摘要
为提高秦淮河流域东山站水位预报的精度,基于BP神经网络算法建立经验预报模型,分别根据降雨历时、起涨水位两种模式对水位涨幅进行预报。分析了两种模式预报结果,选出最优的预报模式,并用混合线性回归模型作为预报精度的参考验证。结果显示,BP神经网络模型的预报精度高于混合线性回归模型,而且BP神经网络模型两种预报模式的结果都达到了乙级标准以上,根据起涨水位的预报模式效果更好。
In order to improve the accuracy of water level prediction of Dongshan Station in Qinhuai River Basin,an empirical prediction model was established based on BP neural network algorithm,and the water level rise was predicted from two aspects of rainfall duration and rising water level.The prediction results of two patterns were analyzed,and the optimal prediction pattern was selected.The mixed linear regression model was used as the reference to verify the prediction accuracy.The results show that the prediction accuracy of BP model is higher than that of the mixed linear regression model.Moreover,the results of the two prediction patterns of BP neural network model have reached the class B standard or above,and better results have been achieved according to the prediction model of rising water level.
作者
张轩
张行南
江唯佳
闻余华
聂青
徐荣嵘
ZHANG Xuan;ZHANG Xingnan;JIANG Weijia;WEN Yuhua;NIE Qing;XU Rongrong(College of Hydrology and Water Resources,Hohai Universiy,Nanjing 210098,China;National Cooperative Innovation Center for Water Safety&Hydro-Science,Hohai University,Nanjing 210098,China;National Engineering Research Center of Water Resources Efficient Utilization and Engineering Safety,Hohai University,Nanjing 210098,China;Jiangsu Province Hydrology and Water Resources Investigation Bureau,Nanjing 210029,China;Hydrology and Water Resources Department,Nanjing Hydraulic Research Institute,Nanjing 210029,China)
出处
《水资源保护》
CAS
CSCD
北大核心
2020年第2期41-46,52,共7页
Water Resources Protection
基金
国家重点研发计划项目(2019YFC0409004)
国家自然科学基金(51420105014)。
关键词
面雨量
降雨特性
BP神经网络模型
水位预报
东山站
area rainfall
rainfall characteristics
BP neural network model
water level prediction
Dongshan Station
