多个随机参数共存条件下地下水污染监测网多目标优化设计研究

The pameters of flow and solute transport models always contain uncertainty simultaneously in the process of groundwater contaminant monitoring using the simulation and optimization method. And, the uncertainty of these parameters will result in the uncertainty of the groundwater monitoring scheme. However, previous studies of groundwater monitoring network design have not considered the coexistence of multiple random parameters of the flow and solute transport models. Therefore, this study is committed through simulation-optimization model method to solving multi-objective optimization problem of groundwater contaminant monitoring network design under the heterogeneity of multiple random parameters. Meanwhile, a new probabilistic multi-objective evolutionary algorithm is developed to improve the previous multi-objective evolutionary algorithm to solve the proposed simulation and optimization model. The proposed probabilistic multi-objective evolutionary algorithm can achieve more efficient implementation to produce a series of the Pareto optimal solutions, which can facilitate the decision-makers in choosing the monitoring program with high precision and low monitoring cost. This study will not only proved new method for groundwater monitoring network design under the coexistence of multiple random parameters and will enriches the research of groundwater monitoring network design at home and abroad.

利用模拟-优化方法对地下水污染监测网进行优化设计时，水流模型和污染物运移模型中的参数往往同时存在不确定性，这些参数的不确定性必然会造成最终监测方案的不确定性。然而，以往对地下水污染监测网进行优化设计的研究并未全面考虑水流模型和污染物运移模型多个随机参数共存的客观事实。鉴于此，本课题致力于通过模拟-优化方法建立多个随机参数共存条件下地下水污染监测网多目标模拟-优化模型来求解地下水污染监测网多目标优化设计问题；同时通过改进现有的多目标进化算法研发随机多目标进化算法对建立的地下水污染监测网多目标模拟-优化模型进行求解，在保证高监测精度和低监测费用等多个目标实现的条件下，寻求满足约束条件的权衡解，进而为决策者提供可供选择的地下水污染监测方案。本申请项目的顺利实施不仅可以为多个随机参数共存条件下地下水污染监测网优化设计问题提供有效的求解方法，而且可以丰富国内外地下水污染监测网优化设计的研究内容。

模拟-优化方法是对地下水污染监测网进行优化设计的有效方法。针对采用模拟-优化方法对地下水污染监测网进行优化设计时水流模型和污染物运移模型中参数存在不确定性的问题，建立随机地下水污染监测网多目标模拟-优化模型，研发出高效的随机多目标进化算法，Probabilistic Pareo Genetic Algorithm （PPGA），在保证高监测精度和低监测费用等多个目标实现的条件下，为决策者提供可供选择的地下水污染监测方案。项目研究成果不仅为地下水污染监测网优化设计提供理论及技术依据，同时可为水资源模拟-优化领域的其他问题如地下水污染修复系统优化设计，提供求解思路。