Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Restructure "Creating Models" notebooks #3844

Open
brosaplanella opened this issue Feb 28, 2024 · 2 comments · May be fixed by #4184
Open

Restructure "Creating Models" notebooks #3844

brosaplanella opened this issue Feb 28, 2024 · 2 comments · May be fixed by #4184
Assignees
@brosaplanella

Comments

@brosaplanella
Copy link
Sponsor Member

After reviewing the "Getting Started" notebooks, the next step is to restructure the "Creating Models" ones. The idea is to have a cohesive set of notebooks that build progressively more complex models up to DFN (potentially with degradation). The inspiration is the "12 steps to Navier Stokes" tutorial.

We will not have 12 steps, at least initially, but so far the structure would be:

  1. Reservoir model (ODEs)
  2. SPM (parabolic PDEs)
  3. SPMe (parabolic PDEs in multiple domains, multiscale formulation appears)
  4. DFN (parabolic & elliptic PDEs in multiple domains)
  5. DFN + degradation (maybe SEI? Includes all of the above, but more complicated)

This is subject to change, as the notebooks are being edited/written (e.g. some might need to be split into two).

This issue would go hand-in-hand with this: #3839 (comment)
@brosaplanella brosaplanella self-assigned this Feb 28, 2024
@agriyakhetarpal agriyakhetarpal mentioned this issue Feb 28, 2024
Merged
@valentinsulzer
Copy link
Member

Do we want all the models to be battery models or should we start with generic models? Pro of battery models is it's more interesting, con is it requires loading in a lot of parameters

If we can have generic models, before 1 could be solving an exponential decay equation (introduce the concept of a variable, rhs, initial conditions, etc) and between 1 and 2 could be diffusion in a cartesian domain.

Other comments:

  • Full SPMe is tricky because it requires a complex expression for the voltage
  • It might be helpful to include some intermediate models between SPMe and DFN that introduce the coupling of potential as an algebraic equation

@valentinsulzer
Copy link
Member

Here are some models I wrote a while back, they don't work any more but might be useful intermediates to explain some of the model structure

import pybamm


class BaseModel(pybamm.lithium_ion.BaseModel):
    def __init__(self, name):
        super().__init__({}, name)
        self.variables = {
            "Time": pybamm.t,
            "x": pybamm.standard_spatial_vars.x,
            "x_n": pybamm.standard_spatial_vars.x_n,
            "x_p": pybamm.standard_spatial_vars.x_p,
            "r_n": pybamm.SpatialVariable("r_n", domain=["negative particle"]),
            "r_p": pybamm.SpatialVariable("r_p", domain=["positive particle"]),
        }

    def new_empty_copy(self):
        return pybamm.BaseModel.new_empty_copy(self)


def spm():
    model = pybamm.lithium_ion.SPM(name="SPM")
    model.variables.update(
        {
            "r_n": pybamm.SpatialVariable("r_n", domain=["negative particle"]),
            "r_p": pybamm.SpatialVariable("r_p", domain=["positive particle"]),
        }
    )
    variables = [
        "X-averaged negative particle concentration",
        "X-averaged positive particle concentration",
    ]
    return model, variables


def spm_no_r():
    model = pybamm.lithium_ion.SPM({"particle": "uniform profile"}, name="SPM_no_r")
    variables = [
        "Average negative particle concentration",
        "Average positive particle concentration",
    ]
    return model, variables


def spme():
    model = pybamm.lithium_ion.SPMe(name="SPMe")
    model.variables.update(
        {
            "r_n": pybamm.SpatialVariable("r_n", domain=["negative particle"]),
            "r_p": pybamm.SpatialVariable("r_p", domain=["positive particle"]),
        }
    )
    variables = [
        "X-averaged negative particle concentration",
        "X-averaged positive particle concentration",
        "Electrolyte concentration",
    ]
    return model, variables


def dfn_no_r():
    model = pybamm.lithium_ion.DFN({"particle": "uniform profile"}, name="DFN_no_r")
    variables = [
        "R-averaged negative particle concentration",
        "R-averaged positive particle concentration",
        "Electrolyte concentration",
        "Electrolyte potential",
        "Negative electrode potential",
        "Positive electrode potential",
    ]
    return model, variables


def dfn():
    model = pybamm.lithium_ion.DFN(name="DFN")
    variables = [
        "Negative particle concentration",
        "Positive particle concentration",
        "Electrolyte concentration",
        "Electrolyte potential",
        "Negative electrode potential",
        "Positive electrode potential",
    ]
    return model, variables


def reduced_c():
    class Model(BaseModel):
        def __init__(self, name="reduced_c"):
            super().__init__(name)
            param = self.param
            c_e = pybamm.Variable(
                "Electrolyte concentration",
                ["negative electrode", "separator", "positive electrode"],
            )

            a_j = pybamm.concatenation(
                pybamm.PrimaryBroadcast(1000 / param.n.L, "negative electrode"),
                pybamm.PrimaryBroadcast(0, "separator"),
                pybamm.PrimaryBroadcast(-1000 / param.p.L, "positive electrode"),
            )
            self.rhs[c_e] = pybamm.div(pybamm.grad(c_e)) + a_j

            self.initial_conditions[c_e] = pybamm.Scalar(1)
            self.boundary_conditions[c_e] = {
                "left": (pybamm.Scalar(0), "Neumann"),
                "right": (pybamm.Scalar(0), "Neumann"),
            }
            self.variables.update({"Electrolyte concentration": c_e})

    model = Model()
    variables = list(model.variables.keys())
    return model, variables


def reduced_c_phi():
    class Model(BaseModel):
        def __init__(self, name="reduced_c_phi"):
            super().__init__(name)
            param = self.param
            c_e = pybamm.Variable(
                "Electrolyte concentration",
                ["negative electrode", "separator", "positive electrode"],
            )
            phi_e = pybamm.Variable(
                "Electrolyte potential",
                ["negative electrode", "separator", "positive electrode"],
            )

            a_j = pybamm.concatenation(
                pybamm.PrimaryBroadcast(1 / param.n.L, "negative electrode"),
                pybamm.PrimaryBroadcast(0, "separator"),
                pybamm.PrimaryBroadcast(-1 / param.p.L, "positive electrode"),
            )
            self.rhs[c_e] = pybamm.div(pybamm.grad(c_e)) + a_j
            self.algebraic[phi_e] = (
                pybamm.div(pybamm.grad(c_e) / c_e - pybamm.grad(phi_e)) - a_j
            )

            self.initial_conditions[c_e] = pybamm.Scalar(1)
            self.initial_conditions[phi_e] = pybamm.Scalar(0)
            self.boundary_conditions[c_e] = {
                "left": (pybamm.Scalar(0), "Neumann"),
                "right": (pybamm.Scalar(0), "Neumann"),
            }
            self.boundary_conditions[phi_e] = {
                "left": (pybamm.Scalar(0), "Dirichlet"),
                "right": (pybamm.Scalar(0), "Neumann"),
            }
            self.variables.update(
                {"Electrolyte concentration": c_e, "Electrolyte potential": phi_e}
            )

    model = Model()
    variables = list(model.variables.keys())
    return model, variables


def reduced_c_phi_j():
    class Model(BaseModel):
        def __init__(self, name="reduced_c_phi_j"):
            super().__init__(name)
            param = self.param
            c_e_n = pybamm.Variable(
                "Negative electrolyte concentration", "negative electrode"
            )
            c_e_s = pybamm.Variable("Separator electrolyte concentration", "separator")
            c_e_p = pybamm.Variable(
                "Positive electrolyte concentration", "positive electrode"
            )
            c_e = pybamm.concatenation(c_e_n, c_e_s, c_e_p)

            phi_e_n = pybamm.Variable(
                "Negative electrolyte potential", "negative electrode"
            )
            phi_e_s = pybamm.Variable("Separator electrolyte potential", "separator")
            phi_e_p = pybamm.Variable(
                "Positive electrolyte potential", "positive electrode"
            )
            phi_e = pybamm.concatenation(phi_e_n, phi_e_s, phi_e_p)

            phi_s_n = pybamm.Variable(
                "Negative electrode potential", "negative electrode"
            )
            phi_s_p = pybamm.Variable(
                "Positive electrode potential", "positive electrode"
            )

            a_j_n = c_e_n**0.5 * pybamm.sinh(phi_s_n - phi_e_n + 1)
            a_j_p = c_e_p**0.5 * pybamm.sinh(phi_s_p - phi_e_p - 4)
            a_j = pybamm.concatenation(
                a_j_n, pybamm.PrimaryBroadcast(0, "separator"), a_j_p
            )

            self.rhs[c_e] = pybamm.div(pybamm.grad(c_e)) + a_j
            self.algebraic[phi_e] = (
                pybamm.div(pybamm.grad(c_e) / c_e - pybamm.grad(phi_e)) - a_j
            )
            self.algebraic[phi_s_n] = pybamm.div(pybamm.grad(phi_s_n)) + a_j_n
            self.algebraic[phi_s_p] = pybamm.div(pybamm.grad(phi_s_p)) + a_j_p

            self.initial_conditions[c_e] = pybamm.Scalar(1)
            self.initial_conditions[phi_e] = pybamm.Scalar(1)
            self.initial_conditions[phi_s_n] = pybamm.Scalar(0)
            self.initial_conditions[phi_s_p] = pybamm.Scalar(4)

            self.boundary_conditions[c_e] = {
                "left": (pybamm.Scalar(0), "Neumann"),
                "right": (pybamm.Scalar(0), "Neumann"),
            }
            self.boundary_conditions[phi_e] = {
                "left": (pybamm.Scalar(0), "Neumann"),
                "right": (pybamm.Scalar(0), "Neumann"),
            }
            self.boundary_conditions[phi_s_n] = {
                "left": (pybamm.Scalar(0), "Dirichlet"),
                "right": (pybamm.Scalar(0), "Neumann"),
            }
            self.boundary_conditions[phi_s_p] = {
                "left": (pybamm.Scalar(0), "Neumann"),
                "right": (pybamm.Scalar(1), "Neumann"),
            }

            self.variables.update(
                {
                    "Electrolyte concentration": c_e,
                    "Electrolyte potential": phi_e,
                    "Negative electrode potential": phi_s_n,
                    "Positive electrode potential": phi_s_p,
                    "Interfacial current density": a_j,
                }
            )

    model = Model()
    variables = list(model.variables.keys())
    return model, variables


def solve_plot(model, variables):
    sim = pybamm.Simulation(model)
    sim.solve([0, 3600])
    sim.plot(variables)


# solve_plot(*spm())
# solve_plot(*spm_no_r())
# solve_plot(*reduced_c())
# solve_plot(*spme())
# solve_plot(*reduced_c_phi())
# solve_plot(*reduced_c_phi_j())
# solve_plot(*dfn_no_r())
# solve_plot(*dfn())

@kratman kratman mentioned this issue Mar 5, 2024
Merged
brosaplanella added a commit that referenced this issue Jun 16, 2024
@brosaplanella
#3844 fix indentation docstring
c74b5e5
brosaplanella added a commit that referenced this issue Jun 16, 2024
@brosaplanella
#3844 initial improvements notebook 1
2c144fe
brosaplanella added a commit that referenced this issue Jun 16, 2024
@brosaplanella
#3844 refactor first notebook: ODE model
7fee8d4
brosaplanella added a commit that referenced this issue Jun 16, 2024
@brosaplanella
#3844 create placeholder for second notebook
17ace12
brosaplanella added a commit that referenced this issue Jun 16, 2024
@brosaplanella
#3844 move old notebooks to another file
2d84dac
@brosaplanella brosaplanella linked a pull request Jun 16, 2024 that will close this issue
Draft
8 tasks
brosaplanella added a commit that referenced this issue Jun 16, 2024
@brosaplanella
#3844 add title to empty notebook so tests don't crash
4c2223c
brosaplanella added a commit that referenced this issue Jun 16, 2024
@brosaplanella
#3844 fix broken link
636ffa8
brosaplanella added a commit that referenced this issue Jun 23, 2024
@brosaplanella
#3844 add basic reservoir model
3f7c65d
brosaplanella added a commit that referenced this issue Jun 23, 2024
@brosaplanella
#3844 fix index & crossref
cfe2a7c
brosaplanella added a commit that referenced this issue Jun 23, 2024
@brosaplanella
#3844 fix a few typos
ba10921
brosaplanella added a commit that referenced this issue Jun 23, 2024
@brosaplanella
#3844 write PDE model notebook
a07983e
brosaplanella added a commit that referenced this issue Jun 23, 2024
@brosaplanella
#3844 initialise model into a class notebook
94eca88
brosaplanella added a commit that referenced this issue Jun 26, 2024
@brosaplanella
#3844 start notebook 3
df9230d
brosaplanella added a commit that referenced this issue Jun 26, 2024
@brosaplanella
#3844 make small changes to make consistent with 3
ed68b44
brosaplanella added a commit that referenced this issue Jun 26, 2024
@brosaplanella
#3844 write models into classes notebook
d8a9359
brosaplanella added a commit that referenced this issue Jun 27, 2024
@brosaplanella
#3844 remove old notebooks
0350804
brosaplanella added a commit that referenced this issue Jun 27, 2024
@brosaplanella
#3844 fix docs
ae96e8a
@brosaplanella brosaplanella mentioned this issue Jul 17, 2024
Open
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@brosaplanella brosaplanella

Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

[WIP] Improve "Creating Models" series pybamm-team/PyBaMM
2 participants
@valentinsulzer @brosaplanella