I am getting convergence issues in the first coupling iteration itself and the solver stops. I tried refining the mesh and also tried reducing the time step size. It is still not working. If anyone who already worked on the similar problem, please help me in overcoming these convergence issues.
Check the multiphysics solver of SU2 which can solve FSI problems. It can deal with unstuctured mesh very well and still easy to use.
https://su2code.github.io/tutorials/Static_FSI/
Check the above tutorial.
Hope it helps.
Hi,
Which software you use for drawing and / or modelling your model?
Hi
You should increase the pressure gradually from zero to your intended pressure. you can define a profile for that and during one second increase the pressure.
Hi have a look at the presentation of Philip Cardiff - it is very insightful in FSI, and also covers Frequent Problems (at the end of the slides) such as convergence issues as you mention.
Conference Paper Training: Solid mechanics and fluid-solid interaction using ...
Hi,
That depends on a lot of factors. I would start with: what are the physical models you are using for the fluid and solid parts? What is your FSI coupling strategy? Monolithic or partitioned? If partitioned (which I suppose is the case), the physics of FSI in arteries make it a very challenging numerical problem: there are several publications, for example, on using more robust techniques to solve the FSI non-linear system, also on the added-mass effect that yields some computations unstable (see the work by Causin and co-workers in Article Added-mass efiect in the design of partitioned algorithms fo...
).For your specific problem for example, depending on the temporal discretization schemes you used, decreasing the time-step won't help (see this other work by Forster and co-workers Article Artificial added mass instabilities in sequential staggered ...
).I hope this helps,
iago
- Badges
- Science topic
- Similar topics
- Indexes