混合多目标遗传算法求解地下水污染修复管理模型

为了提高多目标遗传算法Pareto解的局部最优性,本文将快速非支配遗传算法(NSGAII)与一种迭代式的局部搜索算法(Hill Climber with Step,HCS)相结合,开发了一种新的混合多目标遗传算法NSGAII-HCS.利用CONV1和ZDT6两个经典的多目标优化函数对NSGAII-HCS的性能进行测试,与传统的多目标算法NSGAII相比,CONV1得到的Pareto锋面与真实Pareto最优解锋面的平均距离由5.49减小到1.74,ZDT6则由0.16减小到0,表明NSGAII-HCS在保证解多样性的前提下,能使解接近或收敛到真实的Pareto最优解锋面.最后,将NSGAII-HCS与地下水流模拟软件MODFLOW和溶质运移模拟软件MT3DMS相耦合,并应用到一个理想的二维地下水污染修复管理模型中,结果分析表明该方法可为地下水污染治理提供多样的和收敛的Pareto管理策略,是一种稳定可靠的多目标优化方法.

A new hybrid multi-objective genetic algorithm for optimal design of groundwater remediation systems

This study presents an algorithm that uses a novel iterative local search (Hill Climber with Step, HCS) in the nondominated sorting genetic algorithm II (NSGAII) as a hybrid multi-objective algorithm for improving convergence to the true Pareto front. We present some numerical results on two benchmark problems (CONV1, ZDT6). For the Pareto optimal fronts achieved by NSGAII-HCS with CONV1 and ZDT6, we compute the Euclidean distances of the Pareto fronts from the true Pareto optimal fronts. Comparing with NSGAII, NSGAII-HCS is able to decrease the average distance from 5.49 to 1.74 for CONV1 and from 0.16 to 0 for ZDT6, indicating that the proposed NSGAII-HCS is able to find much better spread of solutions and better convergence to the true Pareto optimal front. Finally, the proposed NSGAII-HCS is coupled with the commonly used flow and transport code, MODFLOW and MT3DMS, and applied to a synthetic pump-and-treat (PAT) groundwater remediation system. Comparing with the existing nondominated sorting genetic algorithm II (NSGAII), the proposed NSGAII-HCS can find Pareto optimal solutions with lower variability and higher reliability and is a promising tool for optimizing the multi-objective design of groundwater remediation systems.