Hello:
In the velocity correction scheme for incompressible flow , the bc for pressure equation is all Neumann bc , i want to know how to solve this linear system in Nektar.
In my own code , I arbitrily set some point pressure value to 0. But it seems I cann't solve the 2d konasky flow problem to high accuracy , I only get 1E-8 compared with the eaxct solution.
and the pressure gradient comparison gives an error of 1E -3. i cann't get them lower nomater how to increase N and decrease timestep .
Could you please give me some advices.
regards .
 | ztdepyahoo |
ztdepyahoo@gmail.com |