摘要
为完善地下水优化管理模型,减小因模型参数不确定性产生的误差,需对优化管理模型进行参数求解性能及敏感性评价。本研究以大连市周水子研究区为例,利用程序MOSWTGA建立海水入侵区地下水优化管理模型,通过引入参数求解性能评价指标及参数敏感性评价方法,对求解得到的近似Pareto解的收敛性及多样性进行评价,并对优化管理模型中二进制位数、交叉概率、种群数及变异概率等参数进行敏感性评价。研究结果表明:MIROC气候模式下,二进制位数为4、交叉概率p_(cross)为0.6、种群数n_(popsiz)为50、变异概率p_(mutate)为(1/50)×3时,收敛性指标最小,优化得到的Pareto解收敛性最好,越接近真实Pareto解;二进制位数为4、交叉概率p_(cross)为0.6、种群数n_(popsiz)为150、变异概率p_(mutate)为(1/150)时,多样性指标最小,其优化得到的近似Pareto解沿着权衡曲线上的分布最均匀。通过参数求解性能评价及敏感度分析能够确定出模拟优化模型中的不确定性因素,从而避免模型参数调整的盲目性,提高地下水优化管理的可靠性。
In order to improve the precision of groundwater simulation-optimization modeling and reduce the errors caused by the uncertainties of simulation-optimization model parameters,performance assessment of solutions and sensitivity analysis of parameters are needed for the numerical simulation-optimization modeling.In this paper,MOSWTGA is applied to optimal control of seawater intrusion in Zhoushuizi district,Dalian.Computational performance is introduced to measure the convergence and the diversity of the near Pareto-optimal solutions,and sensitivity analysis of parameters is performed including the length of the binary string,crossover probability,population size and mutation probability.The results show that in MIROC climate mode when the length of the binary string is 4,the crossover probability pcross is 0.6,the population size npopsiz is 50 and the mutation probability pmutate is(1/50)×3,the value of convergence metrics is the smallest with the best convergence towards the Pareto-optimal front;when the length of the binary string is 4,the crossover probability pcross is 0.6,the population size npopsiz is 150 and the mutation probability pmutate is(1/150),the diversity index is the smallest,while the Pareto-optimal solution along the tradeoff curve is distributed uniformly.The uncertain factors in the simulation-optimization model are determined by computational performance assessment and sensitivity analysis of parameters,and the blindness of parameters adjustment is avoided.The reliability of optimal groundwater management is improved.
作者
赵洁
林锦
吴剑锋
吴吉春
Zhao Jie;Lin Jin;Wu Jianfeng;Wu Jichun(School of Water Conservancy,North China University of Water Resources and Electric Power,Zhengzhou 450046,China;Key Laboratory of Surficial Geochemistry,Ministry of Education,Department of Hydrosciences,School of Earth Sciences and Engineering,Nanjing University,Nanjing 210093,China;Nanjing Hydraulic Research Institute,Nanjing 210029,China)
出处
《工程勘察》
2020年第3期32-37,78,共7页
Geotechnical Investigation & Surveying
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(51709106,40902069)
河南省教育厅高校重点科研项目(18A170011).
关键词
MOSWTGA
海水入侵
模拟优化
参数求解性能
敏感性评价
MOSWTGA
seawater intrusion
simulation-optimization
computational performance
sensitivity evaluation
