[xzdbjx]

Dear all, I am trying to simulate the neutral equilibrium atmospheric boundary layer using Fluent 16.0. When define the upper boundary of my model, the fluid has been considered either to be driven by tangential velocity or shear stress. For shear stress scheme, i have defined the upper boundary as symmetry. However, I am not sure how to define the shear stress using the UDF in Fluent. The equation of the shear stress are given as follows.
I really need your help.
Best wishes!

[AlexanderZ]

Check DEFINE_PROFILE macro in Ansys Fluent Customization manual

best regards

[xzdbjx]

Hi AlexanderZ

Thanks for your reply

Perhaps i did not illustrate my problem clearly. Now I am working on the simulation of the equilibrium atmospheric boundary layer using Fluent. A cuboid has been established as fluid region. For the upper boundary condition, the literature has shown that it should be driven by a constant shearing stress as shown in the aforementioned equation. I wanna prescribe an equivalent volume momentum source by creating a sub-domain one cell thick at the top boundary to drive the flow.

As far as i am concerned, since the shearing stress has the unit N/m2, while the volume momentum has the unit N/m3. Thus, if the constant shearing stress is divided by the cell height, it must be the wanted momentum source. With this consideration, the following code has been listed:

DEFINE_SOURCE(xmom_source,c,t,dS,eqn)
{
real CeC[ND_ND];
real yC,source;
real ymax=1.75; //the y-coordinate of the cell-centroid of the upper cell
real height=0.1; //the height of the upper cell
real rol=1.225; // the density of the air 1.225kg/m3
real tau;
Domain *domain=Get_Domain(1);
tau=rol*pow(ustar,2);
thread_loop_c(t,domain)
{
begin_c_loop(c,t)
{
C_CENTROID(CeC,c,t);
yC=CeC[1];
if (yC>=ymax)
{
source=tau/height;
dS[eqn]=0.;
return source;
}
}
end_c_loop(c,t)
}
}

Please tell me if my consideration is reasonable? Is there any other things i need to pay attention?

Thank you for the help.

Best regards!

[LuckyTran]

Is all this udf work necessary? You can apply a shear stress using a wall type boundary condition.

[xzdbjx]

Originally Posted by LuckyTran

Is all this udf work necessary? You can apply a shear stress using a wall type boundary condition.

Thank you for your reply. Some typical literature have been investigated such as:
[1] Richards, P. J. , & Norris, S. E. . (2011). Appropriate boundary conditions for computational wind engineering models revisited. Journal of Wind Engineering & Industrial Aerodynamics, 99(4), 257-266.
[2]Blocken, B., Stathopoulos, T., & Carmeliet, J. (2007). CFD simulation of the atmospheric boundary layer: wall function problems. Atmospheric environment, 41(2), 238-252.

In the pioneers' works, they have suggested to consider the shear stress as an equivalent volume momentum source. Attached please find a illustration of the shear stress boundary condition.

[LuckyTran]

Applying the shear stress should be a piece of cake (it's a built-in option). What the wall BC does not do is the provide a sink for the dissipation / gradient of epsilon condition.


The equivalent momentum source you need to provide is the divergence of tau. So you are right that you need to divide by something. I'll have to check later if you did it correctly or not.

[xzdbjx]

Thank you for your reply. The problem has been solved.

Best wishes!

[Gonliveira110]

How were you able to solve the problem?



Best regards.