Tutorial¶
This tutorial is intended to give a short introduction on how to use the PyFMI package to load an FMU into Python and to simulate the given model.
For a more detailed description on how to use PyFMI, see the user’s documentation in JModelica.org
Loading an FMU into Python¶
Loading of an FMU is performed by simply importing the necessary object (load_fmu) from PyFMI. The object takes care of unzipping the model, loading the XML description and connecting the binary for use from Python.
# Import the load function (load_fmu)
from pyfmi import load_fmu
#Load the FMU
model = load_fmu('myFMU.fmu')
Note that this will either return an instance of a class consistent with the FMI for Model Exchange definition or for the Co-Simulation definition (and of the correct FMI version).
Simulating an FMU¶
Simulation of an FMU exported as Model Exchange requires that the additional package Assimulo is available. For Co-Simulation FMUs, no additional package is required as the solver is contained inside the FMU. A simulation is performed simply by using the simulate method:
#Simulate an FMU
res = model.simulate(final_time=10)
This will simulate the model from its default starting time 0.0 to 10.0 using default options. The result is returned and stored in res.
Information about the arguments to simulate is best viewed interactively from for example an IPython shell:
#View the documentation for simulate
model.simulate?
Options (for Model Exchange)¶
The options for an algorithm, which in our case is Assimulo, can be retrieved by calling the method simulate_options:
#Get the default options
opts = model.simulate_options()
This will return the default options for a simulation as a dictionary, for example the default solver is CVode from the Assimulo package. Changing the options specifically related to the solver is done by:
opts["CVode_options"]["atol"] = 1e-6 #Change the absolute tolerance
opts["CVode_options"]["discr"] = "Adams" #Change the discretization from BDF to Adams
All options related to the CVode solver can be found in the Assimulo documentation.
For changing general options such as the number of output points (number of communication points, ncp), they are accessed directly:
opts["ncp"] = 100 #One hundred output points
To use the options in an simulation, pass them in the call to the simulate method:
res = model.simulate(final_time=10, options=opts)
Options (for Co-Simulation)¶
The simulation options for a Co-Simulation FMU is retrieved and set consistent as for a Model Exchange FMU. The only difference is the actual options.
Result Object¶
The result object returned from a simulation contains all trajectories related to the variables in the model and are accessed as a dictionary.
res = model.simulate()
y = res['y'] #Return the result for the variable/parameter/constant y
dery = res['der(y)'] #Return the result for the variable/parameter/constant der(y)
This can be done for all the variables, parameters and constants defined in the model and is the preferred way of retrieving the result.