|
[Sponsors] |
September 12, 2001, 19:58 |
Node-by-node FEM
|
#1 |
Guest
Posts: n/a
|
I am currently working on an incompressible FEM solver. Basically, the solution methods I am looking into are various CG-methods (problem: pre-conditioning) and an explicit node-by-node solution. For CG and its variants, most of the publications dealing with pre-conditioning are for basic problems (driven cavity, etc) with small number of unknowns. The most effective pre-conditioners seem to require ILU-decomposition, this becomes very difficult for a large system (O(10^6) unknowns) as the original matrix already becomes too large to store in one array. In order to get ILU done, one would need the original matrix.... On the other hand, I am not familiar with papers dealing with explicit FEM solutions of this kind of problems.
My questions here are: 1. do you know of any papers that deal with pre-conditioning for incompressible N-S equations for large problems? 2. do you know of papers with node-by-node solution? Thank you. -Jarmo |
|
|
|
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 |
License server not visible from master node | Charles | FLUENT | 0 | October 30, 2007 18:48 |
comments on FDM, FEM, FVM, SM, SEM, DSEM, BEM | kenn | Main CFD Forum | 2 | July 18, 2004 19:28 |
The reasons why FEM has more computational quantit | Liu | Main CFD Forum | 0 | June 20, 2004 23:00 |
FEM or FVM for CFD | Astrid | Main CFD Forum | 18 | December 15, 2000 01:02 |