pybammeis is not compatible with grouped parameter models.

[oarcelus]

I am trying to run impedance simulations with the grouped parameter model SPMe provided in PyBOP. Here a example:

import pybamm
import pybop
import numpy as np
import matplotlib.pyplot as plt
import pybammeis

model = pybop.models.lithium_ion.GroupedSPMe()

parameter_values = model.default_parameter_values
parameter_values.update({"Initial SoC": 0.5})
eis_sim = pybammeis.EISSimulation(model, parameter_values=parameter_values)
freq = np.logspace(-4, 4, 30)
eis_sim.solve(freq)
eis_sim.nyquist_plot()

I have an error saying that KeyError: "'Total current density [A.m-2]' not found. Best matches are ['Current [A]']". This comes from the following code in EISSimulation class

symbol_replacement_map = {
      new_model.variables[name]: variable
      for name, variable in external_circuit_variables.items()
}

Where external circuit FunctionControl is exposing a current density variable, while the GroupedSPMe class does not. I see that PyBOP has its own EIS simulator. I see that there is a slightly different code for symbol_replacement_map. Where you explicitly check if external circuit variables are exposed to the model.

# Define the variables to replace
symbol_replacement_map = {}
for name, variable in external_circuit_variables.items():
    if name in model.variables.keys():
        symbol_replacement_map[model.variables[name]] = variable

Is there any other changes that EIS simulator does in PyBOP to handle these grouped parameter models?

[NicolaCourtier]

Hi @oarcelus

PyBOP's EISSimulator implements the same numerical computation as pybamm-eis, as described by Hallemans et al. (2025) in Appendix B. We have generalised the code to a broader class of models, as you have identified, since our grouped models use only current and not current density as a variable.