地下水污染监测网多目标优化设计模型及进化求解

An evolutionary-based multi-objective optimization model for groundwater monitoring network design

采用模拟-优化方法建立了一个用于地下水污染监测网设计的多目标优化模型,该模型包括最小化监测费用、污染物质量评估误差、污染羽一阶矩评估误差和二阶矩评估误差等4个目标函数,以充分揭示减少地下水污染监测费用与提高污染监测精度之间的权衡关系。将改进小生境Pareto遗传算法与地下水流模拟程序和污染物运移模拟程序相耦合用于求解地下水污染监测网多目标设计模型。算例研究表明,采用进化算法求解监测网的多目标模型,能真实地反映各个目标函数间的权衡关系,并且不用考虑传统方法中惩罚因子的影响。与单目标优化模型相比,多目标优化模型可在较短的时间内得到优化问题的一系列Pareto权衡解,以利于相应条件下决策者选择最为经济有效的地下水污染监测方案。

In this paper,a multi-objective simulation-optimization model is developed for the optimal design of groundwater sampling network for contaminant plume monitoring.The multi-objective simulation-optimization model includes four objectives:(1) minimization of total sampling and analysis costs for contaminant plume monitoring,(2) minimization of mass estimation error of the plume,(3) minimization of the first moment estimation error of the plume,and(4) minimization of the second moment estimation error of the plume,to adequately balance the trade-off between decreasing the monitoring / sampling cost and increasing the estimation accuracy of plume monitoring.A multi-objective evolutionary algorithm,the improved niched Pareto genetic algorithm(INPGA),is also combined with the groundwater flow and transport simulation model to solve the multi-objective optimization problem associated with groundwater monitoring network design for contaminant plume monitoring.Moreover,application of the proposed methodology to a hypothetical monitoring network design problem show that the optimization results can really reflect the trade-off between the various objectives,while avoiding the impacts of the subject-dependent penalty factors used in the traditional methods on the optimization results.Compared with the single objective optimization model,the multi-objective optimization model can achieve more efficient implementation to produce a series of the Pareto optimal solutions,which can facilitate the decision-makers in choosing the most cost-effective monitoring strategy consistent with the actual field conditions.