|
[Sponsors] |
September 30, 2007, 23:22 |
FEM vs finite-differences
|
#1 |
Guest
Posts: n/a
|
Hi!
I just starting to work with a project and free to choose a computational method which I'll use in my code. I may be wrong but I think that it would be better to use FEM if one wants to solve RANS eqs for a flow with strong shocks and complex geometry. I expect the solution to belong W2 but not C2, so it's not smooth enough for finite-differences and will converge only in a "weak" sence. What is Your professional opinion, FEM or FDM? If FDM, continuous or discrete? Does FEM require much more CPU time than Finite-Difference Method (for strong shocks)? Thank You! |
|
October 1, 2007, 09:58 |
Re: FEM vs finite-differences
|
#2 |
Guest
Posts: n/a
|
You've not considered Finite Volume Method (FVM)? This has the potential for capturing strong shocks and enforces the unlying conservation laws.
|
|
October 1, 2007, 14:04 |
Re: FEM vs finite-differences
|
#3 |
Guest
Posts: n/a
|
I think You are right. Finite volumes would be the best choise for flow with strong shocks. Thank You very much for advice! //natalie
|
|
October 1, 2007, 15:31 |
Re: FEM vs finite-differences
|
#4 |
Guest
Posts: n/a
|
with simple walk in literature, u see that fdm, fem or fvm could solve PDE with discintinuty or singularity, previously FDM has limitation to resolve complex geometry, that with appearance of local refinement/immersed bc/multiblock it is almost resolved, also fem had conservation problem which is resolved by appearence discontinuous galerkin based methods, ...
so selection of method seems to be arbitrary (only implementation difficulty and computational cost is issue !) in my opinion: FEM is more easy to implement, particularly for a general solver, but it almost needs more computational cost, also mathematical theory and related open source codes of FEM have more matuarity which is another interesting issue. |
|
Thread Tools | Search this Thread |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
I developed an FEM toolkit in Java: FuturEye | nkliuyueming | Main CFD Forum | 7 | January 29, 2016 14:28 |
comments on FDM, FEM, FVM, SM, SEM, DSEM, BEM | kenn | Main CFD Forum | 2 | July 18, 2004 19:28 |
Multigrid for finite differences | Enda Bigarelli | Main CFD Forum | 4 | November 6, 2001 13:04 |
Finite differences | Zdravko Stojanovic | Main CFD Forum | 8 | March 24, 2001 23:31 |
U.F.D. :Finite Differences on Unstructure grids | STROHM | Main CFD Forum | 4 | August 6, 1998 16:25 |