基于改进的拉丁超立方参数随机地下水数值模拟研究

发布时间：2018-04-18 14:51

本文选题：参数随机 + 改进的拉丁超立方　；参考：《中国地质大学(北京)》2015年硕士论文

【摘要】：数值模拟是研究地下水流动及溶质运移规律的重要手段,多孔介质的空间变异性使得确定性的数值模拟在刻画复杂的水文地质结构方面存在局限性,本文进行地下水流随机模拟和溶质运移随机模拟,模拟开采条件下的傍河水源地地下水水位和溶质浓度分布。本文首先综述了相关领域的研究进展,进而尝试用对偶变数改进LHS,并借助c#语言完成改进的LHS程序的编写,对比两种采样方法样本的统计特征和分布特征,结果表明改进的LHS样本离散程度增加。在浑河冲洪积扇的相关资料的基础上,建立水文地质概念模型,用两种方法获取渗透系数、孔隙度、弥散度的随机样本,继而建立参数随机水流模型和参数随机溶质运移模型,通过对模拟水头和模拟溶质浓度统计量和分布特征的分析,改进的LHS方法的水头标准差偏小,与改进的LHS样本的分布特征和水头对渗透系数的变化率有关,改进的LHS方法的溶质浓度标准差偏小,与改进的LHS样本的分布特征和溶质浓度对孔隙度的变化率有关。最后参照《地下水质量标准》,用地下水溶质运移随机模拟预测了研究区内污染物超标浓度的覆盖范围,为地下水污染的治理与修复了提供必要的依据。
[Abstract]:Numerical simulation is an important method to study the law of groundwater flow and solute migration. The spatial variability of porous media makes deterministic numerical simulation have limitations in depicting complex hydrogeological structures.In this paper, stochastic simulation of groundwater flow and solute transport is carried out to simulate the distribution of groundwater level and solute concentration in the riverside water source under the condition of mining.This paper first summarizes the research progress in related fields, and then attempts to improve the LHSs with dual variables, and completes the improved LHS program with the help of c# language, and compares the statistical characteristics and distribution characteristics of the samples of the two sampling methods.The results show that the improved LHS sample dispersion increases.Based on the relevant data of the alluvial fan of the Hunhe River, a hydrogeological conceptual model is established. The random samples of permeability coefficient, porosity and dispersion are obtained by two methods, and then the parametric stochastic flow model and the parametric stochastic solute transport model are established.By analyzing the statistical and distribution characteristics of simulated water head and simulated solute concentration, it is found that the standard deviation of water head of the improved LHS method is small, which is related to the distribution characteristics of the improved LHS sample and the change rate of the water head to the permeability coefficient.The standard deviation of solute concentration in the improved LHS method is small, which is related to the distribution characteristics of the modified LHS sample and the change rate of solute concentration to the porosity.Finally, with reference to "groundwater quality Standard", the coverage of pollutants exceeding the standard concentration in the study area is predicted by stochastic simulation of solute transport in groundwater, which provides the necessary basis for the treatment and remediation of groundwater pollution.
【学位授予单位】：中国地质大学(北京)
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P641;X523

【参考文献】