Hello guys,
I am using PreCICE to reproduce some VIV phenomenon using openfoam & mbd tools. But my fluid solver always crashes when it comes to large deformation per time step. (As you can see in this pic, it crashed around the initial position where displacement is small while velocity is big.)
I know reducing time step size usually reliefs this behavior. However, the cost of it is formidable
considering the amount of my element number of fluid partition[around 10 million].
I am seeking some numerical tricks could let me have large time step and stability at the same time.[hahahaa]
My friends, kindly inspire me if you have any good suggestions. Thanks in advance.
Could you upload your precice-config.xml?
Common suggestions:
- Implicit coupling with IQN or Aitken
- RBF mapping
- Small coupling time window size
- Avoid subcycling for debugging
- Check the mesh quality and potentially use a better mesh motion solver