哈里斯鹰算法在广义非线性马斯京根参数优化中的应用——以洛河为例

Application of Harris Hawks Optimization Algorithm in Optimization of Generalized Nonlinear Muskingum Parameters--A Case Study of the Luohe River

马斯京根模型在河道洪水演算中发挥着重要作用,其演算精度在于参数的优选。针对目前马斯京根参数率定中存在的求解复杂、精度不高等问题,提出利用哈里斯鹰算法对其参数进行优化,这种方法具有广泛的全局搜索能力,且需要调节的参数较少。以黄河支流洛河为研究对象,利用广义非线性马斯京根模型对宜阳—白马寺段的河道进行洪水演算,且分别用哈里斯鹰算法、粒子群算法和蚁群算法对其参数进行优化。结果表明,基于哈里斯鹰算法的广义非线性马斯京根模型在洛河宜阳—白马寺段的演算精度较高,其Min.SSD为1237,洪峰误差DPO仅为5,均优于粒子群算法和蚁群算法优化后的结果,其成果适合应用于洛河宜阳—白马寺段的洪水预报工作。

The Muskingum model plays an important role in river flood simulation,and its simulation accuracy relies on the optimal selection of parameters.To address the current challenges in parameter calibration for the Muskingum model,such as complex solution processes and low accuracy,the use of the Harris Hawks optimization(HHO)algorithm was proposed to optimize its parameters.HHO algorithm has a wide range of global search capabilities,with fewer parameters to be adjusted.Taking Luohe River,a tributary of the Yellow River,as the research object,the generalized nonlinear Muskingum model was used to simulate the flood in the Yiyang-Baimasi section of the river.The parameters were optimized by employing the HHO algorithm,particle swarm optimization(PSO)algorithm,and ant colony optimization(ACO)algorithm,respectively.The results show that the generalized nonlinear Muskingum model based on the HHO algorithm achieved high simulation accuracy in the Yiyang-Baimasi section of the Luohe River,with a Min.SSD of 1237 and the flood peak error(DPO)of only 5,outperforming those obtained through optimization using PSO algorithm and ACO algorithm.The results are suitable for application in flood forecasting in the Yiyang-Baimasi section of the Luohe River.