基于SMSA混合算法的地下水污染源强度反演研究

Recovering the History of Groundwater Contaminant by hybrid Simplex-Simulated Annealing Method

摘要:

根据浓度监测数据反求地下水污染源信息是一类典型的地下水逆问题。本文基于单纯形模拟退火混合算法(SMSA混合算法)，结合污染物迁移问题的解析解反演地下水污染源的强度变化历时曲线。SMSA混合算法结合了单纯形法的确定性搜索和模拟退火算法的全局概率搜索机制，是一种高效的混合优化算法；同时本文采用Yeh提出的解析解，它具有可靠性强、易于编程实现和扩展性强的特点。计算结果显示一维、二维、三维情形下点污染源的反演浓度均较好地再现了真实的污染物释放过程，这表明基于SMSA混合算法和Yeh解析解的反演方法是一种有效的地下水污染源重建方法。

Abstract:

To recover groundwater contaminant source’s release history according to the obtained monitoring data belongs to groundwater inverse problem. In this study, a hybrid simplex-simulated annealing method is adopted and incorporated with the analytical solutions of the contaminant transport equation to recover the history of groundwater contaminant. The hybrid SMSA algorithm incorporates the deterministic search pattern of simplex method and the global probabilistic search mechanism of simulated annealing algorithm, and has proved to be a robust hybrid optimization algorithm. Meanwhile, the analytical solution to contaminant transport equation proposed by Yeh is adopted and realized in this study, for the characteristics of this analytical solution are reliability, easy programming and scalability. As shown by the calculation results, the recovered release history obtained by hybrid SMSA method and Yeh’s analytical solution is well reproduce the real process of contaminant release. It is further concluded that hybrid SMSA method and its incorporation with Yeh’s analytical solutions can be widely used in reconstructing the release history of groundwater contaminant.