Physics-informed neural networks for phase-resolved data assimilation and prediction of nonlinear ocean waves

This work presents a novel Physics-Informed Neural Network (PINN) approach for nonlinear ocean wave data assimilation and prediction. The method leverages potential flow theory to parameterize wave dynamics and integrates physical constraints into the loss function, enabling accurate reconstruction and prediction of phase-resolved, nonlinear, and dispersive wave fields from sparse measurements. Validated against analytical solutions and laboratory experiments, this PINN framework infers full spatiotemporal wave and velocity potential fields efficiently and thus advances the use of physics-informed machine learning for ocean wave modeling and understanding.