基于水文水动力模型和机器学习模型耦合的河道水位预报方法  

River Water Level Prediction Method Using Hydrodynamic Model and Machine Learning Model

作　　者：胡义明[1] 陈钰 周瑛 李彬权[1] 陈丞 许栋 梁忠民[1] HU Yi-ming;CHEN Yu;ZHOU Ying;LI Bin-quan;CHEN Cheng;XU Dong;LIANG Zhong-min(College of Hydrology and Water Resources,Hohai University,Nanjing 210098,China;Institute of Water Science and Technology,Hohai University,Nanjing 210098,China;College of Water Conservancy and Hydropower Engineering,Hohai University,Nanjing 210098,China;Nansha District Water Bureau of Guangzhou,Guangzhou 511455,China)

机构地区：[1]河海大学水文水资源学院,江苏南京210098 [2]河海大学水科学研究院,江苏南京210098 [3]河海大学水利水电学院,江苏南京210098 [4]广州市南沙区水务局,广东广州511455

出　　处：《水电能源科学》2024年第10期29-32,共4页Water Resources and Power

基　　金：国家自然科学基金项目(42371045,52379007);广州市南沙区水务局科技项目(2022-263)。

摘　　要：为探讨利用水文水动力模型和机器学习模型来提高河道水位预报精度的可行性,首先利用水文水动力模型进行河道水位预报,采用支持向量机模型对水文水动力模型的预报结果进行校正,进而构建了一种耦合水文水动力模型和机器学习模型的河道水位预报模型。在广州市南沙区蕉西水闸的应用结果表明,构建的耦合模型的预报效果优于单一的水文水动力模型,明显地提高了不同预见期下的水位预报精度;尽管随着预见期的增加,耦合模型的预报精度有一定的衰减趋势,但整体上仍优于水文水动力模型提供的水位预报结果。This study explores the feasibility of utilizing hydrodynamic model and machine learning models to improve the accuracy of river water level prediction.A hydrodynamic model is first used to predict river water levels,and the support vector machine model is used to correct the prediction results of the hydrodynamic model.Then,a river level prediction model coupled with hydrodynamics model and machine learning model is constructed.The application results of Jiaoxi Sluice in Nansha District show that the coupled model has better prediction performance than a single hydrodynamic model,significantly improving the accuracy of water level prediction under different forecast periods.As the forecast periods increase,the prediction effect of the coupled model shows a decreasing trend,but overall,it is still better than the water level prediction process of the hydrodynamic model.

关 键 词：水位预报 水文水动力模型 机器学习模型 耦合模型 

分 类 号：TV124[水利工程—水文学及水资源]