Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.
Error with finer meshes
Posted 11 oct. 2016, 09:36 UTC−4 Geometry, Mesh 1 Reply
Please login with a confirmed email address before reporting spam
Dear,
My program which is about the solution of some diffusion and reaction equations in a cube works very well for coarser and extra coarser meshes, however when I repeat the analyse with finer meshes, the program does not converge after a certain time step. And still gives good results up to that point. Can you please help me to understand the reason of this. Thanks in advance.
My program which is about the solution of some diffusion and reaction equations in a cube works very well for coarser and extra coarser meshes, however when I repeat the analyse with finer meshes, the program does not converge after a certain time step. And still gives good results up to that point. Can you please help me to understand the reason of this. Thanks in advance.
1 Reply Last Post 12 oct. 2016, 10:57 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
8 years ago
Hi there,
If nothing else has changed within the model other than the mesh density, then it may well be a memory limitation on the machine you are running the simulation on.
Presumably you can justify the need to increase mesh density by studying the results from the models with coarser meshes? Is the solution accuracy not quite good enough with the coarser meshes?
I know that sometimes the discretisation settings and method of result recovery can be experimented with in order to improve the result accuracy, negating the need to cram more elements into the model.
Aside from these general suggestions, you may need to dig into the solver settings for the particular Physics that you are dealing with, (I'm afraid I can't help with that though)!
Hope you manage to get a solution!
Mark
If nothing else has changed within the model other than the mesh density, then it may well be a memory limitation on the machine you are running the simulation on.
Presumably you can justify the need to increase mesh density by studying the results from the models with coarser meshes? Is the solution accuracy not quite good enough with the coarser meshes?
I know that sometimes the discretisation settings and method of result recovery can be experimented with in order to improve the result accuracy, negating the need to cram more elements into the model.
Aside from these general suggestions, you may need to dig into the solver settings for the particular Physics that you are dealing with, (I'm afraid I can't help with that though)!
Hope you manage to get a solution!
Mark