how to judge that mesh we are using is fine enough for getting the good approximate result.
Re: boundary condition
It's called a mesh dependacy check. Do a simulation with one mesh, then repeat the simulation with a second mesh at a different grid resolution. If the results of important parameters are the same to within an acceptable accuracy than you may have achieved grid convergence (preferably check with a third grid resolution to be sure). If they are not the same, then re-run with a finer mesh until you manage to get a mesh-indepenant answer.
The book "Computational Fluid Dynamics" by Roache is recommended for further reading on grid dependancy and other accuracy issues.
|All times are GMT -4. The time now is 12:38.|