Re: Concepts about model reduction
106 views
Skip to first unread message
Prosper
Oct 8, 2012, 2:34:35 PM10/8/12
to mor4...@googlegroups.com
Based on my search online, almost all the MOR techniques are for dynamic systems. Is there any MOR method for static finite element problem?
On Monday, October 8, 2012 1:55:52 PM UTC-4, Prosper wrote:
On Monday, October 8, 2012 1:55:52 PM UTC-4, Prosper wrote:
Hi all!I have some basic understanding about the model reduction concepts. If, for example, I can reduce the order of the stiffness matrix for a static 2d or 3d problem, thus saving computational efforts, is that considered to be model reduction?Also, are there people doing something like that? I heard that people can reduce the model order by sonar measurement of displacement. What should I search to learn more about that?Thanks!Best,
Evgenii Rudnyi
Oct 8, 2012, 2:46:41 PM10/8/12
to mor4...@googlegroups.com
For a static problem the simplest is just to solve it. Here the question
would be what a solver will be the fastest, direct, iterative,
multi-grid. In a way an iterative solver could be thought as a MOR
version of the direct solver but I am afraid that this analogy is not good.
If you think of an optimization though, you may want to look at
reduced-basis methods.
Evgenii
On 08.10.2012 20:34 Prosper said the following:
would be what a solver will be the fastest, direct, iterative,
multi-grid. In a way an iterative solver could be thought as a MOR
version of the direct solver but I am afraid that this analogy is not good.
If you think of an optimization though, you may want to look at
reduced-basis methods.
Evgenii
On 08.10.2012 20:34 Prosper said the following:
Reply all
Reply to author
Forward
0 new messages