基于BP神经网络的复合单沙模型在西江泥沙监测的应用初探——以梧州水文站为例  

Application of Composite Single Sand Model for Sediment Monitoring Based on BP Neural Network in Xijiang River:A Case Study of Wuzhou Hydrometric Station

作　　者：朱颖洁 ZHU Yingjie(Wuzhou Hydrological Center,Wuzhou 543002,China)

机构地区：[1]梧州水文中心,广西梧州543002

出　　处：《水文》2024年第3期36-40,共5页Journal of China Hydrology

基　　金：国家自然科学基金项目(41461005);广西自然科学基金重点项目(2022GXNSFDA080009);广西水利厅科技项目(201618)。

摘　　要：梧州水文站河段水流形态、泥沙输移规律复杂,适用的仪器少,实现泥沙自动在线监测是一大难题。通过多元回归和BP神经网络建立单沙模型,研究复合单沙模型在西江黄金水道梧州水文站河段泥沙监测的适用性。结果表明,基于BP神经网络的复合单沙模型推算出的单沙满足规范要求,应用效果较好,可在梧州水文站河段泥沙监测中应用;通过使用浊度自动监测系统进行浊度观测,并利用BP神经网络方法建立单沙与浊度和水位之间的关系来推算单沙,实现含沙量的自动在线监测功能,具有一定应用推广意义。The automatic online monitoring of sediment in Wuzhou River is a difficult problem due to the complex flow patterns and sediment transport laws,and the lack of suitable equipments.A composite single sand model was established by multiple re-gression and BP neural network for the sediment monitoring at Wuzhou hydrometric station in Xijiang River.The results show that the single sand calculated by the neural network model is compliant with specifications,which could be applied to sediment monitoring of Wuzhou hydrometric station.Using turbidity monitoring equipment and neural network to establish the relationship among single sand,turbidity and water level is applicable to the online monitoring of sediment.

关 键 词：BP神经网络 浊度 单沙 西江 梧州 

分 类 号：P33[天文地球—水文科学] TV121[水利工程—水文学及水资源]