The Vortex Induced Motion (VIM) response of multi-column semi-submersible floaters may have a significant impact on the predicted fatigue life of mooring and riser systems. Over the last decades physical model testing using Froude-scaled floater geometries has been the main method to estimate the VIM response in the design stage. However, available field measurements seem to indicate that the VIM response when the floaters are installed, is typically lower than what is predicted in the model tests. Overly conservative design guidance for moorings and risers may be the result of only using standard model test information, with significant impact on costs.
Recent joint research projects, e.g. the VIM JIP led by MARIN and the RPSEA project led by Houston Offshore Engineering, investigated the effects of waves, current inflow conditions, external damping, mass ratio's and Reynolds number using both Computational Fluid Dynamics (CFD) and model testing to explore the main reason for the observed reduction in VIM response in the field. According to these investigations external damping, which may come from mooring and risers being dragged through the water during the VIM of the platform, was identified as the main cause for the reduced VIM response.
In this paper a coupled approach between the CFD code ReFRESCO and the time-domain mooring code aNySIM-XMF is utilized to investigate the effect of the damping from the riser and mooring systems on the VIM response of a semi-submersible platform. The VIM excitation from the hydrodynamic loading on the platform is solved by the CFD code while the damping from mooring and risers is simulated using dynamic anchor line models within the mooring code. A representative deep-water mooring and riser system consisting of 14 anchor lines and two risers was built based on a recent wave basin and VIM model test campaign. CFD simulations with an equivalent linearized mooring system are first carried out to identify the VIM response without damping from mooring and risers. Coupled simulations are then carried out to identify the effect of damping from mooring and risers on the VIM response of the platform. Different mooring models are investigated as well as the effect of different current profiles.
The calculated results for the equivalent linearized mooring system are first blindly benchmarked against available model test data which gives confidence in the CFD results. Additional sensitivity studies on the influence of the time step size showed that the calculated VIM results lie within 5% from each other.
Comparing the co-simulation results using different mooring models, i.e. an equivalent linearized stiffness matrix, a quasi-static catenary model and a dynamic lumped mass model, it can be observed that with the dynamic lumped mass model the VIM response reduces by 35-60% depending on the reduced velocity compared to the other two mooring models. This indicates that the hydrodynamic effects, i.e. added mass and damping, from the mooring and risers significantly reduce the VIM response of the semi-submersible platform. By setting the CD value to zero in the lumped mass model, a very similar VIM response was found compared to when using the catenary mooring model indicating that the drag component is responsible for the reduction in VIM response.
The novelty of the presented co-simulation approach is in the ability to estimate the hydrodynamic effects, originating from mooring and risers, on the VIM response of the semi-submersible floater. As a consequence more realistic and less conservative predictions for VIM response can be obtained in the design stage of projects, which is important for the prediction of the associated fatigue life of mooring and riser systems.