Short-Term Prediction of the Sea State Dynamics

Forecasting of the sea level plays a key role to control on- and offshore facilities. First, we start with a determinstic time series method based on the state space embedding to determine the vector field of the nonlinear dynamical system and deduce the solution of its corresponding high-order differential equation. Second, We assume that the sea state is a stochastic process governed by a deterministic part and by noise so that this dynamical system can be modelled by the Langevin equation. We extract the nonlinear dynamical system considering fluctuations directly from a measured time series by estimating the drift vector and the diffusion matrix of the Fokker-Planck equation. In order to determine the prediction accuracy, the numerical solutions of the deterministic model and the Langevin equation are compared to the data values at future time. (© 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim)