxi section of Liangxi River and Beijing-Hangzhou Canal as the study area, the river network water system is generalized according to the regional hydrodynamic path and the water flow control conditions. Based on the MIKE11 AD model, the water quality model of Liangxi River was established, and the water quality model was determined and verified by using the hydrological data and water quality monitoring data from 2013-2014. The sensitivity of the water quality model of Liangxi River was analyzed by Morris method. The uncertainty of input data and parameters of the water quality model of Liangxi River was studied by using GLUE method. The results show that the 1 / 1 / MiKE11 model has good applicability in the study area. 2) the roughness coefficient of river bed is the third type sensitivity coefficient. The higher the roughness of the river bed, the higher the simulated water level. The uncertainty of COD is mainly derived from the parameters n and D; The uncertainty of TP mainly comes from the diffusion coefficient D. compared with COD. TP is more susceptible to the uncertainty of input data, and the uncertainty of water quality model of Liangxi River in Wuxi mainly comes from the uncertainty of parameters.
【学位授予单位】：南京大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：X52

【参考文献】