PyMAPDL Cheat Sheet

This is a quick-reference guide for PyMAPDL. You can jump to a section of your choosing by using these links:

Launching PyMAPDL

To launch a PyMAPDL instance locally and exit it

# To launch an instance
from ansys.mapdl.core import launch_mapdl

mapdl = launch_mapdl()

# To exit the instance
mapdl.exit()

To specify a job name, number of processors, and working directory

jname = 'user_jobname' 
path = '<path of directory>'
mapdl = launch_mapdl(nproc=2, run_location=path, jobname=jname)

To connect to an existing instance of MAPDL at IP 192.168.1.30 and port 50001.

mapdl=launch_mapdl(start_instance=False, ip='192.168.1.30', port=50001)

To create and exit a pool of instances

# To create a pool of 10 instances
from ansys.mapdl.core import LocalMapdlPool

pool = mapdl.LocalMapdlPool(10)
# To exit the pool 
pool.exit()

PyMAPDL Language

PyMAPDL commands are Python statements that act as a wrapper for APDL commands. For instance, ESEL, s, type, 1 is translated as:

mapdl.esel('s', 'type', vmin=1)

Commands that start with " * " or " / " have those characters removed.

mapdl.prep7()    # /PREP7 
mapdl.get()    # *GET

In cases where removing " * " or " / " will cause conflict with other commands, a prefix "slash" or "star" is added.

mapdl.solu()    # SOLU
mapdl.slashsolu()    # /SOLU 
mapdl.vget()    #VGET 
mapdl.starvget()    #*VGET

Converting an existing APDL script to PyMAPDL format

inputfile = 'ansys_inputfile.inp' 
pyscript = 'pyscript.py' 
mapdl.convert_script(inputfile, pyscript)

MAPDL Class

Load a table from Python to MAPDL

mapdl.load_table(name, array, var1='', var2='', var3='', csysid='')

To access from or write parameters to MAPDL database

# Save a parameter to a NumPy array nparray 
nparray = mapdl.parameters['displ_load']

# Create a parameter from a NumPy array nparray 
mapdl.parameters['exp_disp'] = nparray

To access information using GET and VGET directly to NumPy arrays

#Runs *GET command and returns a Python value
mapdl.get_value(entity='', entnum='',  item1='', it1num='', item2='', it2num='', **kwargs) 

#Runs *VGET command and returns a Python array 
mapdl.get_array(entity='', entnum='', item1='',it1num='', item2='', it2num='', kloop='', **kwargs)

Mech Class

Store the finite element mesh as a VTK UnstrucuredGrid data object.

grid = mapdl.mesh.grid

Save element & node numbers to Python arrays.

# Array of nodal coordinates 
nodes = mapdl.mesh.nodes 

# Save node numbers of selected nodes to array 
node_num = mapdl.mesh.nnum 

# Save node numbers of all nodes to array 
node_num_all = mapdl.mesh.nnum_all 

# Element numbs. of currently selected elements 
elem_num = mapdl.mesh.enum 

# All element numbers incl. those not selected 
elem_num_all = mapdl.mesh.enum_all

Post-Processing Class

This class is used for plotting and saving results to NumPy arrays.

mapdl.post1()
mapdl.set(1, 2) 

# Plot the nodal equivalent stress 
mapdl.post_processing.plot_nodal_eqv_stress() 

# Save nodal eqv. stresses to a Python array 
nod_eqv_stress = mapdl.post_processing.nodal_eqv_stress() 

# Plot contour legend using dictionary 
mapdl.allsel() 
sbar_kwargs = {
    "color": "black",
    "title": "EquivalentStress(psi)",
    "vertical": False,
    "n_labels": 6
} 

mapdl.post_processing.plot_nodal_eqv_stress(
    cpos='xy',
    background='white',
    edge_color='black',
    show_edges=True,
    scalarbar_args=sbar_kwargs,
    n_colors=9
)

Plotting Class

Plotting is interpolated with PyVista by saving the resulting stress and storing within the underlying UnstructureGrid

pl = pyvista.Plotter()
pl0 = mapdl.post_processing.plot_nodal_stress(
    return_plotter=True
)
pl.add(pl0.mesh)
pl.show()

# Plot the currently selected elements 
mapdl.eplot(show_node_numbering, vtk)

# Plot the selected volumes 
mapdl.vplot(nv1, nv2, ninc,degen,scale,...) 

# Display the selected areas 
mapdl.aplot(na1, na2, ninc, degen, scale,...) 
# Display the selected lines without 
# MAPDL plot symbols 
mapdl.lplot(vtk=True, cpos='xy', line_width=10) 

# Save png file of line plot with MAPDL 
#coordinate symbol \
mapdl.psymb('CS', 1) 
mapdl.lplot(vtk=False)