PyDPF - Select Surface nodes

TommasoMarella

Hi guys,

Is there an method to extract all the surface nodes from a rst file?

For the moment, I create a new component on mechanical with all the surface nodes, and I extract it during the postprocess. But I cannot always modify the model on mechanical.

Thanks,

Tommaso

Pierre Thieffry

Hi @TommasoMarella , you can look at the skin operator. It will allow you to extract the surface of any mesh https://dpf.docs.pyansys.com/api/ansys.dpf.core.operators.mesh.skin.html?highlight=skin#module-ansys.dpf.core.operators.mesh.skin

Mike.Thompson

@Pierre@Pierre

Do you have any tangible example of the skin operator. Ideally for DPF in Mechanical?

I am specifically wanting to use DPF to extract surfaces stresses. This would prevent users from needing to create a surface coating object (and APDL surface elements), and DPF would do the skin operation and post processing of new "skin" elements for EQV stress based on those DPF surface elements.

Would this process be any different than the stresses extracted from actual APDL surface elements generated from a Mechanical surface coating?

Pernelle Marone-Hitz

@Mike.Thompson This example uses the skin operator: https://discuss.ansys.com/discussion/2082/get-successive-elements-with-max-stress-in-depth-of-structure

Mike.Thompson

@Pernelle Marone-Hitz ,

Thanks, I see this uses the skin operator, but as far as I can tell, this is used to exclude elements that have a free surface in the iterative calculation. The stress values obtained are still for solid elements that exist in the APDL results file.

In my case, I would like to post process a stress value on the skin elements that do not exist in the APDL results file. I am not sure if this is possible, but I am trying to find out:

1. Can we extrapolate a stress value onto the skin elements created by DPF. This would be done by knowing the DOF solution for the nodes attached to the solid under the DPF skin element, and doing an integration on the skin elements all via DPF, no APDL.
2. If we can, are the values calculated exactly the same as what is done if we have surface elements calculating stress via APDL, or is it different.

Pernelle Marone-Hitz

@Mike.Thompson Indeed I was only answering the first part of the question. I haven't tried out the postprocessing part but indeed if it works it'd be great. Please keep me posted if you try it out!