利用差分进化算法反求含水层参数

Estimation of hydrogeology parameters using differential evolution algorithm

为更好地估算含水层参数,引入了差分进化算法。该算法是一种智能优化算法,其特点是借助现有近似解群体间的距离及方向指导未来的搜索行为,兼具有较好的全局寻优能力和较快的局部搜索能力。用此算法求解了均质各向同性及均质各向异性条件下Theis公式中的含水层参数,所用降深数据是在设定参数后利用解析解计算得到的。计算结果表明,该算法较传统方法计算精度高,不受人为因素干扰。

Differential evolution （DE） is firstly introduced into estimating hydrogeology parameters in this paper. The DE is one of the intelligent optimization methods, it extracts the differential information from the current population of solutions to guide further search. DE makes good balance between global optimization and local searching ability by introducing a method like Singleton method. The general principle of the DE is introduced, so as the main computing procedures. Some hydrogeology parameters in Theis formula are estimated with DE while the data is calculated according to analytical solutions with supposing parameters. The result shows that DE performs well and is worth to applying in estimating hydrogeology parameters and without artificial influence.