A Prediction Model of Significant Wave Height in the South China Sea Based on Attention Mechanism

被引:12
|
作者
Hao, Peng [1 ]
Li, Shuang [1 ]
Yu, Chengcheng [1 ]
Wu, Gengkun [2 ]
机构
[1] Zhejiang Univ, Inst Phys Oceanog & Remote Sensing, Ocean Coll, Zhoushan, Peoples R China
[2] Shandong Univ Sci & Technol, Coll Comp Sci & Engn, Qingdao, Peoples R China
基金
中国国家自然科学基金;
关键词
CBA-Net; significant wave height (SWH); deep learning; South China Sea; attention mechanism; STATISTICAL-MODELS; TERM PREDICTION; NEURAL-NETWORKS; SIMULATION; FORECASTS;
D O I
10.3389/fmars.2022.895212
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Significant wave height (SWH) prediction plays an important role in marine engineering fields such as fishery, exploration, power generation, and ocean transportation. Traditional SWH prediction methods based on numerical models cannot achieve high accuracy. In addition, the current SWH prediction methods are largely limited to single-point SWH prediction, without considering regional SWH prediction. In order to explore a new SWH prediction method, this paper proposes a deep neural network model for regional SWH prediction based on the attention mechanism, namely CBA-Net. In this study, the wind and wave height of the ERA5 data set in the South China Sea from 2011 to 2018 were used as input features to train the model to evaluate the SWH prediction performance at 1 h, 12 h, and 24 h. The results show that the single use of a convolutional neural network cannot accurately predict SWH. After adding the Bi-LSTM layer and attention mechanism, the prediction of SWH is greatly improved. In the 1 h SWH prediction using CBA-Net, SARMSE, SAMAPE, SACC are 0.299, 0.136, 0.971 respectively. Compared with the CNN + Bi-LSTM method that does not use the attention mechanism, SARMSE and SAMAPE are reduced by 43.4% and 48.7%, respectively, while SACC is increased by 5%. In the 12 h SWH prediction, SARMSE, SAMAPE, and SACC of CBA-Net are 0.379, 0.177, 0.954 respectively. In the 24 h SWH prediction, SARMSE, SAMAPE, and SACC of CBA-Net are 0.500, 0.236, 0.912 respectively. Although with the increase of prediction time, the performance is slightly lower than that of 12 h, the prediction error is still maintained at a small level, which is still better than other methods.
引用
收藏
页数:12
相关论文
共 50 条
  • [41] A hybrid model based on chaos particle swarm optimization for significant wave height prediction
    Yang, Can
    Kong, Qingchen
    Su, Zuohang
    Chen, Hailong
    Johanning, Lars
    OCEAN MODELLING, 2025, 195
  • [42] Significant wave height prediction based on variational mode decomposition and dual network model
    Chen, Jiaxin
    Li, Shibao
    Zhu, Jinze
    Liu, Menglong
    Li, Ran
    Cui, Xuerong
    Li, Lianghai
    OCEAN ENGINEERING, 2025, 323
  • [43] Significant Wave Height Prediction Based on MSFD Neural Network
    Wang, Huan
    Fu, Dongyang
    Liao, Shan
    Wang, Guancheng
    Xiao, Xiuchun
    2019 TENTH INTERNATIONAL CONFERENCE ON INTELLIGENT CONTROL AND INFORMATION PROCESSING (ICICIP), 2019, : 39 - 43
  • [44] Prediction of significant wave height based on feature decomposition and enhancement
    Wu, Lin
    An, Yi
    Qin, Pan
    Hu, Huo-Sheng
    EXPERT SYSTEMS WITH APPLICATIONS, 2025, 277
  • [45] Prediction of significant wave height based on EEMD and deep learning
    Song, Tao
    Wang, Jiarong
    Huo, Jidong
    Wei, Wei
    Han, Runsheng
    Xu, Danya
    Meng, Fan
    FRONTIERS IN MARINE SCIENCE, 2023, 10
  • [46] Prediction of significant wave height in hurricane area of the Atlantic Ocean using the Bi-LSTM with attention model
    Luo, Qin-Rui
    Xu, Hang
    Bai, Long-Hu
    OCEAN ENGINEERING, 2022, 266
  • [47] Comparison of ECMWF Significant Wave Height Forecasts in the China Sea with Buoy Data
    Wang, Juanjuan
    Li, Benxia
    Gao, Zhiyi
    Wang, Jiuke
    WEATHER AND FORECASTING, 2019, 34 (06) : 1693 - 1704
  • [48] An evaluation of input/dissipation terms in WAVEWATCH Ⅲ using in situ and satellite significant wave height data in the South China Sea
    WANG Jichao
    ZHANG Jie
    YANG Jungang
    BAO Wendi
    WU Guoli
    REN Qifeng
    Acta Oceanologica Sinica, 2017, 36 (03) : 20 - 25
  • [49] Short-term prediction of significant wave height based on AR-SVR model
    Zhang Z.
    Li X.
    Yang S.
    Taiyangneng Xuebao/Acta Energiae Solaris Sinica, 2021, 42 (07): : 15 - 20
  • [50] Wavelet-Based ResNet: A Deep-Learning Model for Prediction of Significant Wave Height
    Yu, Xiangjun
    Liu, Yarong
    Sun, Zhiming
    Qin, Pan
    IEEE ACCESS, 2022, 10 : 110026 - 110033