Plotting nodal displacements not working

[thiesblaak]

Dear all,

I'm trying to perform a static structural analysis in which I would like to plot the displaced results. On one of my models it works fine and I can choose at which time step I want to plot. My other model only works if I set show_displacedment to False in the plot_nodal_displacement() function. The error I recieve is the following:

ValueError Traceback (most recent call last)
Cell In[7], line 9
5 print(result)
8 mapdl.post1()
----> 9 result.plot_nodal_displacement(
10 result_set,
11 background = 'w',
12 cpos="xy",
13 displacement_factor=10,
14 show_displacement=True,
15 # show_edges=False,
16 # show_axes=False,
17 show_scalar_bar = False,
18 add_text=False,
19 color = 'red',
20 text_color="k",
21
22 )

File ~\anaconda3\envs\Ansys\Lib\site-packages\ansys\mapdl\reader\rst.py:860, in Result.plot_nodal_displacement(self, *args, **kwargs)
856 if self._is_thermal:
857 raise AttributeError(
858 "Thermal solution does not contain nodal " "displacement results"
859 )
--> 860 return self.plot_nodal_solution(*args, **kwargs)

File ~\anaconda3\envs\Ansys\Lib\site-packages\ansys\mapdl\reader\rst.py:840, in Result.plot_nodal_solution(self, rnum, comp, show_displacement, displacement_factor, node_components, element_components, sel_type_all, treat_nan_as_zero, **kwargs)
837 grid = self.grid
839 kwargs.setdefault("scalar_bar_args", {"title": f"{comp}\n{label}\n"})
--> 840 return self._plot_point_scalars(
841 scalars,
842 rnum=rnum,
843 grid=grid,
844 show_displacement=show_displacement,
845 displacement_factor=displacement_factor,
846 node_components=node_components,
847 element_components=element_components,
848 sel_type_all=sel_type_all,
849 treat_nan_as_zero=treat_nan_as_zero,
850 **kwargs,
851 )

File ~\anaconda3\envs\Ansys\Lib\site-packages\ansys\mapdl\reader\rst.py:3021, in Result._plot_point_scalars(self, scalars, rnum, grid, show_displacement, displacement_factor, add_text, animate, n_frames, overlay_wireframe, node_components, element_components, sel_type_all, movie_filename, treat_nan_as_zero, progress_bar, **kwargs)
3019 new_points[:, :-1] += disp
3020 else:
-> 3021 new_points = disp + grid.points
3022 grid = grid.copy()
3023 grid.points = new_points

ValueError: operands could not be broadcast together with shapes (711,3) (1335,3)

Has anyone ever experienced something similar before? Or does anyone maybe know what is going wrong?

With kind regards,
Thies

[thiesblaak]

To give some context, the code I'm using to call the plot_nodal_displacement is the following:

result = mapdl.result
result_set = 180 # Plotting the first results
disp_fact = 1e10

print(result)

mapdl.post1()
result.plot_nodal_displacement(
result_set,
background = 'w',
cpos="xy",
displacement_factor=10,
show_displacement=True,
# show_edges=False,
# show_axes=False,
show_scalar_bar = False,
add_text=False,
color = 'red',
text_color="k",

)

[germa89]

Hi @thiesblaak

I guess there is some elements there which are not compatible with PyMAPDL reader. Can you post a minimal working example of the model?

[thiesblaak]

Below youcan find the minimal working (not really) example for my analysis. To give a little context, I'm creating a surface, which I would like to turn around an axis using an MPC element.

mapdl.finish()  # some commands to clear the workspace
mapdl.clear()
mapdl.run("/OUTPUT")
mapdl.run("/GOPR")

# parameters
mapdl.run("E = 1.8e9")  # Young's modulus (Pa)
mapdl.run("v = 0.38")  # Poisson ratio (-)

mapdl.run("t = 0.4e-3")  # Original thickness Helicoid
mapdl.run("pi=acos(-1)") #Pi

# preprocessing

mapdl.prep7()  # enter preprocessor menu

# Define element type
mapdl.et(1, "SHELL181")  # element type 1
mapdl.et(2, "MPC184")    # Rigid element
mapdl.keyopt(1, 5, 2)    # Allows for initial curvature effects

mapdl.keyopt(2, 1, 1)  # Allow rigid link

# Create surface from geometry

mapdl.k(1,0,0,0)
mapdl.k(2,1,0,0)
mapdl.k(3,1,1,0)
mapdl.k(4,0,1,0)

mapdl.a(1, 2, 3, 4)


# Define shell geometry
mapdl.eshape()
mapdl.sectype(1, "SHELL")
mapdl.secdata("t", 1, "", 9)  # More integration points makes it more vulnerable to excessive distortions

mapdl.sectype(2, "beam", "RECT")
mapdl.secdata("t", "t")  # same for buckled beam


# Define material model
mapdl.mp("EX", 1, "E")  # set Young's modulus of mat 1
mapdl.mp("PRXY", 1, "v")  # set Poisson ratio of mat 1


mapdl.allsel("ALL")
# Mesh using smart sizing just in case
mapdl.type(1)
mapdl.smrtsize(6)  
mapdl.amesh("ALL")
mapdl.allsel("ALL")  # Make sure to have selected everything again

mapdl.type(2)  # select element type for mesh
mapdl.secnum(2)  # select section for mesh
mapdl.lsel("S", "LINE", "", 3)  # Select top line to remesh for rigid link
mapdl.smrtsize(1)
mapdl.lmesh("ALL")  # mesh all lines
mapdl.allsel("ALL")


mapdl.allsel("ALL")

mapdl.finish()  # exit pre-processor

mapdl.slashsolu()  # enter solution menu


# Settings
mapdl.nlgeom(1)  # Include large deflection effects
mapdl.neqit(300)  # Allow n substeps per iteration (good for non-linear analysis)
mapdl.nropt("FULL", "", "ON")
mapdl.nsubst(1000, 1000, 1)
mapdl.autots("ON")
mapdl.allsel("ALL")

mapdl.outres("ALL", "ALL")

# Constrain bottom, spine and top lines, and remove unnecessary ones
mapdl.dl(1, "", "ALL", 0)  # Bottom line

mapdl.dl(2, "", "ALL", 0)  # Symmetry axis
mapdl.dldele(2, "ROTY")
mapdl.dldele(2, "UY")

mapdl.dl(3, "", "ALL", 0)  # Top line
mapdl.dldele(3, "ROTY")
mapdl.dldele(3, "UX")
mapdl.dldele(3, "UY")
mapdl.dldele(3, "UZ")

# STEP 1: Twist
mapdl.time(1)
mapdl.dl(3, "", "ROTY", "pi/2")  # Rotate mechananism

# mapdl.f(105, "MY", -0.0005)  # Parametriseren voor mesh?
mapdl.allsel("ALL")
mapdl.solve()


mapdl.finish()

#plot nodal displacements
result = mapdl.result
result_set = 17  # Plotting the first results
disp_fact = 1e10

mapdl.post1()
result.plot_nodal_displacement(
    result_set,
    background = 'w',
    cpos="xy",
    displacement_factor=1,
    show_displacement=True,
    show_edges=False,
    # show_axes=False,
    show_scalar_bar = False,
    add_text=False,
    color = 'red',
    text_color="k",
    
)

[mikerife]

Hi @thiesblaak line 858 tells what happened; nodal displacements are being plotted for a thermal analysis. In which case there are no displacements. Mike

[thiesblaak]

Dear @mikerife, the strange thing is that I'm doing the exact same analysis for a different mechanism. In that analysis however I don't get the error.

[mikerife]

@thiesblaak Mmmm...odd. I'll have to treat this like going to court: I trust what you say but will need you to prove it. Can you supply a test model that shows the same behavior? Mike

[germa89]

Any update @thiesblaak ??