Abstract: Climate change has caused frequent extreme sea conditions in China's coastal regions. This has resulted in significant dynamic movements of the semi-submersible floating fan foundation at sea, leading to the relaxation-tensioning phenomenon of the mooring lines. This phenomenon greatly reduces the useful life of the mooring system and poses a safety risk to the overall operation of the wind turbine platform. A database of wave height and semi-submersible floating fan foundation load and dynamic response is created using numerical simulation. The full connection neural network method is then applied to establish a prediction model for the characteristics of the semi-submersible floating fan foundation load and dynamic response. This model enables the prediction of the load, dynamic response characteristics, and tension characteristics of the mooring system under extreme sea state conditions. By utilizing neural network prediction, it is possible to forecast the tension of the mooring lines, the dynamic response, and the load of the floating wind turbine foundation based on the wave conditions. This enables the assessment of the tension of the mooring lines, as well as the dynamic response and load of the floating wind turbine foundation.

Key words: semi-submersible floating platform, mooring system, fully connected neural network, deep learning, dynamic response