Linearised Navier-Stokes time-dependant base flow
Hi Nek's, I have a question regarding the linearised Navier-Stokes solver while the underlying base flow is time-dependant. As far as I understand, one should simply have to set param(31) = 1 to achieve this (and p31 = -1 for a steady state base flow). What about the boundary conditions however? For instant, I want to investigate the linear dynamics of perturbations evolving onto a lid-driven cavity flow exhibiting a periodic limit cycle I have computed. How could I impose in userbc: ux = 1 uy = 0 uz = 0 for the periodic base flow time-evolution, and in the meantime: ux = 0 uy = 0 uz = 0 for the perturbation? Cheers, JC -- Jean-Christophe Loiseau Homepage <https://sites.google.com/site/loiseaujc/>
Dear JP, In perturb.f, the driver reads: do jp=1,npert if (nid.eq.0.and.igeom.eq.2) write(6,1) istep,time,jp 1 format(i9,1pe14.7,' Perturbation Solve:',i5) call perturbv (igeom) enddo jp=0 ! set jp to zero, for baseline flow So, the appropriate discriminator in userbc would be: if (jp.eq.0) then ux=1 else ux=0 end Best regards, Paul On Thu, 30 Jan 2014, [email protected] wrote:
Hi Nek's,
I have a question regarding the linearised Navier-Stokes solver while the underlying base flow is time-dependant. As far as I understand, one should simply have to set param(31) = 1 to achieve this (and p31 = -1 for a steady state base flow). What about the boundary conditions however?
For instant, I want to investigate the linear dynamics of perturbations evolving onto a lid-driven cavity flow exhibiting a periodic limit cycle I have computed. How could I impose in userbc:
ux = 1 uy = 0 uz = 0
for the periodic base flow time-evolution, and in the meantime:
ux = 0 uy = 0 uz = 0
for the perturbation?
Cheers, JC
-- Jean-Christophe Loiseau Homepage <https://sites.google.com/site/loiseaujc/>
Thanks a lot Paul. That's pretty much the idea I had, but I just wanted to make sure. Regards, JC 2014-01-30 <[email protected]>:
Dear JP,
In perturb.f, the driver reads:
do jp=1,npert
if (nid.eq.0.and.igeom.eq.2) write(6,1) istep,time,jp 1 format(i9,1pe14.7,' Perturbation Solve:',i5)
call perturbv (igeom)
enddo
jp=0 ! set jp to zero, for baseline flow
So, the appropriate discriminator in userbc would be:
if (jp.eq.0) then
ux=1
else
ux=0
end
Best regards,
Paul
On Thu, 30 Jan 2014, [email protected] wrote:
Hi Nek's,
I have a question regarding the linearised Navier-Stokes solver while the underlying base flow is time-dependant. As far as I understand, one should simply have to set param(31) = 1 to achieve this (and p31 = -1 for a steady state base flow). What about the boundary conditions however?
For instant, I want to investigate the linear dynamics of perturbations evolving onto a lid-driven cavity flow exhibiting a periodic limit cycle I have computed. How could I impose in userbc:
ux = 1 uy = 0 uz = 0
for the periodic base flow time-evolution, and in the meantime:
ux = 0 uy = 0 uz = 0
for the perturbation?
Cheers, JC
-- Jean-Christophe Loiseau Homepage <https://sites.google.com/site/loiseaujc/>
_______________________________________________
Nek5000-users mailing list [email protected] https://lists.mcs.anl.gov/mailman/listinfo/nek5000-users
-- Jean-Christophe Loiseau Homepage <https://sites.google.com/site/loiseaujc/>
Dear Nek's, 1. How to do LES or DNS? 2. How to do variable time stepping? 3. How to build a mesh for conic shape instead the cylindrical one given in turboJet? Thanks in advance, Barak
participants (1)
-
nek5000-users＠lists.mcs.anl.gov