水文地质参数反演的Hooke-Jeeves粒子群混合算法

水文地质参数寻优结果的好坏会直接影响到地下水数值模拟的精度,而参数寻优结果很大程度上取决于寻优方法的选择。粒子群算法是一种基于群智能的随机全局寻优方法,算法的缺陷是后期搜索效率低劣。基于随机寻优算法的混合策略,引入有效的约束处理手段和粒子群算法惯性因子的动态非线性调整技术,有机融合粒子群算法与Hooke-Jeeves方法,提出一种适用于水文地质参数反演的HJPSO混合算法。应用研究表明,HJPSO混合算法在参数反演计算中求解精度高、收敛速度快、寻优性能强,是一种值得推广的水文地质参数识别方法。

Inverse analysis of hydrogeological parameters using hybrid Hooke-Jeeves and particle swarm optimization method

The performance of groundwater modeling relies largely on the accurate estmiation of hydrogeological parameters,which are often influenced by the choice of optimization methods used.The particle swarm optimization(PSO)is a swarm in telligence technique for global optimization.However,the search efficiency of PSO decreases as the iteration process approaching to the end.It is thus desirable to search for the enhanced version of PSO.In this study,a hybrid global optimization algorithm that uses the Hooke-Jeeves(HJ)method for the bcal optimization and PSO for the global optimization is proposed to address the inverse problems in ground water modeling.Hydrogeological parameters can be determined using the new HJPSO algorithm.The result of a case study shows that HJPSO is characterized as an algorithm with accuracy,fast convergence and high robustness in the estmiation of hydrogeological parameters,and applicable to hydrogeologic parameters identification.