混合变分模态长短期神经网络的水库表面位移形变预测
DOI:
作者:
作者单位:

1.东华理工大学 测绘与空间信息工程学院;2.河北省水利水电勘测设计研究院集团有限公司 河北 石家庄;3.江西理工大学 土木与测绘工程学院

作者简介:

通讯作者:

中图分类号:

P228

基金项目:

国家自然科学基金项目(面上项目,重点项目,重大项目)(项目编号:42374040,42061077,42104023,42364002);江西省主要学科学术和技术带头人培养计划(20225BCJ23014)


Mix variational mode decomposition long short-term memory for predicting of reservoir surface displacement and deformation
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为提高水库位移形变预测精度,通过改变变分模态分解(Variational Mode Decomposition,VMD)的分解方式,融合VMD与长短期神经网络 (Long Short-Term Memory,LSTM)对非线性非平稳的水库位移形变进行预测,提出了一种混合变分模态长短期神经网络(Mix Variational Mode Decomposition Long Short-Term Memory,MVMDLSTM)模型预测方法;对不同单一预测模型与组合模型采用多源数据集验证新方法的可靠性。实验结果表明:MVMDLSTM模型能有效减弱单一预测模型与经验模态分解组合模型估计的偏差,MVMDLSTM模型预测精度更优,为稳定、监测水库慢滑移和蠕动等微小变形预测预警提供有效的数据决策。

    Abstract:

    In order to improve the prediction accuracy of the displacement and deformation of YANFGHE reservoir, the displacement and deformation of non-linear and non-stationary reservoir was predicted by changing the decomposition method of VMD(variational mode decomposition) and integrating VMD and LSTM (long short-term memory). A MVMDLSTM (mixed variational mode decomposition long short-term memory) model prediction method was proposed. The reliability of the new method was analyzed by comparing different single prediction models with the combined model and different data sets. The experimental results show that the MVMDLSTM model can effectively attenuate the bias of the single prediction model and the empirical mode decomposition combination model estimation, and the prediction accuracy of the MVMDLSTM model is better, which provides an effective data decision-making for the stabilization and monitoring of the prediction and warning of the reservoir's slow sliding and creeping and other small deformations.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2023-04-10
  • 最后修改日期:2024-06-21
  • 录用日期:2023-11-29
  • 在线发布日期: 2025-04-03
  • 出版日期:
文章二维码