用改进的遗传算法反演地下水数值模型参数  

Parameter identification of groundwater flow numerical modeling by means of improved genetic algorithm

作　　者：姚磊华[1] 李竞生[2] 李钊 

机构地区：[1]中国地质大学工程技术学院,北京100083 [2]煤炭科学研究总院西安分院,陕西西安710054 [3]河南省长葛市水利局,河南长葛461500

出　　处：《水利学报》2003年第12期40-46,共7页Journal of Hydraulic Engineering

基　　金：国家自然科学基金资助项目(49972085)

摘　　要：在简单的遗传算法的基础上,结合优体克隆操作、子体优生操作和多代调环操作,提出了一种综合改进的遗传算法,并以非均质各向同性承压二维非稳定流动为算例,以有限元法讨论了用遗传算法反演地下水数值模型参数的过程。计算结果表明,综合改进的遗传算法进行参数识别时,具有收敛速度快、解的精度高和避免出现早熟等优点,可用于地下水渗流和水资源评价计算中。On the basis of simple genetic algorithm,an improved genetic algorithm applying the best chromosome clone operation,younger generation chromosomes prepotency operation and multi-generations adjusting environment operation is proposed.The 2-D unsteady flow in inhomogeneous and isotropic confined aquifer is used as an example to illustrate the procedure of the algorithm to be applied to the inverse problem of hydrology parameter identification by means of FEM.The calculation result indicates that the algorithm possesses the advantages of expediting constringency.avoiding earliness and improving precision,when it is used in parameter identification of groundwater flow modeling.

关 键 词：遗传算法 改进 反演 地下水水流模型 参数 

分 类 号：P641.2[天文地球—地质矿产勘探]