You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
As I explained in my latest issue #12386 Im trying to use the mapping app to initialize values from one mesh reading them from another one.
For the case of Stresses, they are stored in the elemental database in the origin model part. I was wondering if it is possible somehow to use the elements instead of the nodes for the mapping. From what I see in the code I think it is not.
I'd like to know if anyone has encountered this problem before and if they managed to solve it. Of course, one solution would be to send the data to the nodes doing some kind of volume based averaging, but I don't know if that might be too diffusive.
The other possibility would be to try to implement a new mapping myself. My first idea would be to use the nearest element using the element center location as reference.
Any suggestions or ideas are very welcome! @KratosMultiphysics/altair @philbucher @KratosMultiphysics/technical-committee @KratosMultiphysics/implementation-committee
The text was updated successfully, but these errors were encountered:
As explained in the readme, mapping on elements/conditions is not supported. In the past we could not agree on the interface (elemental vs GP values etc). In the end, there was not enough interest to really push for it
Long story short, only nodes can be used for mapping, i.e. you need to interpolate/extrapolate from/to elements or wherever your data lives
How badly do you need it? Ofc there is always a way but depend how much effort you want to put 🤔
As I explained in my latest issue #12386 Im trying to use the mapping app to initialize values from one mesh reading them from another one.
For the case of Stresses, they are stored in the elemental database in the origin model part. I was wondering if it is possible somehow to use the elements instead of the nodes for the mapping. From what I see in the code I think it is not.
I'd like to know if anyone has encountered this problem before and if they managed to solve it. Of course, one solution would be to send the data to the nodes doing some kind of volume based averaging, but I don't know if that might be too diffusive.
The other possibility would be to try to implement a new mapping myself. My first idea would be to use the nearest element using the element center location as reference.
Any suggestions or ideas are very welcome! @KratosMultiphysics/altair @philbucher @KratosMultiphysics/technical-committee @KratosMultiphysics/implementation-committee
The text was updated successfully, but these errors were encountered: