Use FElupe inside MATLAB

[adtzlr]

This code snippet demonstrates the usage of FElupe inside MATLAB. First, ensure to have a version of Python installed which is supported by MATLAB.

pyenv

should give something like

  PythonEnvironment with properties:

          Version: "3.10"
       Executable: "C:\...\python-3.10.8.amd64\python.EXE"
          Library: "C:\...\python-3.10.8.amd64\python310.dll"
             Home: "C:\...\python-3.10.8.amd64"
           Status: Loaded
    ExecutionMode: InProcess
        ProcessID: "4200"
      ProcessName: "MATLAB"

While most function calls are straight-forward, the ramp-argument of the Step-definition needs special attention. As MATLAB doesn't support a dict with objects as keys, the ramp-argument has to be created by py.dict.fromkeys(). Also, integers have to be created explicitely, i.e. int8(3).

% import FElupe
fem = py.importlib.import_module('felupe');

% create a hexahedron-region on a cube
mesh = fem.Cube(n=11);
region = fem.RegionHexahedron(mesh);

% add a field container (with a vector-valued displacement field)
field = fem.FieldContainer({fem.Field(region, dim=int8(3))});

% apply a uniaxial elongation on the cube
loadcase = fem.dof.uniaxial(field, clamped=true);
boundaries = loadcase{1};

% define the constitutive material behaviour 
% and create a nearly-incompressible (u,p,J - formulation) solid body
umat = fem.NeoHooke(mu=1);
solid = fem.SolidBodyNearlyIncompressible(umat, field, bulk=5000);

% prepare a step with substeps
move = fem.math.linsteps({0, 2}, num=int8(10));
step = fem.Step( ...
    items={solid}, ...
    ramp=py.dict.fromkeys({boundaries{'move'}}, move), ...
    boundaries=boundaries ...
);

% add the step to a job, evaluate all substeps and create a plot
job = fem.CharacteristicCurve(steps={step}, boundary=boundaries{'move'});
job.evaluate(filename='result.xdmf');

fig = job.plot( ...
    xlabel='Displacement $u$ in mm $\longrightarrow$', ...
    ylabel='Normal Force $F$ in N $\longrightarrow$'...
);
fig{1}.show()