Learning Forum

[277623]

As part of my university project, I aim to simulate floating objects (solid, not deformable) moored with spring-like lines in a marine environment, subject to propagating waves. For fluid simulation, I am limited to using ANSYS FLUENT (AQUA and CFX are not available), though mechanical solvers are accessible—for example, for a two-way fluid–structure interaction (FSI) setup.

While simulating a freely floating object seems relatively straightforward, I have concerns about modeling mooring lines. I want to omit the physics of interaction of mooring lines with air and water but capture displacements of moored object and forces on tag lines due to wave impact.

I previously consulted an internet expert who mentioned that this is not possible in FLUENT due to certain limitations, including the following

1. Dynamic Mesh Requires a Rigid Body Definition (6DOF):

In ANSYS Fluent, when we enable dynamic mesh with 6DOF, the floating object (in this case, the donut) must be treated as a rigid body. This means Fluent only solves for its motion based on external hydrodynamic forces — it does not accept internal spring restoring forces (like F=kdx) directly in the dynamic mesh motion solver.

2. Spring Effects Require Structural Deformation (FSI/Mechanical):

For the spring forces to act on the body, the body needs to deform under those forces. This is only possible in ANSYS Mechanical, not within Fluent’s 6DOF model. But if we instead try to model the structure as deformable and use FSI coupling, then Fluent requires a fluid-structure interface—something that doesn’t work cleanly here because: Attempting to couple mooring springs directly onto a Fluent 6DOF rigid body through an FSI boundary leads to insurmountable modeling conflicts. In Fluent, a surface must be governed either by the 6DOF rigid body solver (for dynamic mesh motion) or by the FSI exchange solver (for structural feedback)—it cannot be both. Suppose you tag the donut as a 6DOF body and simultaneously declare its spring attachment faces as FSI. In that case, Fluent has no valid mesh connectivity or node definitions for those faces since a rigidbody 6DOF object has no deformable mesh. Therefore, The solver cannot interpret “spring endpoints,” and the competing motion definitions cause cell inversions and negativecell errors. In short, Fluent does not support a hybrid rigidbody + deformable interface on the same geometry, making proper two-way coupling of dynamic mesh waves and nonlinear spring lines impossible.

Could you please advise if it is possible or not to simulate floating object constrained with mooring lines in sea with waves? What are possible approaches (2-way FSI, custom made UDF)?

Are specified limitations above relevant ot they are wrong?

Regards,

Anton