EFFORT. Test of a lifeboat at a free fall from a very high ramp.
flow-induced motion of the lifeboat (which is regarded as a rigid body here) are computed as a coupled system. The flow-induced forces and moments acting on the lifeboat, and the resulting new position of the boat, are deter- mined in each time step.
Since the change in the boat’s position in- fluences the flow, the coupling must be con- sidered iteratively: the iterations in each time step are repeated until neither the flow field nor the position of the boat exhibits any ap- preciable change. Typically, five to ten itera- tions are needed for each time step. Since the number of the simulations required for an optimization study is large, it is important to ensure that the method is efficient from the viewpoint of computing effort. This necessi- tates local mesh refinement and efficient han- dling of the mesh adaptation to the position of the moving lifeboat.
In this case, the approach involves the use of overlapping meshes. With this method, a background mesh is fitted to the free surface and outer boundaries (the seabed, the oil rig or the ship), while the overlapping mesh is at- tached to the lifeboat, moving together with it but without deformation. The benefit offered here is that the method of overlapping meshes can be applied to any movements (also cap- sizing), and the boundary conditions (such as wave generation) can be implemented more easily than with other approaches.
The position of the centre of gravity and the inclination of the lifeboat relative to the water surface serve as initial conditions for the sim- ulation. The initial velocities in the horizontal and vertical direction as well as the angular speeds must also be specified. For a free-fall launch, they are equal to zero; in the case of ramp launching, the velocities must be deter- mined separately for the instant when the boat leaves the ramp.
When calculating the motion of the lifeboat in free fall through the air, simplifying meth- ods can be used to reduce the computing time required for CFD simulation. In this case, the simulation can be started with an initial condi- tion determined by means of alternative meth- ods at a height of several metres above the wa- ter surface.
Frequently, the simulation is only conducted for the water entry phase, since this is when the greatest loads occur – in terms of the load on the structure as well as the acceleration affecting the crew. This phase lasts for about two seconds. Usually, time steps of about a millisecond are used, so that the simula-
FINE MESH. Pressure distribution during water entry (corres- ponding to the last graph in the row on page 28).