| 涂俊豪.基于机器学习的长江口含沙量动态预测方法研究[J].水运工程,2024,(12):206-211 |
| 基于机器学习的长江口含沙量动态预测方法研究 |
| Dynamic prediction method of sediment content in Yangtze River estuary based on machine learning |
| |
| DOI： |
| 中文关键词: 长江口 机器学习 含沙量 预测 |
| 英文关键词:Yangtze River estuary machine learning sediment content prediction hydrological parameter |
| 基金项目: |
|
| 摘要点击次数: 186 |
| 全文下载次数: 0 |
| 中文摘要: |
| 长江口作为中国水量最大河流的出海口，其含沙量变化和预测直接影响河口地区的生态环境、航道维护和防洪安全等。开发一个基于机器学习的模型，用于预测长江口区域的含沙量动态。考虑到含沙量受多种水文环境因素的影响，通过收集长江口区域1年的水文数据包括流速、潮位、含沙量等，运用时间序列分析方法，提取关键的特征和模式，选取长短期记忆网络（LSTM）对数据进行训练和测试。分析结果表明，基于LSTM的模型在预测长江口区域含沙量方面表现出了较高的准确性，模型的平均绝对误差为0.146 5，决定系数为0.931 4。 |
| 英文摘要: |
| The Yangtze River estuary is the outlet of the largest river in China,the changes and predicions of sediment content directly affect the ecological environment,waterway maintenance and flood control safety in the estuary area.This study develops a machine learning-based model for predicting sediment content dynamics in the Yangtze River estuary region.Considering that sediment content is affected by a variety of hydrological environmental factors,this study collects hydrological data for a year in the Yangtze River estuary area,including flow velocity,tide level,sediment content,etc.and uses time series analysis methods to extract key features and patterns,and a long short-term memory network(LSTM) is selected,trained and tested.The analysis results show that the LSTM-based model shows high accuracy in predicting sediment content in the Yangtze River estuary region.The mean absolute error of the model is 0.146 5,and the coefficient of determination is 0.931 4. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| 关闭 |
|
|
|
