How can I get the nodal result for a single material if the node is shared between two materials?

I have a model with two materials and shared topology. So the nodes between the two bodies are shared.

Node 13 is a corner node between the two geometries.

Due to material props, you can get different strains at the node so Mechanical returns two values for a equivalent strain at node 13.

**My goal is to get the value for equivalent strain at node 13 for material 1 (or 2). **

I tried to use a named selection w/ boolean (material - convert to nodes - filter to node 13 and scope to that name selection but that gives the same results as node 13 direct).

The result I want is best demonstrated by picking a non shared topo node where a single value is returned because there is only one material.

The op = dpf.operators.scoping.split_on_property_type() should allow me to scope to a material (by number) but I don't have it working yet.

first attempts with dpf which get me the averaged nodal result:

def post_started(sender, analysis):# Do not edit this line
    define_dpf_workflow(analysis)

# Uncomment this function to enable retrieving results from the table/chart
# def table_retrieve_result(value):# Do not edit this line
    # import mech_dpf
    # import Ans.DataProcessing as dpf
    # wf = dpf.Workflow(this.WorkflowId)
    # wf.Connect('contour_selector', value)
    # this.Evaluate()

def define_dpf_workflow(analysis):
    import mech_dpf
    import Ans.DataProcessing as dpf

    node = 13
    my_data_sources = dpf.DataSources(analysis.ResultFileName)
    my_time_scoping = dpf.Scoping()
    my_time_scoping.Ids = [1] # the first set

    my_scoping = dpf.Scoping()
    my_scoping.Location = "Nodal" #optional
    my_scoping.Ids = [node]

    st_eqv_op = dpf.operators.result.elastic_strain_eqv()
    st_eqv_op.inputs.requested_location.Connect('Nodal')
    st_eqv_op.inputs.data_sources.Connect(my_data_sources)
    st_eqv_op.inputs.time_scoping.Connect(my_time_scoping)

    st_eqv_op.inputs.mesh_scoping.Connect(my_scoping)

    #mesh_op = dpf.operators.mesh.mesh_provider() 
    #mesh_op.inputs.data_sources.Connect(my_data_sources) 
    #mesh = mesh_op.outputs.mesh.GetData() 

    op = dpf.operators.scoping.split_on_property_type() # operator instantiation
    #op.inputs.mesh_scoping.Connect(st_eqv_op)# optional
    #op.inputs.mesh.Connect(mesh)
    #op.inputs.requested_location.Connect(13)

    dpf_workflow = dpf.Workflow()
    dpf_workflow.Add(st_eqv_op)
    dpf_workflow.SetOutputContour(st_eqv_op)
    dpf_workflow.Record('wf_id', True)
    this.WorkflowId = dpf_workflow.GetRecordedId()

@Vishnu @Pierre Thieffry @Ayush Kumar

Find more posts tagged with

Accepted answers

Ayush Kumar

Below is an example based on the idea from @Mike.Thompson

import Ans.DataProcessing as dpf

rst_path = Model.Analyses[0].ResultFileName

nid = 134

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

mesh_split = dpf.operators.mesh.split_mesh(mesh=mesh, property="mat")
meshes = mesh_split.outputs.meshes.GetData()

for index in range(meshes.Length):
    mat_id = meshes.GetLabelSpace(index)["mat"]
    ms = dpf.operators.scoping.from_mesh(mesh=meshes.Get(index), requested_location="Elemental")
    ms_f = ms.outputs.scoping.GetData()

    sx = model.CreateOperator("SX")
    sx.inputs.requested_location.Connect("Nodal")
    sx.inputs.mesh_scoping.Connect(ms_f)
    sx_f = sx.outputs.fields_container.GetData()[0]

    s_nid = sx_f.GetEntityDataById(nid)[0]
    print("Material ID %s Averaged Stress for Node ID %s: %s" % (mat_id, nid, s_nid ))

model.ReleaseStreams()

All comments

Mike.Thompson

@M, What you need is to use an elemental scoping for only the elements of interest (in your case mat 1 or 2). You can get this in a number of ways. This will mean the results operator getting the data will only pull information from those nodes and ignore the others. You can still have requested location = nodal, and the operator will do the element nodal to nodal averaging, but only per the elements in the elemental scoping.

I hope this makes sense the difference between the mesh scoping being elemental and the requested location being nodal.

Mike.Thompson

FYI, this is also very helpful when you have things like mixed dimensional modeling. I first ran into this when I wanted to extract stress from a surface coating and the nodes were attached to the surface shells as well as the solids. The data being extracted was decomposed into two types, one for each. I made an elemental mesh scoping for only the shell coatings and all was fixed to get only the results per the coating elements as desired.

Ayush Kumar

Below is an example based on the idea from @Mike.Thompson

import Ans.DataProcessing as dpf

rst_path = Model.Analyses[0].ResultFileName

nid = 134

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

mesh_split = dpf.operators.mesh.split_mesh(mesh=mesh, property="mat")
meshes = mesh_split.outputs.meshes.GetData()

for index in range(meshes.Length):
    mat_id = meshes.GetLabelSpace(index)["mat"]
    ms = dpf.operators.scoping.from_mesh(mesh=meshes.Get(index), requested_location="Elemental")
    ms_f = ms.outputs.scoping.GetData()

    sx = model.CreateOperator("SX")
    sx.inputs.requested_location.Connect("Nodal")
    sx.inputs.mesh_scoping.Connect(ms_f)
    sx_f = sx.outputs.fields_container.GetData()[0]

    s_nid = sx_f.GetEntityDataById(nid)[0]
    print("Material ID %s Averaged Stress for Node ID %s: %s" % (mat_id, nid, s_nid ))

model.ReleaseStreams()