黄河下游河道悬沙与床沙粒径智能预测研究

扫码查看

原文链接

万方数据
维普

中文摘要：探究河道悬沙和床沙粒径分布有助于分析河道泥沙整体淤积情况,并反馈指导优化水库运用方式.为系统掌握黄河下游河道悬沙与床沙粒径的分布规律,收集整理黄河下游河道花园口等 6 个断面的水沙系列数据,采用递归特征消除算法进行变量筛选,并基于不同机器学习算法构建黄河下游河道泥沙粒径的预测模型.结果表明:变量筛选算法能够有效提取模型中的主要影响因子,而利用机器学习算法进行不同断面泥沙粒径预测时,在测试集上整体表现较好,各断面优选模型预测悬沙粒径的决定系数R2均在 0.64～0.89 之间,床沙的均在 0.37～0.72 之间.同时,采用 2020 年数据对模型预测效果进一步验证时,悬沙和床沙粒径预测值与实测值R2分别达 0.609 7 和 0.445 6,表明所建模型能够有效实现黄河下游河道泥沙粒径预测.

外文标题：Research on Intelligent Prediction of Suspended Load and Bed Material Load Particle Size in the Lower Yellow River

外文摘要：Exploring the particle size distribution of suspended load and bed material load in river channels helps to analyze the overall sedi-mentation situation of river sediment and provide feedback to guide the optimization of reservoir operation.In order to systematically grasp the distribution pattern of suspended load and bed material load particle size in the lower Yellow River,this paper collected and organized the water and sediment series data from 6 key sections such as Huayuankou in the lower Yellow River.Recursive feature elimination algorithm was used for variable selection,and a prediction model for sediment particle size in the lower Yellow River was built based on different ma-chine learning algorithms.The results show that the variable selection algorithm can effectively extract the main influencing factors in the mod-el,and when using machine learning models to predict the sediment particle size of different sections,the overall performance is good on the test set.The R2 values of suspended load and bed material load particle size prediction under each section optimization model are between 0.64 and 0.89,and 0.37 and 0.72 respectively.At the same time,when using 2020 data to further verify the predictive effect of the model,the predicted and actual values of suspended load and bed material load particle sizes R2 are also up to 0.609 7 and 0.445 6 respectively,which indicates that the built model can effectively achieve the prediction of sediment particle sizes in the lower Yellow River channel.

外文关键词：

particle size of sedimentvariable screeningmachine learningintelligent predictionlower Yellow River channel

作者：

董庆豪、孙龙飞、王远见、赵万杰

展开 >

作者单位：

郑州大学水利与交通学院,河南郑州 450001

黄河水利委员会黄河水利科学研究院,河南郑州 450003

水利部黄河下游河道与河口治理重点实验室,河南郑州 450003

河海大学水利水电学院,江苏南京 210098

展开 >

关键词：

泥沙粒径变量筛选机器学习智能预测黄河下游河道

基金：

国家重点研发计划项目水利部重大科技项目水利部重大科技项目国家自然科学基金资助项目国家自然科学基金资助项目

项目编号：

2021YFC3200400SKR-2022021SKS-2022088U2243601U2243241

出版年：

2024

DOI：

10.3969/j.issn.1000-1379.2324.02.006

人民黄河

水利部黄河水利委员会

人民黄河

CSTPCD北大核心

影响因子：0.494

ISSN：1000-1379

年,卷(期)：2024.46(2)

参考文献量11