ALE - modify mesh viscosity at each time step
Posted: Mon Jun 29, 2020 11:23 am
Hello,
For a ALE simulation, the mesh viscosity defined inside the user_physical_properties.f90 is defined only once at the beginning of a simulation.
In my case, I fix the cell mesh viscosity value based on geometrical conditions based on the distance from the moving boundary. Under significant mesh deformation, this becomes critical since mesh viscosity values are kept constant based on the initial mesh configuration, but the cell distance from the moving boundary changes significantly due to the movement of the boundary. This leads to negative volume issues quite soon.
My question is: is there a way to change the mesh viscosity value of cells at each time step, thus taking into account the actual position of the moving wall? I tried to implement the mesh viscosity definition inside "usstr2" of the cs_user_fluid_structure_interaction.f90, which is meant to be called at each time step. Do you think this is a feasible way?
Do you have any more appropriate suggestion in order to achieve a mesh viscosity varying with time?
Thank you very much for your help.
Best regards,
Daniele
For a ALE simulation, the mesh viscosity defined inside the user_physical_properties.f90 is defined only once at the beginning of a simulation.
In my case, I fix the cell mesh viscosity value based on geometrical conditions based on the distance from the moving boundary. Under significant mesh deformation, this becomes critical since mesh viscosity values are kept constant based on the initial mesh configuration, but the cell distance from the moving boundary changes significantly due to the movement of the boundary. This leads to negative volume issues quite soon.
My question is: is there a way to change the mesh viscosity value of cells at each time step, thus taking into account the actual position of the moving wall? I tried to implement the mesh viscosity definition inside "usstr2" of the cs_user_fluid_structure_interaction.f90, which is meant to be called at each time step. Do you think this is a feasible way?
Do you have any more appropriate suggestion in order to achieve a mesh viscosity varying with time?
Thank you very much for your help.
Best regards,
Daniele