Abstract: In complex marine environments, the irregular six-degree-of-freedom (6-DoF) motion responses of floating platforms significantly impact structural stability and energy production efficiency. Predicting platform motion trajectories enables proactive motion control, effectively mitigating operational risks under harsh environmental conditions. To address the challenge of motion response prediction under diverse wave conditions, this study employed Computational Fluid Dynamics (CFD) simulations to develop a numerical model of the floating platform and generate corresponding datasets. A Convolutional Neural Network (CNN)-based motion prediction model was subsequently established using these datasets. The model integrates mooring line tension data with real-time platform motion parameters to predict the platform's 5-second motion trajectory. The results demonstrate that the CNN-based prediction model achieves high-precision forecasting with an average computational latency below 2 ms on general-purpose computing hardware. Moreover, the prediction errors remain consistently confined within a low tolerance range across various operational scenarios, demonstrating robust real-time performance and exceptional environmental adaptability.