[mathis302]

Hello Openfoam community !

I am currently modelling a floater for floating wind turbines (Volturn US) using Openfoam 2012, waves2foam as the incompressible two phase solver and Moody for the moorings. I am using a dynamic meshing technique based on morphing strategy to deform the mesh around the floater.

I am varying some parameters on the mesh in order to select the mesh that is the best computationally wise and in terms of accuracy, here the domain size. I use surge decay tests to run simulations, no waves are added.

When running the case with the smaller domain, the code runs until the end. However when I increase the size of the domain essentially width and length (see in MyVars) to the full domain (600m x 600m, I multiplied the width and the length by 2), WITHOUT changing the size of the base cells nor the expansion ratios in BlockMesh, the simulation runs for a certain time but stops after 90s of simulated surge test (see log file). I did not change anything else in the setup of the file so I believe it has something to do with the mesh.

I have done several other runs to try to understand what is wrong with the simulation:
With the full domain (600m x 600m base) : I have discovered that when I decrease the accelerationRelaxation in DynamicMesh Dict the simulation runs for longer (but still crashes after around 200s). This makes sense in a way that a lower accelerationRelaxation reduces the acceleration, thus the possible sudden increase in forces.

With the full domain (600m x 600m base) : I have increased the refinement level of the volumes surrounding the floater, which are two concentric elliptical cylinders with different refinement levels. It has to be noted that the refinement level of the surface of the floater does not change with the increase (refinement level 3). In those cases the simulations run until the end (400s).

When looking through the log file of the faulty case, I could not find anything anormal preceding the time step that could lead to the sudden increase of the velocity of the floater and then the divergence of the simulation. However I am relatively new to Openfoam and I hope that the community could help me out here !

Thanks in advance

Mathis