CNN-BiLSTM-Attention Model in Forecasting Wave Height over South-East China Seas

作     者：Lina Wang Xilin Deng Peng Ge Changming Dong Brandon J.Bethel Leqing Yang Jinyue Xia 

作者机构：School of Artificial Intelligence(School of Future Technology)Nanjing University of Information Science and Technology Nanjing210044China Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)Zhuhai519080China School of Marine SciencesNanjing University of Information Science and TechnologyNanjing210044China International Business Machines Corporation(IBM)New York10504USA 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2022年第73卷第10期

页      面：2151-2168页

核心收录：

学科分类：07[理学] 0707[理学-海洋科学] 0706[理学-大气科学] 0714[理学-统计学（可授理学、经济学学位）] 0701[理学-数学] 

基　　金：This study is supported by the project supported by the Southern Marine Science and Engineering Guangdong Laboratory(Zhuhai)(SML2020SP007) the National Natural Science Foundation of China(Nos.61772280 and 62072249) 

主　　题：Conv2D CNN-BiLSTM-Attention wave forecasting significant wave height typhoon 

摘      要：Though numerical wave models have been applied widely to significant wave height prediction,they consume massive computing memory and their accuracy needs to be further *** this paper,a two-dimensional(2D)significant wave height(SWH)prediction model is established for the South and East China *** proposed model is trained by Wave Watch III(WW3)reanalysis data based on a convolutional neural network,the bidirectional long short-term memory and the attention mechanism(CNNBiLSTM-Attention).It adopts the convolutional neural network to extract spatial features of original wave height to reduce the redundant information input into the BiLSTM ***,the BiLSTM model is applied to fully extract the features of the associated information of time series ***,the attention mechanism is used to assign probability weight to the output information of the BiLSTM layer units,and finally,a training model is *** to 24-h prediction experiments are conducted under normal and extreme conditions,*** the normal wave condition,for 3-,6-,12-and 24-h forecasting,the mean values of the correlation coefficients on the test set are 0.996,0.991,0.980,and 0.945,*** corresponding mean values of the root mean square errors are measured at 0.063 m,0.105 m,0.172 m,and 0.281 m,*** the typhoon-forced extreme condition,the model based on CNN-BiLSTM-Attention is trained by typhooninduced SWH extracted from the WW3 reanalysis *** 3-,6-,12-and 24-h forecasting,the mean values of correlation coefficients on the test set are respectively 0.993,0.983,0.958,and 0.921,and the averaged root mean square errors are 0.159 m,0.257 m,0.437 m,and 0.555 m,*** model performs better than that trained by all the WW3 reanalysis *** result suggests that the proposed algorithm can be applied to the 2D wave forecast with higher accuracy and efficiency.