Data-Driven Digital Twin-Based Prediction Approach for Buoy Motion and Mooring Forces in Extreme Sea States

This study proposes a three-level data-driven digital twin prediction framework that integrates physical mechanisms, a sample database, and intelligent algorithms, enabling high-precision and rapid prediction of buoy motion response and mooring force under extreme sea conditions. Taking a coastal microwave antenna buoy as the research object, a high-fidelity physical model is first established. A standardized sample database covering extreme sea conditions is then constructed through batch simulations under multiple working conditions.

The nonlinear fitting performance of full second-order polynomial regression and nonparametric regression is systematically compared, and the optimal algorithm is selected to develop the intelligent prediction model. The results indicate that the relative error between the simulated values of the physical model and experimental measurements is less than 10%, confirming the high reliability of the model.