人民黄河2026，Vol.48Issue(1)：84-88,95,6.DOI:10.3969/j.issn.1000-1379.2026.01.013
基于机器学习的输水河道沿线地下水位预测分析
Prediction and Analysis of Groundwater Level Around Water Conveyance River Based on Machine Learning
摘要
Abstract
The water diversion project from the Yangtze River to the Huaihe River is a major water resources allocation project.Its Henan section supplies water from the Xifei River in Anhui Province to the upstream.The project uses the Qingshui River to transport water for 48.40 km,and the operation of the project will cause the change of groundwater level.Therefore,the influence of the change of groundwater level a-round the river on the safe and stable operation of the project is studied.In this study,BPNN,SVM and XGBoost algorithms were used to es-tablish machine learning models to predict the groundwater level around the water conveyance channel.The training effects of different models were compared and the optimal prediction model was selected.The influence of different vertical distances from the channel on the prediction of groundwater level was analyzed.The results show that SVM has the best effect in predicting groundwater level,and the vertical distance from the channel has no effect on the prediction results of groundwater level.关键词
引调水工程/地下水位/机器学习/预测分析/引江济淮工程Key words
water diversion project/groundwater level/machine learning/forecasting analysis/Yangtze-to-Huaihe Water Diversion Project分类
建筑与水利引用本文复制引用
ZHANG Yibo,WANG Hui,ZHAO Shougang,LAN Yan..基于机器学习的输水河道沿线地下水位预测分析[J].人民黄河,2026,48(1):84-88,95,6.基金项目
引江济淮工程(河南段)工程科研服务项目(HNYJJH/JS/FWKY-2021001) （河南段）
黄河水利科学研究院科技发展基金资助项目(黄科发202406) （黄科发202406）
水利部堤防安全与病害防治工程技术研究中心开放课题(NQ-ZX-2023-0418) （NQ-ZX-2023-0418）
