Deep Koopman Embeddings for Unidirectional Phase-Resolved Wave Prediction

The phase-resolved wave prediction is of vital im-portance for improving safety in ocean engineering. However, it is a challenge for traditional physical methods to predict wave elevation accurately and efficiently at the same time. In this work, we develop a phase-resolved wave prediction model based on combined deep learning and the Koopman theory. The model can calculate in real-time while ensuring prediction accuracy. In addition, the model can predict wave elevation under the sea state 4, demonstrating certain generalization performance.