利用单纯形-粒子群混合算法确定越流含水层参数  

Estimation of leakage aquifer parameters with simplex-particle swarm optimization algorithm

作　　者：袁帆[1] 刘元会[1] 郭建青[2] 王璐璐 

机构地区：[1]长安大学理学院,西安710064 [2]长安大学环境科学与工程学院,西安710051 [3]中国水利水电第三工程局有限公司,河北承德067411

出　　处：《南水北调与水利科技》2015年第4期729-732,755,共5页South-to-North Water Transfers and Water Science & Technology

基　　金：国家自然科学基金(11171043)

摘　　要：针对粒子群优化算法后期存在的收敛速度慢、早熟、易陷入局部极小等问题,将局部搜索能力强的单纯形算法和粒子群算法结合,构造单纯形-粒子群混合算法。以第一类越流系统情况下的非稳定井流问题的解析解为基础,将单纯形-粒子群混合算法应用于分析抽水试验数据,计算含水层参数的问题。数值实验结果表明:单纯形-粒子群混合算法能有效地应用于分析抽水试验数据,确定含水层参数,且具有局部搜索能力强、运算速度快和计算精度高等优点。The particle swarm optimization algorithm has slow convergence, prematurity, and local minimum problems; therefore, the sim plex method algorithm w as combined with the part icle sw arm algorithm to develop a hybrid algorithm called simplex-particle sw arm algorit hm given that the simplex met hod algorithm has strong local search ability. Based on the analytical solu-tions to unst eady w ell flow problems in t he first type leakage system, the simplex-part icle sw arm algorithm w as employed to an-alyze the pumping test data in order to determine aquifer parameters. Numerical results show ed that the simplex-particle sw arm algorithm can analyze the pumping test dat a effectively for the estimation of aquifer parameters, and this met hod has strong local search ability, fast calculation abilit y, rapid convergence rate, and high accuracy.

关 键 词：越流系统 含水层参数 单纯形算法 粒子群算法 混合算法 

分 类 号：TV211.1[水利工程—水文学及水资源]