Flow over a flat plate at an AoA

[airwarriorg91]

Hello,
I am using xcompact3d to simulate the flow over flat plate at an angle of attack of 5 degrees at Re=10000. This is for the purpose of validation. I have modified the bccylinder.f90 file to input the geometry of the flat plate.
The 2.5% t/c flat plate used for the simulation.

Problem: I am neither getting the correct values of drag/lift coefficients nor the correct value of St number. Any suggestions/improvements to get the correct values ?

The parameters for simulation are as follows,

p_row=20 !20             ! Row partition
p_col=10 !10              ! Column partition

! Mesh
nx=321               ! X-direction nodes
ny=321                 ! Y-direction nodes
nz=160                ! Z-direction nodes
istret = 1            ! y mesh refinement (0:no, 1:center, 2:both sides, 3:bottom)
beta = 0.5    ! Refinement parameter (beta)

! Domain
xlx = 4      ! Lx (Size of the box in x-direction)
yly = 4            ! Ly (Size of the box in y-direction)
zlz = 2            ! Lz (Size of the box in z-direction)

! Boundary conditions
nclx1 = 2
nclxn = 2
ncly1 = 1
nclyn = 1
nclz1 = 0
nclzn = 0


! Flow parameters
iin = 0               ! Inflow conditions (1: classic, 2: turbinit)
re  = 1000 ! nu=1/re (Kinematic Viscosity)
u1  = 20              ! u1 (max velocity) (for inflow condition)
u2  = 20              ! u2 (min velocity) (for inflow condition)
init_noise  = 0.005   ! Turbulence intensity (1=100%) !! Initial condition
inflow_noise = 0.0    ! Turbulence intensity (1=100%) !! Inflow condition

! Time stepping
dt = 0.00001              ! Time step
ifirst = 0            ! First iteration
ilast = 500001     ! Last iteration

! Enable modelling tools
ilesmod=0             ! if 0 then DNS
iscalar=0             ! If iscalar=0 (no scalar), if iscalar=1 (scalar)
iibm=2                ! Flag for immersed boundary method

Thanks,
Gaurav

[slaizet]

Dear Gaurav,
What is the size of your control volume for the computation of the lift and drag? It should include the plate.
Best.
Sylvain

[airwarriorg91]

Yeah, It includes the flat plate. The plate is centered at 1.5, 2 and the control volume boundaries are,

xld(1) = 0.8 !X left for volume control
xrd(1) = 2.2 !X right for volume control
yld(1) = 1.5 !Y bottom for volume control
yud(1) = 2.5 !Y top for volume control

[slaizet]

I guess you resolution is quite coarse so I would aim to increase the resolution if possible.
Other than that, I do not see any issue with your set-up.
Best,
Sylvain

[airwarriorg91]

Okay sir ! I will try with a finer mesh. Also sir, I have one more question regarding the mesh refinement factor.
What does the mesh refinement factor (Beta) actually represent, is it a ratio of something ? I wanted to know about the mathematical relation which determines the refinement in the y-direction.
Thank You !

[slaizet]

You will find all the details in Laizet, S., & Lamballais, E. (2009). High-order compact schemes for incompressible flows: A simple and efficient method with quasi-spectral accuracy. Journal of Computational Physics, 228(16), 5989-6015.

The metric is quite complicated :-)

Best,

Sylvain

[airwarriorg91]

Okay sir. I will look into it. Thank you !

[airwarriorg91]

Merry Christmas Sir !
Sir, I am trying to implement another approach here i.e. by rotating the input velocity instead of the geometry. Thus, making the flat plate perfectly in the domain .

I modified the inflow subroutine of the BC-Cylinder.f90 file and did the following changes to implement the approach,

    A = 5.0 

    ca = 0.996194698 

    sa = 0.087155743 


    do k=1,xsize(3)

       do j=1,xsize(2)

          bxx1(j,k)=u1*ca+bxo(j,k)*inflow_noise

          bxy1(j,k)=u1*sa+byo(j,k)*inflow_noise

          bxz1(j,k)=zero+bzo(j,k)*inflow_noise

       enddo

    enddo

The error in drag and lift was comparatively more than the approach where we rotate the geometry also the flow was different.

Comparision.mp4

I feel I might have missed some modifications to correctly implement the approach. Can you help regd. the same ?

Thank You.