Prediction of Significant Wave Height Based on Gated Recurrent Unit and Sequence-to-Sequence Networks in the Taiwan Strait

Authors

  • Yindong Zeng College of Oceanic and Atmospheric Sciences, Ocean University of China, Qingdao 266071, China & Marine Forecasting Center of Fujian Province, Fuzhou 350003, China
  • Jinshan Ma Institute of Computer Applications, Fujian Agriculture and Forestry University, Fuzhou 350000, China
  • Xinping Chen National Marine Hazard Mitigation Service, Ministry of Natural Resources of China, Beijing 100194, China
  • Minquan Guo Marine Forecasting Center of Fujian Province, Fuzhou 350003, China
  • Zaichang Ren Marine Forecasting Center of Fujian Province, Fuzhou 350003, China
  • Yanqi Jiang College of Harbour, Coastal and Offshore Engineering, Hohai University, Nanjing 210098, China
  • Zhenchang Zhang Institute of Computer Applications, Fujian Agriculture and Forestry University, Fuzhou 350000, China

DOI:

https://doi.org/10.31577/cai_2022_3_885

Keywords:

Wave forecasting, significant wave height, gated recurrent unit, long short-term memory, sequence-to-sequence

Abstract

Wave forecasting approaches based on deep learning techniques have recently made a great progress. In this study, we developed a deep learning model based on Gated Recurrent Unit (GRU) and sequence-to-sequence neural networks (GRUS), to improve the forecasting accuracy of significant wave heights for the Taiwan Strait, where ocean waves and winds own their unique characteristics. The performances of our proposed GRUS model and the other deep learning models based on WaveNet and Long Short-Term Memory (LSTM) were compared by means of wind and wave observations at three buoys in the study area. Model parameters were optimized by means of various model experiments. Performance comparison illustrates that our proposed GRUS model outperforms the other models in 24-hour Hs forecasting, while the GRUS has extraordinary ability for short-term prediction (prediction horizon is less than 6 h). Moreover, for high wave states prediction (e.g., wave height over 4 m), the GRUS has the strongest prediction ability among the models, in which forecasted wave heights are mostly lower than the corresponding observations.

 

Downloads

Download data is not yet available.

Downloads

Published

2022-09-08

How to Cite

Zeng, Y., Ma, J., Chen, X., Guo, M., Ren, Z., Jiang, Y., & Zhang, Z. (2022). Prediction of Significant Wave Height Based on Gated Recurrent Unit and Sequence-to-Sequence Networks in the Taiwan Strait. COMPUTING AND INFORMATICS, 41(3), 885–904. https://doi.org/10.31577/cai_2022_3_885