Dear Libmesh Developers,
Recently, I am solving a complex number-based PDE. Thank you very much for
your help. Now, I meet a problem about error estimation. I discretize the
solved domain into coarse and fine meshes.
I get the solution on two meshes. I also adaptively refine the coarse mesh
to get another solution. The number of the refinement is 2 and 4. After I
get these solutions. I calculate the relative errors, that is
abs(Solution_{coarse, 2 refinements, 4
refinements}-Solution_{fine})/Solution_{fine}, for real component of the
complex number-based solution. You can find the error comparison in the
attached file. The relative errors between adaptive and fine meshes are
larger than those between coarse and fine meshes. I also compare the
relative errors for abs(solution). I used two error estimators (Kelly,
Patchrecovery). The same phenomena are observed. Could you give me some
advice? thanks a lot.
Regards,
Yujie
------------------------------------------------------------------------------
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
_______________________________________________
Libmesh-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/libmesh-users
