项目名称: 地下水区间型不确定性数值模拟
项目编号: No.41202179
项目类型: 青年科学基金项目
立项/批准年度: 2013
项目学科: 天文学、地球科学
项目作者: 董贵明
作者单位: 中国矿业大学
项目金额: 23万元
中文摘要: 地下水数值模拟中的各种不确定性,直接影响了模拟结果的可靠性,因此,开展不确定性数值模拟,深入分析不确定性就显得尤为迫切和必要。本课题以徐州张集岩溶地下水源地为研究对象,从区间型不确定性角度出发,基于非概率集合理论,开展地下水区间型不确定性数值模拟研究,分析地下水数值模拟的不确定性。采用凸模型方法和区间分析方法求解数值模型,使用区间参数摄动方法反演模型参数、边界条件和源汇项,用多尺度全变分正则化方法处理反演的不适定和效率问题。区间型不确定性明显区别于概率及其它不确定性处理方法,不需要知道变量的概率密度函数,基于较少的数据(不确定变量的上下界)就可以解决一系列不确定性问题,并得到令人满意的结果。地下水区间型不确定性数值模拟开辟了新的不确定性分析途径。 通过本课题的研究,将从区间型不确定性角度,揭示地下水数值模拟的不确定性响应规律,增加数值模拟不确定性新的认识,提高数值模拟的可靠性。
中文关键词: 地下水;区间;不确定性;数值模拟;非概率集合理论
英文摘要: The reliability of groundwater numerical simulation is vulnerable to many uncertain factors. Thus, it is necessary to quantify the uncertainties in groundwater simulation. Interval uncertainty method is used into Zhangji groundwater source area, Xuzhou in Jiangsu province, to analyze the uncertainties of groundwater numerical simulation based on non-probabilistic theory. The numerical model is mainly solved by convex model theory and interval method. Its parameters, boundary conditions and source terms are inversed with interval parameter perturbation algorithm. During the inversion, we use multi-scale regularization method to deal with the ill-posed problems and efficiency. Interval uncertainty method is a good method to solve kinds of reliabilities with fewer dates with unknown probability density function of variables, which is different from probability and other uncertainty methods. It is a new method to simulate groundwater with interval uncertainty, which will reveal the uncertainties of groundwater numerical simulation and increase its reliability better.
英文关键词: Groundwater;interval;uncertainty;numerical;non-probabilistic theory