Inlet velocity profile
OpenLB – Open Source Lattice Boltzmann Code › Forums › on OpenLB › General Topics › Inlet velocity profile
- This topic has 4 replies, 2 voices, and was last updated 2 years, 12 months ago by Anand.
-
AuthorPosts
-
January 20, 2022 at 1:08 pm #6253AnandParticipant
Hello,
I am simulating turbulent single-phase flow in a horizontal pipe using the D3Q27 descriptor, SmagorinskyBGKdynamics, no-slip condition at the wall and I use CirclePoiseuille3D velocity profile at the inlet.
Are there other options for inlet velocity such as constant velocity etc?Following is my code for boundary values:
PolynomialStartScale<T,int> StartScale( iTmaxStart, converter.getCharLatticeVelocity() );
// Creates and sets the Poiseuille inflow profile using functors
// int iTvec[1] = {iT};
//T frac[1] = {};
// StartScale( frac,iTvec );
//std::vector<T> maxVelocity( 3,0 );
//maxVelocity[0] = 2.25*frac[0]*converter.getCharLatticeVelocity();
int iTvec[1]= {iT};
T maxVelocity[1]= {T()};
StartScale( maxVelocity,iTvec );
CirclePoiseuille3D<T> poiseuilleU( superGeometry,3,maxVelocity [0] );
sLattice.defineU( superGeometry,3,poiseuilleU );Thank you
Regards
AnandaJanuary 20, 2022 at 9:19 pm #6254AnandParticipantI check the velocity profile at the end of the pipe.
I see the profile is exactly the same at inlet (Poiseuille flow profile) even though I Reynolds number is 10005 and I use SmagorinskyBGKdynamics.Please can someone explain to me what is wrong with my code, am I missing something?
Thank you in advance.
Regards
AnandJanuary 20, 2022 at 10:07 pm #6256AdrianKeymasterHow did you determine that your simulation models Re = 10005? At which resolution and for what timespan are you simulating? Did you check the entire Paraview output? Did you base your setup on some existing example of ours?
January 20, 2022 at 10:20 pm #6257AnandParticipant1) I calculated using: Re = (inletvelocity*pipediameter)/(kinematicviscosity); (inletvelocity=0.9423m/s, Pipediameter=0.04m; kinematicviscosity=0.0044 (kg/m-s)/1168 (kg/m3))
I also see the same Reynolds number in the UnitConversion Information Parameters output.2) I use N=50 and M=50
3) Timespan = 10 second
4) I use .pvd file in paraview and put a vertical line at the outlet and plot velocity vs diameter of pipe
5) I just modify your the existing examples.
January 20, 2022 at 10:31 pm #6258AnandParticipantHi Adrian,
One more thing I need to mention.
If I keep everything the same and just change the descriptor to either D3Q19 or D3Q27, the results are different.
At the outlet
D3Q27 gives me the same velocity profile as inlet (Poiseuille)
D3Q19 give me fluctuation at the outlet no particular pattern -
AuthorPosts
- You must be logged in to reply to this topic.