How to create a inflow generator for LES?

Questions and remarks about code_saturne usage
Forum rules
Please read the forum usage recommendations before posting.
Post Reply
MartianDuan
Posts: 74
Joined: Fri Aug 02, 2013 4:26 pm

How to create a inflow generator for LES?

Post by MartianDuan »

Dear users,

I am trying to do a simulation on the supercritical fluids flow in a heated pipe by using code_saturne 3.0.1. I need to create an inflow generator in the front of the flow domain, to make the flow develop.

Two line system will be introduced to generate the turbulent flow generator, as shown in the attached figure.
The 1st line is copying the velocity and temperature field from mesh layer 4 to mesh layer 1.
The 2nd velocity and temperature field at mesh layer 2 would be copied to mesh layer 4 to complete the circle.

Does anyone know how to achieve this goal in code_saturne 3.0.1.

Thanks in advance.
Attachments
2line systemn.png
(6.93 KiB) Not downloaded yet
Yvan Fournier
Posts: 4070
Joined: Mon Feb 20, 2012 3:25 pm

Re: How to create a inflow generator for LES?

Post by Yvan Fournier »

Hello,

I believe some colleagues have done this sort of thing occasionally, but am not sure how.

You could use a periodic duct with a velocity source term, but you then need to couple it to the main domain. This could probably be done with code/code coupling, but balancing the load between the 2 domains so as to have an efficient calculation may be tricky. Dumping time-dependent data and reading it is simple, but worse efficiency-wise (I am not fond of generating extra I/O).

Doing this in a more automated manner has been on the wish list for some time, but nothing definite is planned yet for this as far as I know. For non-LES calculations, the auto-inlet profile user example can be used, but that approach wouldn't work with LES.

Regards,

Yvan
MartianDuan
Posts: 74
Joined: Fri Aug 02, 2013 4:26 pm

Re: How to create a inflow generator for LES?

Post by MartianDuan »

Hi Yvan,

Thanks for your reply.

I will keep on searching solutions.

Thanks again,

Yu.
Yvan Fournier wrote:Hello,

I believe some colleagues have done this sort of thing occasionally, but am not sure how.

You could use a periodic duct with a velocity source term, but you then need to couple it to the main domain. This could probably be done with code/code coupling, but balancing the load between the 2 domains so as to have an efficient calculation may be tricky. Dumping time-dependent data and reading it is simple, but worse efficiency-wise (I am not fond of generating extra I/O).

Doing this in a more automated manner has been on the wish list for some time, but nothing definite is planned yet for this as far as I know. For non-LES calculations, the auto-inlet profile user example can be used, but that approach wouldn't work with LES.

Regards,

Yvan
Post Reply