进化粒子群优化算法及其在各向异性含水层参数估计中的应用

Evolutionary particle swarm optimization algorithm and its application to estimation of anisotropic aquifer parameters from pumping test data

在粒子群算法的基础上,引入进化思想和组群思想提出一种新的智能优化算法——进化粒子群算法(EPSO)。基于抽水试验数据,将EPSO算法应用到各向异性含水层参数估计中,对算法性能进行研究并与其他算法进行了对比,发现标准粒子群算法及其一般改进算法已不能有效求解各向异性含水层参数,而EPSO算法进行多次计算后,1)结果可靠;2)目标函数值及待估参数稳定;3)对初始范围的鲁棒性好。结果表明EPSO算法对各向异性含水层参数估计问题具有可靠性、收敛性和鲁棒性,可望应用到更广泛的参数识别问题中。

By introducing evolution thought and group thought into Particle Swarm Optimization （ PSO） algorithm, this paper proposed a new algorithm-Evolutionary Particle Swarm Algorithm （ EPSO） . Based on the pumping test data, the paper used EPSO to estimate the anisotropic aquifer parameters and made a comparative study with other calculation methods. The experimental results demonstrate that sandard PSO algorithm and its general improvements can＇t evaluate the anisotropic aquifer parameters but EPSO for multiple calculation show that： 1 ） results are reliable; 2 ） the objective function value and estimated parameters remain stable; 3） parameters show good robustness for original scope. The results show that the EPSO for estimating the anisotropy of aquifer parameters is reliable, convergent and robust.