Estimating parameters of the variable infiltration capacity model using ant colony optimization.
Estimating parameters of the variable infiltration capacity model using ant colony optimization.
复制标题
DOI：
10.2166/wst.2016.282
复制
发表时间：
2016-08
期刊：
影响因子：
--
通讯作者：
Jiajia Yue;B. Pang;Zongxue Xu
中科院分区：
文献类型：
--
作者：
Jiajia Yue;B. Pang;Zongxue Xu
由于水文模型在解决环境问题中的重要性，参数校准是应用水文模型的一项基本任务。在过去的20年里，一种广泛使用的方法来获得模型参数是进化算法。该方法可以通过模拟演化过程来估计一组未知的模型参数。蚁群优化算法是一种具有较强的组合优化能力的进化算法，适用于水文模型的率定。将基于网格划分策略的蚁群算法应用于黑河上游和西苕溪流域变入渗能力（维克）模型的参数率定。采用混合复杂进化（SCE-UA）算法对蚁群算法进行了验证。结果表明，ACO能够对维克模型进行模型校正，黑河上游和西苕溪流域的Nash-Sutcliffe效率系数分别为0.62和0.81，验证效率系数分别为0.65和0.86，与SCE-UA结果基本一致。尽管取得了令人鼓舞的结果，迄今为止，仍然可以进行进一步的研究，对ACO的参数优化，以扩大其适用性更分布式水文模型。
Because hydrological models are so important for addressing environmental problems, parameter calibration is a fundamental task for applying them. A broadly used method for obtaining model parameters for the past 20 years is the evolutionary algorithm. This approach can estimate a set of unknown model parameters by simulating the evolution process. The ant colony optimization (ACO) algorithm is a type of evolutionary algorithm that has shown a strong ability in tackling combinatorial problems and is suitable for hydrological model calibration. In this study, an ACO based on the grid partitioning strategy was applied to the parameter calibration of the variable infiltration capacity (VIC) model for the Upper Heihe River basin and Xitiaoxi River basin, China. The shuffled complex evolution (SCE-UA) algorithm was used to test the applicability of the ACO. The results show that ACO is capable of model calibration of the VIC model; the Nash-Sutcliffe coefficient of efficiency is 0.62 and 0.81 in calibration and 0.65 and 0.86 in validation for the Upper Heihe River basin and Xitiaoxi River basin respectively, which are similar to the SCE-UA results. Despite the encouraging results obtained thus far, further studies could still be performed on the parameter optimization of an ACO to enlarge its applicability to more distributed hydrological models.