FEM vs FVM on the Web
News | Posted on February 3rd, 2012 by Phil Kinnane
I noticed a lot of activity when I posted the question concerning “FEM vs FVM” in a previous blog post, so I started looking into it a bit. A lot of people went to the COMSOL Discussion Forum thread on the matter, from the blog, which got me thinking that this must be a hot topic out there.
How disappointed I was when I typed ‘FEM vs FVM’ into Google and saw that there wasn’t much to sink your teeth into. My blog post ranked pretty highly, which explains how little this has been debated.
While there were a couple of decent presentations summarizing both methods, I was beginning to feel that the Discussion Forum thread was the most comprehensive summary I was going to get on the subject.
Yet, there was one nugget I found on Youtube (a picture is worth a thousand words, as you can see below). A 3-second video shows FEM being able to handle the retrograde oscillatory flow (apparently pretty common in biomechanics) in the figure on the right, while FVM’s correction method either ignores or rubs out this effect. And in the abstract connected to this research, the author concludes “the computation time required for convergence was found to be significantly smaller for the finite element approach” (you can find the abstract here). Get your teeth into that!! I’ll keep looking at this subject for sure.
Here’s the 3-second video I found on YouTube: