A transformer-based method for correcting significant wave height numerical forecasting errors

被引:1
作者
Kang, Xianbiao [1 ]
Song, Haijun [1 ]
Zhang, Zhanshuo [2 ]
Yin, Xunqiang [3 ]
Gu, Juan [1 ]
机构
[1] Civil Aviat Flight Univ China, Coll Aviat Meteorol, Guanghan, Peoples R China
[2] Natl Supercomp Ctr Jinan, Shandong Comp Sci Ctr, Jinan, Peoples R China
[3] Minist Nat Resources, Inst Oceanog 1, Key Lab Marine & Numer Modeling, Qingdao, Peoples R China
基金
中国国家自然科学基金;
关键词
significant wave height; transformer; 2D-Geoformer; numerical forecasting; error correcting; COASTAL REGIONS; MODEL; ENERGY; ASSIMILATION; PREDICTION; VALIDATION;
D O I
10.3389/fmars.2024.1374902
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Accurate significant wave height (SWH) forecasting is essential for various marine activities. While traditional numerical and mathematical-statistical methods have made progress, there is still room for improvement. This study introduces a novel transformer-based approach called the 2D-Geoformer to enhance SWH forecasting accuracy. The 2D-Geoformer combines the spatial distribution capturing capabilities of SWH numerical models with the ability of mathematical-statistical methods to identify intrinsic relationships among datasets. Using a comprehensive long time series of SWH numerical hindcast datasets as the numerical forecasting database and ERA5 reanalysis SWH datasets as the observational proxies database, with a focus on a 72-hour forecasting window, the 2D-Geoformer is designed. By training the potential connections between SWH numerical forecasting fields and forecasting errors, we can retrieve SWH forecasting errors for each numerical forecasting case. The corrected forecasting results can be obtained by subtracting the retrieved SWH forecasting errors from the original numerical forecasting fields. During long-term validation periods, this method consistently and effectively corrects numerical forecasting errors for almost every case, resulting in a significant reduction in root mean square error compared to the original numerical forecasting fields. Further analysis reveals that this method is particularly effective for numerical forecasting fields with higher errors compared to those with relatively smaller errors. This integrated approach represents a substantial advancement in SWH forecasting, with the potential to improve the accuracy of operational SWH forecasts. The 2D-Geoformer combines the strengths of numerical models and mathematical-statistical methods, enabling better capture of spatial distributions and intrinsic relationships in the data. The method's effectiveness in correcting numerical forecasting errors, particularly for cases with higher errors, highlights its potential for enhancing SWH forecasting accuracy in operational settings.
引用
收藏
页数:12
相关论文
共 51 条
[11]  
Dias F., 2018, P ASME 2018 37 INT C, V11B, DOI [10.1115/OMAE2018-77291, DOI 10.1115/OMAE2018-77291]
[12]   Prediction of significant wave height by artificial neural networks and multiple additive regression trees [J].
Elbisy, Moussa S. ;
Elbisy, Ahmed M. S. .
OCEAN ENGINEERING, 2021, 230
[13]   A novel model to predict significant wave height based on long short-term memory network [J].
Fan, Shuntao ;
Xiao, Nianhao ;
Dong, Sheng .
OCEAN ENGINEERING, 2020, 205
[14]   Significant wave height and energy flux range forecast with machine learning classifiers [J].
Fernandez, J. C. ;
Salcedo-Sanz, S. ;
Gutierrez, P. A. ;
Alexandre, E. ;
Hervas-Martinez, C. .
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, 2015, 43 :44-53
[15]   Significant Wave Height Prediction in the South China Sea Based on the ConvLSTM Algorithm [J].
Han, Lei ;
Ji, Qiyan ;
Jia, Xiaoyan ;
Liu, Yu ;
Han, Guoqing ;
Lin, Xiayan .
JOURNAL OF MARINE SCIENCE AND ENGINEERING, 2022, 10 (11)
[16]  
HASSELMANN S, 1988, J PHYS OCEANOGR, V18, P1775
[17]   The ERA5 global reanalysis [J].
Hersbach, Hans ;
Bell, Bill ;
Berrisford, Paul ;
Hirahara, Shoji ;
Horanyi, Andras ;
Munoz-Sabater, Joaquin ;
Nicolas, Julien ;
Peubey, Carole ;
Radu, Raluca ;
Schepers, Dinand ;
Simmons, Adrian ;
Soci, Cornel ;
Abdalla, Saleh ;
Abellan, Xavier ;
Balsamo, Gianpaolo ;
Bechtold, Peter ;
Biavati, Gionata ;
Bidlot, Jean ;
Bonavita, Massimo ;
De Chiara, Giovanna ;
Dahlgren, Per ;
Dee, Dick ;
Diamantakis, Michail ;
Dragani, Rossana ;
Flemming, Johannes ;
Forbes, Richard ;
Fuentes, Manuel ;
Geer, Alan ;
Haimberger, Leo ;
Healy, Sean ;
Hogan, Robin J. ;
Holm, Elias ;
Janiskova, Marta ;
Keeley, Sarah ;
Laloyaux, Patrick ;
Lopez, Philippe ;
Lupu, Cristina ;
Radnoti, Gabor ;
de Rosnay, Patricia ;
Rozum, Iryna ;
Vamborg, Freja ;
Villaume, Sebastien ;
Thepaut, Jean-Noel .
QUARTERLY JOURNAL OF THE ROYAL METEOROLOGICAL SOCIETY, 2020, 146 (730) :1999-2049
[18]   Deciphering "the language of nature": A transformer-based language model for deleterious mutations in proteins [J].
Jiang, Theodore T. ;
Fang, Li ;
Wang, Kai .
INNOVATION, 2023, 4 (05)
[19]   A wavelet - Particle swarm optimization - Extreme learning machine hybrid modeling for significant wave height prediction [J].
Kaloop, Mosbeh R. ;
Kumar, Deepak ;
Zarzoura, Fawzi ;
Roy, Bishwajit ;
Hu, Jong Wan .
OCEAN ENGINEERING, 2020, 213
[20]   Regional ocean wave height prediction using sequential learning neural networks [J].
Kumar, N. Krishna ;
Savitha, R. ;
Al Mamun, Abdullah .
OCEAN ENGINEERING, 2017, 129 :605-612