Dam deformation prediction model based on mixed attention mechanism and deep learning
Author:
Affiliation:

(1.State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, China;2.College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China;3.College of Hydraulic & Environmental Engineering, China Three Gorges University, Yichang 443002, China;4.Gansu Water Resources and Hydropower Survey and Design Research Institute Co., Ltd., Lanzhou 730000, China)

Clc Number:

TV698.1

Fund Project:

  • Article
  • |
  • Figures
  • |
  • Metrics
  • |
  • Reference
  • |
  • Related
  • |
  • Cited by
  • |
  • Materials
  • |
  • Comments
    Abstract:

    In order to deeply mine the dynamic correlation between the front and back information in the time series data,explore the internal influence mechanism of dam deformation, and effectively improve the prediction accuracy of the model, a bidirectional gated recurrent unit (BiGRU) prediction model based on mixed attention mechanism and whale optimization algorithm (WOA) was constructed. WOA was used to optimize the hyperparameters of BiGRU to effectively mine the deep information of deformation data in the time dimension.A mixed attention mechanism that combines factor attention mechanism (FATT) and temporal attention mechanism (TATT) was introduced to calculate the contribution rate of each impact factor, so as to visualize the model and improve the ability to capture the dynamic changes of environmental factors. Taking a high arch dam as an example, the prediction results of this model were compared with those of various commonly used models. The results show that the prediction accuracy of this model is significantly improved, and the calculation of contribution rate is in line with the research results of dam deformation, which verifies the superiority and rationality of the model in dam deformation prediction.

    Reference
    Related
    Cited by
Get Citation

向镇洋,包腾飞,白妍丽,等.基于混合注意力机制和深度学习的大坝变形预测模型[J].水利水电科技进展,2023,43(2):96-101.(XIANG Zhenyang, BAO Tengfei, BAI Yanli, et al. Dam deformation prediction model based on mixed attention mechanism and deep learning[J]. Advances in Science and Technology of Water Resources,2023,43(2):96-101.(in Chinese))

Copy
Share
Article Metrics
  • Abstract:
  • PDF:
  • HTML:
  • Cited by:
History
  • Received:May 19,2022
  • Revised:
  • Adopted:
  • Online: March 10,2023
  • Published: