Mechanical 2022R2 DPF VTK Export of Assembly & Plotting Parts.

This is possible by adding an identifier to the VTK Export such as the element material ID number. As an example we can use the Vise geometry that is included in the Ansys installation and which has 12 parts. There will be 36 contact pairs in the model for a total of 48 material IDs:

Using the Mechanical Scripting tool we can define a Field and fill it in with the internal element and material ID:

import mech_dpf
import Ans.DataProcessing as dpf

analysis = ExtAPI.DataModel.Project.Model.Analyses[0]
mech_dpf.setExtAPI(ExtAPI)

dataSource = dpf.DataSources(analysis.ResultFileName)

model =dpf.Model(dataSource)
mesh = model.Mesh

path = str("D:\\Temp\\")

e_field = dpf.FieldsFactory.CreateScalarField(12237, "Elemental")
for item in range(0,12237):
    a = [float(mesh.GetPropertyField("mat").Data[item])]
    b = mesh.ElementScoping.Ids[item]
    e_field.Add(b, a)

The VonMises stress will be gathered and it along with the 'e_field' can then be written to a VTK file:

vm_op = dpf.operators.result.stress_von_mises()
vm_op.inputs.data_sources.Connect(dataSource)
vm_fc = vm_op.outputs.fields_container.GetData()
vm_op.Run()

vtk_op = dpf.operators.serialization.vtk_export() # operator instantiation
vtk_op.inputs.file_path.Connect(path + "seqv.vtk")
vtk_op.inputs.fields1.Connect(vm_fc)
vtk_op.inputs.fields2.Connect(e_field)
vtk_op.Run()

Moving to an Python IDE like VSC we can read in the VTK and plot the result:

import pyvista as pv
import numpy as np

path = str("D:\\Temp\\")

part_1 = pv.read(path + 'seqv.vtk')

part_1.plot(scalars = "2_stress_1.s_eqv_(psi)")
print(part_1.array_names)

The list of arrays in the VTK file is:
['node_id', '2_stress_1.s_eqv_(psi)', 'element_id', '3_DimensionLess()']

'3_DimensionLess()' is the name of the array of material IDs.

The resulting plot:

The second part of the assembly is the SlidingJaw. To plot just it a numpy array of all the other material IDs can be created (here named ghosts), and the PyVista "remove cell" used to remove them from the plot:

ghosts = np.argwhere(part_1["3_DimensionLess()"] != 2.0 )
part_1 = part_1.remove_cells(ghosts)
print(part_1.array_names)
part_1.plot(scalars = "2_stress_1.s_eqv_(psi)")

This will add vtkGhostType array to our VTK object:

['node_id', '2_stress_1.s_eqv_(psi)', 'element_id', '3_DimensionLess()', 'vtkGhostType']

Then plotting just the single part show with the scalar bar (legend) updated: