[ziemowitzima]

Dear All,
I am solving flow in abdominal aorta. To see my first results please visit and go to the bottom of the page (two last animations):


https://sites.google.com/site/engcalculusev/research


Main problem in this simulation is inlet velocity boundary condition. This boundary condition is space and time varying. It is described by sin function to simulate blood enter to the aorta. But the main problem is the back flow of the blood. It happened for some period of time. The inlet changes to outlet and two outlets becomes inlets
(to see full velocity bc at the inlet visit:
http://www.youtube.com/watch?v=7YrnD39cQYQ
).
In the case you have seen by clicking the above link the inlet bc are simplified (there is no back flow).


Standard bc for velocity at the outlets (exits) is zeroGradient.


And here complications begin. When keeping zeroGradient bc on the outlets during back flow phase errors arise and solution diverge. In my opinion it is caused by vortex structures at the outlets which cause that on some regions of the exit fluid enters the aorta on some others fluid exits (see figure ). With time velocity become larger and larger...
Then solver is trying to keep constant mass flow.
Does anyone have an idea how to solve this problem ?
cross_section_y-plane_vectors_velosity_mag_t=0.2.jpg

[ljrmicroport]

Dear, have you solved this problem? I met almost the same one.