基于含水层DNAPL污染修复替代模型的多目标优化研究

发布时间：2018-04-01 13:07

本文选题：表面活性剂增强含水层修复　切入点：DNAPL　出处：《中国环境科学》2016年11期

【摘要】：基于Kriging方法建立表面活性剂强化修复DNAPL污染含水层的替代模型,与混合多目标算法NSGAII-HCS(Nondominated sorting genetic algorithm II-Hill climber with step)耦合,实现修复成本最小化和治理效率最大化的多目标优化.以三维非均质承压含水层中PCE污染物的运移与修复过程为例,采用UTCHEM程序模拟表面活性剂强化修复含水层过程.将Kriging替代模型与多相流模型的输出结果进行对比,两种模型得到的含水层中PCE去除效率的平均相对拟合误差为0.80%,相关系数为0.9992,表明Kriging模型可以有效替代多相流模型.进一步将替代模型的Pareto最优解与相应的多相流模型的模拟值进行比较,得到两种模型的平均相对拟合误差仅为0.70%,相关系数达0.9998,表明在多目标优化的迭代求解过程中可以直接调用Kriging替代模型,而无须重复调用多相流模型的大负荷运算,从而为制定表面活性剂强化含水层修复决策提供一种稳定可靠的多目标优化方法.
[Abstract]:Based on the Kriging method, an alternative model of surfactant enhanced restoration of DNAPL polluted aquifer was established and coupled with the hybrid multiobjective algorithm NSGAII-HCS(Nondominated sorting genetic algorithm II-Hill climber with step. Multi-objective optimization to minimize the cost of restoration and maximize the efficiency of remediation. Taking the transport and remediation process of PCE pollutants in a three-dimensional heterogeneous confined aquifer as an example. The UTCHEM program was used to simulate the surface active agent enhanced aquifer restoration process. The output results of the Kriging substitution model and the multiphase flow model were compared. The average relative fitting error and correlation coefficient of PCE removal efficiency obtained from the two models are 0.80 and 0.9992 respectively, indicating that the Kriging model can effectively replace the multiphase flow model. Further, the Pareto optimal solution of the replacement model and the corresponding multiphase flow can be replaced by the Kriging model. The simulated values of the model are compared, The average relative fitting error of the two models is only 0.70 and the correlation coefficient is 0.9998, which indicates that Kriging can be directly used to replace the model in the iterative solution of multi-objective optimization, without repeating the heavy load operation of the multiphase flow model. Therefore, a stable and reliable multi-objective optimization method is provided for the decision-making of surfactant-enhanced aquifer restoration.
【作者单位】：南京大学地球科学与工程学院水科学系表生地球化学教育部重点实验室;河海大学地球科学与工程学院;
【基金】：国家重点研发计划项目(2016YFC0402807) 国家自然科学基金资助项目(41372235,41402198,U1503282)
【分类号】：X523

【相似文献】