DNS of the turbulent flow around a square cylinder at Re=22000
Вставка
- Опубліковано 5 вер 2024
- A direct numerical simulation (DNS) of the turbulent flow around a square cylinder at Reynolds number 22000 (based on the diameter of the cylinder, D, and the inflow velocity, U). The dimensions of the computational domain are 30.5D × 54D × piD in the stream-wise, cross-stream and span-wise direction, respectively. The upstream face of the cylinder is located at 10D from the inflow and centered in the cross-stream direction. The origin of coordinates in placed at the center of the cylinder. At the inflow, a constant velocity profile, U, is prescribed and convective boundary condition are used at the outflow. The Neumann boundary condition is adopted at the cross-stream direction. Periodic boundary conditions are imposed in the span-wise direction. Finally, non-slip boundary conditions are imposed at the surface of the cylinder. We have used a 1272 × 1174 × 216 (~323M grid points) staggered mesh to cover the computational domain. The grid is stretched out away from the cylinder surface by means of a piece-wise tanh-function. A fourth-order energy-conserving spatial discretization has been used.
You can find more details in:
F.X.Trias, A. Gorobets, and A. Oliva. "Turbulent flow around a square cylinder at Reynolds number 22000: a DNS study", Computers & Fluids, 123:87-98, 2015. dx.doi.org/10.1...
Data is publicly available in:
www.cttc.upc.ed...
Authors: F.X.Trias, A.Gorobets, A.Oliva
A numerical solution just as beautiful as the fluid motion in nature! A very good reason to go into CFD.
Extremely beautiful, mind-blowing, outstanding and mesmerizing.
This is awesome!
Very beautiful, great work.
Reynolds? More like "Really cool videos!" Thanks for sharing.
Really cool! I was curious if you could share how you produced such a stunning video? Is this showing the turbulence intensity with a black-white color map? Did you use Paraview?
What is shown is the magnitude of the pressure gradient, i.e | grad p |. We used Tecplot but Paraview should be fine too.
Very beautiful and ethereal. Did you visualize a cross-stream slice of |grad(p)| , or did you visualize some sort of average, or did you visualize a few cross-stream slices superimposed on each other?
It is an instantaneous cross-stream slice.
Beauty of Science 🙂
What causes the asymmetry?
This is beautiful!
Hello author, I've recently been working on a mathematical physics chronicles content and I very much need to edit 5 seconds of footage using the footage from the video you created, I really, really need the 5 seconds of material. I'll be sure to credit the source of the footage and @ your channel when I'm done with my work. I very much recognize your work and thanks again. : )
(The above text was translated by me, using a translation program, so please forgive me if I use the wrong words) 😀
No problem at all if you properly refer the source. Thanks for your interests.
Beautiful!
Mesmerizing! Did you publish any paper on this simulation? Would be interested in the numerical part.
It was published in Computers&Fluids: F.X.Trias, A. Gorobets, and A. Oliva. "Turbulent flow around a square cylinder at Reynolds number 22000: a DNS study", Computers & Fluids, 123:87-98, 2015. dx.doi.org/10.1016/j.compfluid.2015.09.013
Thanks a ton!
Great work! Could you share any data points or U/Uref or similar for validation of other codes please?
www.cttc.upc.edu/downloads/SqCyl22K/
ive never seen a simulation of such detail before, AND IT WAS MADE 9 YEARS AGO
What platform did you use for the simulation and the visualization? And how much time did it take to complete?
I used an in-house code (details are in the description). Snapshots were generated with Tecplot.
So this is Static Pressure with a 2-color gradient map? Great work / video.
Brilliant! I'll check out the paper, but I'm assuming it was a spectral code, right? Beautiful postprocessing by the way!
Thanks for your comment! The code is not spectral. It is a finite volume in-house code. It uses a fourth-order energy-preserving spatial discretization. You can find all the details in the paper. Best!
@@FXTrias thank you for your reply! I've recently started my PhD on LES for compressible flows using Galerkin based methods, and one of the possible solutions we'll try is to use Spectral Elements, hence the assumption.
@@lucasgasparino6141 how did your PhD go?
@jpa_fasty3997 writing the thesis jajajajaja. We have a paper on review detailing the algorithm we implemented in an open source code, using spectral continuous galerkin method with our interpretation of guernonds entropy viscosity stabilisation. The code is under a constant development process, but right now we do get some impressive results! Plus on gpus it's ultra fast even when using p8 elements.
What quantity is actually being visualised here? Vorticity? Stress?
Could this be used to simulate/visualize wind coming over the tops of trees and across a runway?
CFD can certainly be used for this, but DNS is currently too computationally-heavy for this to be feasible with current technology.
Great!
Looks like little hurricanes downstream.
von Karman liked this
That computation is 100 times more expensive and time consuming than actually doing the test and recording the video.
Warum Flo?
I don't know why but the turbulence right after the first impact with the square looks like a mandelbrot image lol
this is not in vain - kinectic energy is transmitted from bigger vortices to smaller vortices acoording to the kolmogorov scale until it dissipates as heat
Can you define a square cylinder?
This is very useful in vastu!
broke: cuboid
woke: *square cylinder*
de qualité
i see how hurricanes form
Hi Xavier, please provide details of the numerical implementation or
papers where you presented this work. I am quite interested on the
Numerical implementation and I also have few questions about the order
of your time-space implementation
It was published in Computers&Fluids: F.X.Trias, A. Gorobets, and A. Oliva. "Turbulent flow around a square cylinder at Reynolds number 22000: a DNS study", Computers & Fluids, 123:87-98, 2015. dx.doi.org/10.1016/j.compfluid.2015.09.013
A square.... cylinder....
sir how to active DNS in fluent ....i know for RANS and LES ....BUT what step set up for DNS....help plz
I used an in-house code for this DNS and not Fluent.
why are there always street shitters asking questions that someone capable wouldn't be asking in these kind of videos? fuck off pajeet
WHAT THE F
IS A SQUARE CYLINDER
www.fxtrias.com/forced_convection.html
www.cttc.upc.edu/downloads/SqCyl22K/
Isn't a "square cylinder" an oxymoron? Aren't "cylinders" by definition circular, not square?
soy bien aweonao como que square cylinder pq chucha no poni que es un paralelogramo saco wea