Search

Abstract: A method to compute structural loads on surface ships coupling a CFD solver with rigid or elastic representations of the ship hull is presented. One-way and two-way coupling approaches were used for the cases when the ship was considered elastic, in which the forces obtained from the CFD solver were used to compute the structural loads. In the two-way coupling method the hull deformation influenced the CFD solution, while in the one-way coupling method this feedback was not needed. The flow field around the ship was calculated with the URANS/DES overset solver CFDShip-Iowa version 4, with the structural response obtained using modal superposition. Predictions of ship motions and structural loads were compared with available experimental data for the S175 containership, obtained from a segmented elastic model. Predicted heave and pitch transfer functions in regular waves matched well with the experimental values. The computations showed that slamming events greatly affected the vertical bending moment amplitude, showing sharp peaks for the rigid model. The elastic models predicted well the ringing of the structure and showed that the ringing was mainly triggered by bow flare slamming. The one-way coupled approach with a virtual mass correction provided a solution that in some aspects had almost the same quality as the two-way coupled solution, at a fraction of the cost. [Copyright &y& Elsevier]

CFD simulations are conducted for zero-speed damaged passenger ship SSRC in calm water and waves with 6DOF motions including flooding procedure in calm water, roll decay in calm water and motions in regular beam waves for various wavelengths. The simulations model the 6DOF soft spring experimental mount, the one- and two-room flooding compartment configurations, including both intact and damaged conditions. For flooding and roll decay, simulations show ability to predict the trend of increases in roll period and damping due to flooding, as reported in ITTC. The damping magnitudes were often under-predicted with large errors while the roll period and compartment water height were well predicted. Two-room compartment simulation showed three times larger damping than one-room compartment cases whereas the roll period was similar for both conditions. For wave cases, all motions show primarily 1st order responses, except for parametric roll condition which shows large ½ harmonic responses for the intact ship. The 2nd order responses are small for both damaged and intact ship. The larger roll period and damping for the damaged ship shift the peak of responses to smaller wave frequency and reduce the amplitude of responses. The average error is often large for 1st order intact ship pitch and damaged ship surge and pitch. The errors are larger for most ½ and 2nd order responses. Large errors could be partially due to the complex mounting system in the experiment. Overall, current CFD results show better predictions than those reported for potential flow solvers even though the computational cost is larger. [ABSTRACT FROM AUTHOR]

Abstract: The motions and added resistance of KVLCC2 at Fr=0.142 and 0.25 with free and fixed surge in short and long head waves are predicted using URANS and validated against EFD datafor Fr=0.142. Verification studies show the results are fairly insensitive to the grid size and time step. CFD indicates no significant difference between free and fixed surge while EFD pitch motion and added resistance are affected. Added resistance was largest when the bow relative motion has largest amplitude and is about 180° out of phase with the waves. The decomposition of forces and moments works well for first harmonics but not for higher harmonics originated from radiation in long waves and diffraction in very short waves. Maximum responses occur near the resonance condition and near the maximum wave excitation force which is at λ/L=1.33 and long waves for surge/pitch and heave, respectively. Potential flow predictions for motions and the added resistance are further from the data than CFD. Local flow analyses show that added resistance is mainly induced by high pressure on the upper bow which is correlated with bow relative motion. The unsteady wave pattern is analyzed and the wake flow is compared with PIV measurements. [Copyright &y& Elsevier]