Problem with a model in OMPython

[kmerbcn]

Hello

I have a strange problem. Perhaps I changed something in the original code, but now my modelica model is not able to simulate using the Python interface. The error is as follows :

Traceback (most recent call last):

Cell In[6], line 1
mod_ee.simulate()

File ~\AppData\Roaming\Python\Python312\site-packages\OMPython\ModelicaSystem.py:1121 in simulate
returncode = om_cmd.run()

File ~\AppData\Roaming\Python\Python312\site-packages\OMPython\ModelicaSystem.py:306 in run
raise ModelicaSystemError(f"Error running command {repr(cmdl)}") from ex

ModelicaSystemError: Error running command ['C:/Users/jgarcia/AppData/Local/Temp/tmp1up5r2n9/D_class_amplifier.exe', '-r=C:/Users/jgarcia/AppData/Local/Temp/tmp1up5r2n9/D_class_amplifier_res.mat']

Before yesterday it was working. So I am a bit puzzled. Can someone tell me what was happening?

Br

[kmerbcn]

Code is as follows . Problem comes with mod_ee.simulate() . It is very strange because same code was working before yesterday and modelica models are working in OMEdit perfectly.

-- coding: utf-8 --

"""
Created on Thu Apr 23 11:14:23 2026

@author: jgarcia
"""

from math import pi, sqrt
import numpy as np
import matplotlib.pyplot as plt
from OMPython import ModelicaSystem
from OMPython import OMCSessionZMQ

if name == 'main':

plt.close('all')

omc = OMCSessionZMQ()

mod_ss = ModelicaSystem("C:/Users/jgarcia/Documents/Pruebas_modelica/Prueba_Estabilidad_Buck.mo",
                      "Prueba_Estabilidad_Buck", 
                      ["Modelica",
                       "C:/Users/jgarcia/Documents/Pruebas_modelica/Buck_Conv_SS.mo",
                       "C:/Users/jgarcia/Documents/Pruebas_modelica/Amplitude_Estimation_SS.mo"],
                      commandLineOptions=["-d=initialization"])

mod_ee = ModelicaSystem("C:/Users/jgarcia/Documents/Pruebas_modelica/D_class_amplifier.mo",
                        "D_class_amplifier",
                        ["Modelica",
                         "C:/Users/jgarcia/Documents/Pruebas_modelica/SMC.mo",
                         "C:/Users/jgarcia/Documents/Pruebas_modelica/Amplitude_Estimation.mo"],
                        commandLineOptions=["-d=initialization"])

fi = 0.0099345882657961
A = 0.12566
mod_ss.setParameters("A = " + str(A))
mod_ss.setParameters("fi = " + str(fi))
mod_ss.setParameters("Vg = 311")
mod_ss.simulate()
out_variables = mod_ss.getSolutions(['time', 'sine.y', 'buck_Conv_SS.y'])

Ro, Li, Co, E = mod_ss.getParameters(['Ro', 'Li', 'Co', 'Vg'])
Ro = float(Ro)
Li = float(Li)
Co = float(Co)
E = float(E)

t_ss = out_variables[0,:]*sqrt(Li*Co)
#il = E*out_variables[2,:]*sqrt(Co/Li)
il_ss = out_variables[2,:]

plt.figure(1)
plt.plot(t_ss, il_ss, color='blue', label='State space')

# A = 100
# fi_ee = 1e6
# print('Frecuencia electrica : ' + str(fi_ee/1e6) + ' [MHz]')
# mod_ee.setParameters("A = " + str(A))
# mod_ee.setParameters("fi = " + str(fi_ee))
# mod_ee.setParameters("Vg = 311")
# mod_ee.simulate()
# out_variables_ee = mod_ee.getSolutions(['time', 'currentSensor.i', 'sine1.y'])

# t_ee = out_variables_ee[0,:]
# il_ee = out_variables_ee[1,:]

# plt.plot(t_ee, il_ee, color='red', label='Electrical model')
plt.grid()
plt.legend()
plt.show()

[adeas31]

Which OMPython version are you using? Run pip show OMPython.

The code you provided doesn't call mod_ee.simulate(). Please post the minimalistic code that can be used to reproduce the error.

[kmerbcn]

Name: OMPython
Version: 4.0.0

You are right this script is not calling mod_ee.simulate() because It was commented. As soon as I remove the comment, the error appears in that line basically in the terminal of SPYDER i do the following statements

mod_ee = ModelicaSystem("C:/Users/jgarcia/Documents/Pruebas_modelica/D_class_amplifier.mo",
"D_class_amplifier",
["Modelica",
"C:/Users/jgarcia/Documents/Pruebas_modelica/SMC.mo",
"C:/Users/jgarcia/Documents/Pruebas_modelica/Amplitude_Estimation.mo"],
commandLineOptions=["-d=initialization"])

mod_ee.simulate()

and it appears the error above

Cell In[6], line 1
mod_ee.simulate()

File ~\AppData\Roaming\Python\Python312\site-packages\OMPython\ModelicaSystem.py:1121 in simulate
returncode = om_cmd.run()

File ~\AppData\Roaming\Python\Python312\site-packages\OMPython\ModelicaSystem.py:306 in run
raise ModelicaSystemError(f"Error running command {repr(cmdl)}") from ex

ModelicaSystemError: Error running command ['C:/Users/jgarcia/AppData/Local/Temp/tmp1up5r2n9/D_class_amplifier.exe', '-r=C:/Users/jgarcia/AppData/Local/Temp/tmp1up5r2n9/D_class_amplifier_res.mat']

I cloned the OMPython from github and I didnt see where the error appears. I also checked the temporary folder tmp1up5r2n9D and I do not see the result file .mat. Could it be that? . On the other hand if you see the long code mod_ss.simulate() works

[kmerbcn]

Hello Adeel

I can send the modelica files and the python script if you wish. I am wondering whether it can be my system .

[kmerbcn]

Debugging the code, the error appears here

I have no idea why, because the model works without problems in OMEdit. Very strange.

I would also like to add that in the temporary folder I see the parameters correctly overriden, the xml, a proper json.

[kmerbcn]

It seems the simulation set-up of the omedit is not transferred to the python and the simulation breaks. :)
I will try to add in python same config.
For some reason, tolerance in omedit is 1e-4. I will check

[kmerbcn]

This line was correct : mod_ee.simulate(simflags='-noRootFinding')

That was the problem. Thanks for the mental support :)

[adeas31]

So adding mod_ee.simulate(simflags='-noRootFinding') solved the problem?

[kmerbcn]

Yes i debugged the code and tried to analyze the excepctions inside of OMPython. Later on, I saw in getSolutions that the solver tried to compute something until the simulation failed. It seems in omedit the restart after event was removed and in OMPython not.
This is something that numerically puzzled me. I had a conversation with @casella concerning about non-linear control systems such as the one i am simulating. In slidding systems, I add some fixed delays to avoid chattering issues. I see slightly differences when the amount of delay is added. I suppose the theory behind it has to do with delay differential equations (DDE). Once, @casella sent a very good paper of him dealing with delays and mechanical oscillators. I do not know if somebody knows more about these topics and references to have a look at it. :)

Best regards and thanks for the support. I hope we can find a way to collaborate with the SW.

Javier Martinez

[kmerbcn]

Hello Adeel

I have another question while computing in OMPYthon. I set up a short simulation to read the variables. The screenshot is as follows.
As you can see I set up a simulation with interval 5 and stoptime 350. After reading the structure, I see in the right circle, a variable with several thousdands of time steps. Why? In theory i should have a vector with 70 time steps approximately.

Is there a way to fix it?