改进的蚁群算法在溶质运移参数识别问题中的应用

非饱和土壤水分和溶质运移参数(扩散率、导水率和水动力弥散系数)取值范围较大,往往跨越几个数量级。采用传统离散化蚁群算法求解此类问题,所需节点较多,这会造成算法收敛时间较长。该文在传统蚁群算法基础上,对蚂蚁搜索路径进行改进,改进后的蚁群算法寻优路径由参数精度位数(整数位和小数位)、参数个数以及0–9十个数字构成,并将路径解码公式修改为具有判别参数正负功能的解码公式。采用改进的连续蚁群算法对非饱和溶质运移参数识别优化模型进行求解。数值模拟表明相同迭代次数下改进的蚁群算法比传统蚁群算法耗时少,算法计算时间与迭代次数满足线性关系,含水率和溶质浓度实测值与计算值吻合较好、相关性较高。

Application of improved ant colony algorithm in parameter identification problem of solute transport

The range of parameters (diffusion rate,hydraulic conductivity,and hydrodynamic dispersion coefficient) for water and solute transport usually is wider in unsaturated soil,and it could across several orders of magnitude.If using traditional ant colony algorithm (TACA) to solve such problems with more nodes,it will cause the convergence time longer than the other.Therefore,in this paper an improved ant colony optimization algorithm (IACA) is proposed on the basis of traditional ant colony algorithm.The seeking optimization paths of improved ant colony algorithm (IACA) are composed of parameter digits of precision (include integer and decimal),numbers of parameter and the 10 number digits from 0 to 9.Meanwhile,the parse path formula is modified to have a function to distinguish that the parameter is a positive or a negative.The improved algorithm is used to solve the optimization model for solute transport parameters identification in unsaturated soil.Numerical simulation shows that the improved ant colony consumes less time than the traditional ant colony algorithm with the same iteration.The algorithm computing time and the iterations satisfies the linear relationship.The measured value is similar to calculated value and the correlation is higher for the water content and solute concentrations.