SOC dependent parameters combined with particle size distribution model

[nicschn93]

Hi all,

I want to extend and investigate my DFN model with the particle size distribution model from Kirk et al. 2021.
For my initial model I use SOC dependent parameters, so the parameter set is comparable to the ORegan2022 dataset.
When I try to solve the model I get the error "UnboundLocalError: local variable 'matrix' referenced before assignment".

A small example to reproduce the error is below. When I replace the "ORegan2022" dataset with the "Marquis2019" dataset the simulation works. If I use the MPM approach the model also works.

Can you tell me, what I have to do to use the MP-DFN model with concentration dependent parameters?

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

options = {"particle size": "distribution"}
model = pybamm.lithium_ion.DFN(options=options, name="MP-DFN")

param = pybamm.ParameterValues("ORegan2022")
param = pybamm.get_size_distribution_parameters(param)

solver = pybamm.CasadiSolver(mode="fast with events")

var_pts = {"x_n": 20, "x_s": 15, "x_p": 20, "r_n": 20, "r_p": 20, "R_n":10, "R_p":10}

submesh_types = model.default_submesh_types
submesh_types["negative particle"] = pybamm.MeshGenerator(
    pybamm.Exponential1DSubMesh, submesh_params={"side": "right"}
)
submesh_types["positive particle"] = pybamm.MeshGenerator(
    pybamm.Exponential1DSubMesh, submesh_params={"side": "right"}
)

sim = pybamm.Simulation(
    model,
    parameter_values=param,
    C_rate=1,
    solver=solver,
    var_pts=var_pts,
    submesh_types=submesh_types,
)

sim.solve([0, 10])

Thanks in advance and best regards
Nicolas

[nicschn93]

@tobykirk @valentinsulzer you also discussed/answered in similar questions about the psd-topic (like #2066) .
Can you help me with this issue?

[nicschn93]

Fixed in v25.1.0 refer to #4726 .