人民珠江2025，Vol.46Issue(11)：1-8,8.DOI:10.3969/j.issn.1001-9235.2025.11.001
基于机器学习的珠江河口咸潮上溯实时预报研究
Real-time Forecasting of Tidal Saltwater Intrusion in the Pearl River Estuary Based on Machine Learning
摘要
Abstract
Intensified global climate change and human activities lead to increasingly severe tidal saltwater intrusion in the Pearl River Estuary,and the water supply security of coastal cities is under significant threat.This study employed long short-term memory(LSTM)and gated recurrent unit(GRU)networks to forecast and validate hourly salinity data at the Pinggang Station in the Modaomen Waterway from 2019 to 2023.The results indicate:① Both the LSTM and GRU models demonstrate strong performance in forecasting tidal saltwater intrusion.Compared to the LSTM model,the GRU model exhibits higher forecasting accuracy,smaller prediction errors,and faster computational speed,with its performance advantages being more pronounced in short-term forecasts.② The GRU model achieves a forecasting accuracy of above 0.8 for future 1～24 hours,with the accuracy for future 1～6 hours generally reaching 0.9.关键词
咸潮上溯预报/LSTM模型/GRU模型/深度学习/珠江河口区Key words
tidal saltwater intrusion forecasting/LSTM model/GRU model/deep learning/Pearl River Estuary分类
建筑与水利引用本文复制引用
易晶晶,刘大为,祝雨珂,刘丙军..基于机器学习的珠江河口咸潮上溯实时预报研究[J].人民珠江,2025,46(11):1-8,8.基金项目
广东省水利科技创新项目(2023-01) （2023-01）
南方海洋科学与工程广东省实验室项目(SML2023SP214) （SML2023SP214）
