ATDNNS: An adaptive time-frequency decomposition neural network-based system for tropical cyclone wave height real-time forecasting

Fan Meng, Danya Xu, Tao Song

Abstract: Highlights•Accurate real-time forecasting of tropical cyclone wave heights using offshore distributed sensors.•The study involves 28 TC events with data collected from 14 coastal buoys over 9 years.•Combination of adaptive time–frequency decomposition and LSTM improves forecast performance.•The forecasting performance is superior to that of the baseline model and previous studies.