Giters

p-costa / SNaC

A multi-block solver for massively parallel direct numerical simulations (DNS) of fluid flows

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

implement a convective outflow

p-costa opened this issue · comments

commented

Hi Costa

Thank you very much for sharing this code.

Is the convective outflow boundary condition used for the square cylinder in your paper "A FFT-accelerated multi-block finite-difference solver for massively parallel simulations of incompressible flows"?

Thanks

commented

Hi @Watchmojo,

Thanks for your interest!

To answer your question, no -- in the paper describing the method, the simulation uses a zero pressure outlet at the boundary, not a convective outflow.

A convective outflow is something that certainly needs to be implemented in SNaC soon, and it seems pretty straightforward to implement something that does a decent job for the case in the paper (e.g., a constant velocity outflow). However, I was aiming at implementing something neater and more general (determine a local convective velocity based on the local flow condition near the outflow), and it seems to be more work than I anticipated!

However, I will revisit this problem soon.

commented

Hi Costa

Would you consider the convective boundary condition proposed in "Dong, S. (2015). A convective-like energy-stable open boundary condition for simulations of incompressible flows, J. Comput. Phys. 302: 300–328."? Semtex (https://users.monash.edu.au/~bburn/pdf/semtex-user.pdf) implemented it. Based on my experience, this convective boundary condition does help the stability of simulation for flows with excessive vortex shedding at the outlet.

Regards

commented

Hi again,

Thanks for sharing! Looks like an interesting method.

Looking at the reference I stumbled upon this work too:
C. Bozonnet, O. Desjardins, and G. Balarac. Traction open boundary condition for incompressible, turbulent, single-or multi-phase flows, and surface wave simulations. JCP 443 (2021): 110528.

Both seem like great candidates. I will definitely consider them. You seem to have experience with this, so I'd really appreciate your insight on the matter.

I'll try to implement this soon. If you plan to use SNaC and need this for your setup, just let me know and I can try getting it done sooner.

commented

Hi Costa

I will read this reference soon.

I am not using SNaC soon, and do not have time to develop CFD codes lately, although I would love to.

SNaC is a great platform for DNS/LES, and I look forward to seeing its future development.

I will definitely consider using SNaC for my future research.