SolverError when starting solution

[Johannes-hub22]

Hi guys,

I am trying to simulate battery degradation using the O'Kane dataset. My initial idea is to replicate Figure 1 of the recent O'Kane et al publication (2022). I often get a SolveError when simulating an experimental cycle using initial (starting) solution from the prev. solutions. This is not dependent on the parameter value (e.g. Dsol for SEI) as I have run into this on numerous simulations / investigations.

Here is the error:
SolverError: Could not find acceptable solution: solver terminated successfully, but maximum solution error (0.19144467801435194) above tolerance (1e-06)

I do not want to disable the eSOH functionality as I would like to investigate SOH etc.

Is there a workaround for this?

[valentinsulzer]

Please post the full code that causes the error

[Johannes-hub22]

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

pybamm.set_logging_level("NOTICE")

exp_cycles = 2

save_exp = 1

option={"SEI":"solvent-diffusion limited"}

model = pybamm.lithium_ion.DFN(options=option)

param = pybamm.ParameterValues("OKane2022")

param.update({"Outer SEI solvent diffusivity [m2.s-1]": 7.5E-21})

exp_ini = pybamm.Experiment(
[(
"Hold at 4.2V until 50mA ",
"Rest for 4 hours "
)],
period="30 seconds"
)

exp_char = pybamm.Experiment(
[
(
"Discharge at 0.1C until 2.5 V",
"Charge at 0.3C until 4.2 V",
"Hold at 4.2 V until 50 mA",
)
],
period="30 seconds"
)

exp_cycling = pybamm.Experiment(
[
(
"Discharge at 1C until 2.5 V",
"Hold at 2.5 V until 50mA",
"Rest for 300 seconds",
"Charge at 1C until 4.2 V",
"Hold at 4.2 V until 50 mA",
"Rest for 300 seconds"
)
] * exp_cycles,
termination="20% capacity" # SOH_C of the battery
)

var_pts = {
"x_n": 50, # negative electrode
"x_s": 40, # separator
"x_p": 30, # positive electrode (not interested in this for SEI growth)
"r_n": 120, # negative particle
"r_p": 40, # positive particle (not interested in this for SEI growth)
}

exp = exp_ini
sim_ini = pybamm.Simulation(model, experiment=exp, parameter_values=param, var_pts= var_pts, )
sol_ini = sim_ini.solve()

exp = exp_char
sim = pybamm.Simulation(model=model, experiment=exp, parameter_values=param, var_pts= var_pts)
sol_k = sim.solve(starting_solution=sol_ini, save_at_cycles=save_exp)

exp = exp_cycling
sim = pybamm.Simulation(model=model, experiment=exp, parameter_values=param, var_pts= var_pts)
sol = sim.solve(starting_solution=sol_k, save_at_cycles=save_exp)

[valentinsulzer]

btw you can use three backticks to format your code, starting with python3 and ending with

import pybamm
model = ...

[Johannes-hub22]

Thanks. I am still a beginner and operating on Spyder IDE :)

[Johannes-hub22]

@tinosulzer: One more observation... It seems like the simulation is sensitive when updating parameters.

When I update the "Dead lithium decay constant [s-1]" to the default value of 4E-07, it throws out the same error.
i.e.
param.update({"Dead lithium decay constant [s-1]": 4E-07})

[valentinsulzer]

Should be fixed by #2355