|国家科技期刊平台
| 注册
首页|期刊导航|大气和海洋科学快报（英文版）|Application of the improved dung beetle optimizer,muti-head attention and hybrid deep learning algorithms to groundwater depth prediction in the Ningxia area,China

Application of the improved dung beetle optimizer,muti-head attention and hybrid deep learning algorithms to groundwater depth prediction in the Ningxia area,China

Jiarui Cai Bo Sun Huijun Wang Yi Zheng Siyu Zhou Huixin Li Yanyan Huang Peishu Zong

大气和海洋科学快报（英文版）2025，Vol.18Issue(1)：18-23,6.
✕
大气和海洋科学快报（英文版）2025，Vol.18Issue(1)：18-23,6.DOI:10.1016/j.aosl.2024.100497

Application of the improved dung beetle optimizer,muti-head attention and hybrid deep learning algorithms to groundwater depth prediction in the Ningxia area,China

Application of the improved dung beetle optimizer,muti-head attention and hybrid deep learning algorithms to groundwater depth prediction in the Ningxia area,China

Jiarui Cai 1Bo Sun 2Huijun Wang 2Yi Zheng 1Siyu Zhou 1Huixin Li 1Yanyan Huang 1Peishu Zong3

作者信息

  • 1. Collaborative Innovation Center on forecast and Evaluation of Meteorological Disasters/Key Laboratory of Meteorological Disaster,Ministry of Education/Joint International Research Laboratory of Climate and Environment Change,Nanjing University of Information Science and Technology,Nanjing,China
  • 2. Collaborative Innovation Center on forecast and Evaluation of Meteorological Disasters/Key Laboratory of Meteorological Disaster,Ministry of Education/Joint International Research Laboratory of Climate and Environment Change,Nanjing University of Information Science and Technology,Nanjing,China||Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai),Zhuhai,China
  • 3. China Meteorological Administration,Key Laboratory of Transportation Meteorology,Nanjing,China||Jiangsu Meteorological Observatory,Nanjing,China
  • 折叠

摘要

Abstract

本研究将两个新模型应用于位于中国西北干旱半干旱区的宁夏地区地下水深度预测.这两个模型将改进的蜣螂优化(DBO)算法与两个深度学习模型相结合,即多头注意力-卷积神经网络-长短期记忆网络和多头注意力-回旋神经网络-门控递归单元.带有DBO的模型预测结果表现出更大的相关系数(R),残差预测偏差(RPD)和较低的均方根误差(RMSE),预测结果更好.此外,与DBO模型相比,改进后的DBO模型的R和RPD增加了 1.5％以上,RMSE降低了 1.8％以上,表明预测结果更好.与传统的统计模型多元线性回归模型相比,深度学习模型具有更好的预测性能.

关键词

地下水深度/多头注意力机制/改进的蜣螂优化算法/CNN-LSTM/CNN-GRU/宁夏

Key words

Groundwater depth/Multi-head attention/Improved dung beetle optimizer/CNN-LSTM/CNN-GRU/Ningxia

引用本文复制引用

Jiarui Cai,Bo Sun,Huijun Wang,Yi Zheng,Siyu Zhou,Huixin Li,Yanyan Huang,Peishu Zong..Application of the improved dung beetle optimizer,muti-head attention and hybrid deep learning algorithms to groundwater depth prediction in the Ningxia area,China[J].大气和海洋科学快报（英文版）,2025,18(1):18-23,6.

基金项目

This work was supported by the National Natural Science Foundation of China[grant numbers 42088101 and 42375048]. （）

大气和海洋科学快报（英文版）

1674-2834

访问量3
|
下载量0
段落导航相关论文