How to solve FSI problems with high stiffness using IBFE

[Haowei Du]

Hi all,

I'm trying to solve a fluid-structure interaction problem with high stiffness material (Young's modulus around 1e10 Pa) using IBFE. Currently, I have to use extremely small time steps (around 1e-9s) to maintain stability. This makes the computation very time-consuming. Are there any strategies or methods to increase the time step while maintaining stability? Any suggestions would be greatly appreciated.

Du

[Boyce Griffith]

On Mar 20, 2025, at 9:10 AM, 'Haowei Du' via IBAMR Users <ibamr...@googlegroups.com> wrote:

Hi all,

I'm trying to solve a fluid-structure interaction problem with high stiffness material (Young's modulus around 1e10 Pa) using IBFE. Currently, I have to use extremely small time steps (around 1e-9s) to maintain stability. This makes the computation very time-consuming. Are there any strategies or methods to increase the time step while maintaining stability? Any suggestions would be greatly appreciated.

The way to think about the time stepping scheme that is used currently is that it is like an explicit mechanics solver, and so if you have a very stiff material, you must take very small time steps.

We are working on implicit solvers that could allow for larger time step sizes, but currently we do not yet have an implicit solver that is actually faster than an implicit solver that uses very small time steps.

If the material is stiff enough so that it makes sense to treat it as a rigid body, you also might be able to improve performance by using a rigid body dynamics structural model instead of an elastic model.

— Boyce

Du

--
You received this message because you are subscribed to the Google Groups "IBAMR Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ibamr-users...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ibamr-users/d1f55f20-a643-4e1a-968f-fb778724e231n%40googlegroups.com.