Computational Fluid Structure Interaction - Computational Hemodynamics: Theory, Modelling and Applications - page 148

Fig. 5.41 Fluid-structure coupling flowchart
Transfer of the fluid flow variables (e.g., velocity, pressure, force, temperature)
from the fluid solver to the structural solver, without the return influence from struc-
ture to fluid is referred to as a one-way FSI coupling. One-way transfer is appro-
priate when displacement and temperature differences calculated in the structural
application are small enough to have insignificant impact on the fluid analysis. In
two-way coupling, deformation of the fluid-structure interface on the structural side
is transferred to the fluid solver to deform the fluid mesh. Therefore, the two-way
FSI coupling solution is more needed for cases with larger deflections where the
fluid field is strongly influenced by the structural deformation.
Using partitioned methods, FSI coupling solutions are separated into fluid dy-
namics, and structure dynamics. At interface, information for the solution is shared
between the fluid and structural solver. The information exchanged is dependent on
the coupling method. For one-way coupling, only the fluid pressure acting at the
structure is transferred. For two-way-coupling calculations, the displacement of the
structure is also transferred to the fluid solver.
A structural solver will most probably use finite elements on unstructured grids,
while a fluid solver might use a finite volume approach on Cartesian grids. There-
fore, interpolation methods plays a key role in mapping data from one mesh to the
other. In addition, a fluid with low viscosity might be highly turbulent and require
very small time steps to stay stable, while the structure could advance with much
larger steps. Sub-cycling might improve the situation by allowing the fluid solver
to accumulate several small time steps on its own before transferring the variables
to the structural solver.
One argument for partitioned approaches is their flexibility. Flexibility in both
the choice of coupling algorithms, and the choice of software for solving the equa-
tions. Since the solvers are separated, they use their own discretisation method, lead-
ing to nonmatching grids on both sides of the coupling interface between fluid and
structure. In order to transfer loads across a dissimilar mesh interface, the nodes of
one mesh must be mapped to the local coordinates of an element in the other mesh.
The fluid nodes must be mapped to the solid elements to transfer displacements.
Likewise, solid nodes must be mapped to the fluid elements to transfer stresses.