基于集合平滑器的地下水污染源与含水层参数的同步反演  

Simultaneous Identification of Groundwater Contaminant Source and Aquifer Parameters Based on Ensemble Smoother

作　　者：张春秋 江思珉[1,2] 张瑞城 刘金炳 孔维铭 栗现文 

机构地区：[1]同济大学土木工程学院水利工程系,上海 [2]南京水利科学研究院水文水资源与水利工程科学国家重点实验室,江苏 南京 [3]长江生态环保集团有限公司,湖北 武汉 [4]西北农林科技大学水利与建筑工程学院,陕西 西安

出　　处：《地球科学前沿（汉斯）》2021年第11期1466-1475,共10页Advances in Geosciences

摘　　要：当地下水污染发生时,能够准确及时地得到污染源信息及场地的渗透性分布,对于污染修复方案的制定尤为重要。但这些信息无法直接观测获取,只能通过有限的观测数据来推估。本文利用集合数据同化方法ILUES算法作为求解框架进行地下水污染源与含水层参数的同步反演,重点探讨样本集合数目和观测信息对地下水污染源反演和含水层渗透系数场估计结果的影响。算例结果表明,融合观测水头数据和浓度数据的ILUES算法框架能够实现污染源参数和渗透系数场的同步反演;作为数据同化方法,ILUES算法参数(集合样本数量)和观测信息(观测井数量和空间分布等)都直接影响着地下水污染源反演和渗透系数场估计的准确性。When groundwater pollution occurs, it is particularly important to be able to obtain accurate and timely information on the source of contaminant and the hydraulic conductivity field of the site for the development of contamination remediation plans. However, this information cannot be obtained by direct observation and can only be deduced by limited measurement data. In this paper, the ensemble data assimilation method (ILUES algorithm) is adopted as the solution framework for simultaneous identification of groundwater contaminant source and aquifer parameters, the influence of sample set number and observation information on groundwater contaminant source identification and estimation of aquifer hydraulic conductivity field is discussed. The results show that the ILUES algorithm framework, which assimilates the observation data (hydraulic head and concentration data), can achieve the simultaneous inversion of contaminant source parameters and hydraulic conductivity field. As a data assimilation method, ILUES algorithm parameters (the number of ensemble samples) and observation information (the number and locations of monitoring wells, etc.) directly affect the accuracy of groundwater contaminant source identification and hydraulic conductivity field estimation.

关 键 词：地下水逆问题 污染源反演 渗透系数场 集合平滑器 数据同化 

分 类 号：TP3[自动化与计算机技术—计算机科学与技术]