中文摘要：

基于数值模拟解决地下水修复优化问题通常会给研究人员带来高额的计算成本。提出了一种基于参数的统计方法（逐步二次响应分析）用来创建一系列响应快速、易于使用的代理回归模型从而建立修复策略（井的抽注速率）和修复性能（污染物浓度）之间的关系。逐步二次响应分析主要有以下3个优点：它能够自动选择潜在的解释变量（各个修复井的抽注速率）;灵活检验代理回归模型中的常数项、一次项、交叉项和二次项显著性水平对各个修复情景下污染物萘的浓度的影响;减小了优化过程中产生的巨大计算工作任务。将该改进方法应用于位于安徽省某电厂受石油污染含水层识别最佳的修复策略,结果表明,当识别最佳运行条件的时候,环境标准将会严重影响抽注速率的选择。

英文摘要：

Solving groundwater remediation optimization problems would usually involve high computational cost for the researchers,when based on numerical simulators. As such,this study promotes a parametric statistical method（ SQRSA） to use rapid-response and easy-to-use surrogates to establish the relationships between remediation strategies（ e. g.,pumping rates at the wells） and remediation performance（ e. g.,contaminant concentrations）. The SQRSA has the advantages of automotive screening of potential explanatory variables（ e. g.,the pumping rates at various remediation wells）,providing a flexible manner for investigating the linear,interactive,and quadratic effects of operating conditions on the naphthalene levels,and alleviating computational efforts in the optimization processes. In this study,the developed method was applied to a petroleum-contaminated aquifer located near a power plant in Anhui Province in order to identify the optimal design for the groundwater remediation systems. The results revealed that the environmental standard significantly affects the pumping rates when the optimal remediation strategy is utilized