Objectives Viscous flow numerical simulations are widely applied in ship hull design due to their high accuracy in capturing complex flow field characteristics. However, they require substantial computational resources and long calculation times, limiting their efficiency in ship motion prediction and optimal design. To address this problem, this study integrates potential flow and viscous flow theories to develop a viscous-potential flow coupling method for ship motion prediction. The computational efficiency advantages of this approach are systematically analyzed under various wave conditions, providing technical guidance for balancing simulation accuracy with computational cost in intelligent ship design.

Methods The Wigley-III ship model, at a scale ratio of 1:50, was used as the study object. The flow field around the hull was divided into an outer potential flow region and an inner viscous flow region using a domain decomposition approach, with a coupling interface established at their junction to enable efficient transmission of wave data. The potential flow region employed the Boussinesq approximation and SWENSE field decomposition method for wave generation, while the inner viscous flow region utilized the RANS-VOF method combined with the SSTk-ω turbulence model to simulate ship-wave interactions. Simulations were conducted under various wave heights and wavelengths, and the ship's heave and pitch motions were analyzed. The simulation results were compared with model test data and full viscous flow simulation results to verify the reliability of the coupling method. Meanwhile, computational indicators such as calculation time, grid requirements, computational efficiency, and speed-up ratio were statistically analyzed, and their distribution characteristics under different wave steepness conditions were investigated.

Results The results show that the viscous-potential flow coupling method demonstrates excellent stability in energy transmission between the inner and outer wave domains, with a relative error of less than 1.03% across different wave heights and wavelengths. The simulated ship motions closely match experimental data, with relative errors for heave and pitch ranging from 5.14% to 10.99% and 5.83% to 11.64% respectively. These results are comparable to—or, under certain conditions, even better than—those obtained from full viscous flow simulations. In terms of computational efficiency, the coupling method requires only 39–47% of the grid resources needed by the full viscous flow method, and the average calculation time is reduced by 38.51%. The method achieves an average computational efficiency improvement of 86.8%, with an average speed-up ratio of 0.862. However, the efficiency gains are sensitive to environmental conditions: as wave steepness increases from 0.013 to 0.021, the speed-up ratio rises significantly; when the wave steepness exceeds 0.021, the speed-up ratio fluctuates and decreases due to enhanced wave nonlinearity.

Conclusions The viscous-potential flow coupling method provides high reliability for general engineering applications in ship motion simulation. It offers significant computational efficiency advantages, greatly reducing the computational cost. Its efficiency is sensitive to wave conditions, and the optimal application strategies should be developed based on wave steepness characteristics. This study contributes to establishing an autonomous trade-off mechanism between simulation accuracy and computational resources, supporting the intelligent and efficient design of future ships.