[lemonice87]

Hi All,

Apologies if the question has been answered before, but can you and how do you give unsteady inlet bounday condition, meaning I want to give a decreasing velocity at the inlet boundary as time advances.

Thanks for your help!

[fpalacios]

Originally Posted by lemonice87

Hi All,

Apologies if the question has been answered before, but can you and how do you give unsteady inlet bounday condition, meaning I want to give a decreasing velocity at the inlet boundary as time advances.

Thanks for your help!

This is an important topic. If you are interested in wind gust simulation, there is an implementation of a technique that is based in adding virtual grid velocity. This is a useful method, otherwise it is very easy to generate spurious waves and transients.

If the inflow profile changes very slowly with the time, you can directly modify the boundary conditions in solver_direct_mean.cpp subroutine

void CEulerSolver::BC_Inlet(CGeometry *geometry, CSolver **solver_container, CNumerics *conv_numerics, CNumerics *visc_numerics, CConfig *config, unsigned short val_marker)

Remember that config->GetExtIter() gives you the index of the current iteration.

In the past (SU2 v2.0.5) there was an option in SU2 for unsteady_farfield, you can take a look at that version if you are interested.

Thanks for your interest in SU2,
Francisco

[lemonice87]

Thanks for the help, I will look to both proposed solution.

When you say "If the inflow profile changes very slowly with the time" you mean with respect to my tiume advancment step correct?

Anyway I am simulating a supersonic object slowing down (whilst remaing in supersoninc flow) wouldn't that be less problematic? Especially as I am using the Euler solver for now.

Thanks for the help

[Santiago Padron]

Yes, it will be better to have the inflow profile change slower than the time advancement step.
Yes, simulating the supersonic flow will make the problem less problematic.
Thanks for your interest and lets us know how your implementation goes.

Santiago

[mmj_blackforce]

Originally Posted by fpalacios

This is an important topic. If you are interested in wind gust simulation, there is an implementation of a technique that is based in adding virtual grid velocity. This is a useful method, otherwise it is very easy to generate spurious waves and transients.

Francisco

Dear Francisco:
Can you please explain if adding virtual grid velocity is possible in fluent?
Regards