pybamm-team#3611 use actual cell volume for average total heating (py…

…bamm-team#3707) * pybamm-team#3611 use actual cell volume for average total heating * pybamm-team#3611 changelog * pybamm-team#3611 account for number of electrode pairs * pybamm-team#3611 update variable names
js1tr3 · Aug 12, 2024 · 1d5c245 · 1d5c245
1 parent fe61b05
commit 1d5c245
Show file tree

Hide file tree

Showing 8 changed files with 606 additions and 521 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -1,8 +1,13 @@
 # [Unreleased](https://github.com/pybamm-team/PyBaMM/)
 
-## Bug Fixes
+## Bug fixes
 
 - Fixed a bug where if the first step(s) in a cycle are skipped then the cycle solution started from the model's initial conditions instead of from the last state of the previous cycle ([#3708](https://github.com/pybamm-team/PyBaMM/pull/3708))
+- Fixed a bug where the lumped thermal model conflates cell volume with electrode volume  ([#3707](https://github.com/pybamm-team/PyBaMM/pull/3707))
+
+## Breaking changes
+- The parameters `GeometricParameters.A_cooling` and `GeometricParameters.V_cell` are now automatically computed from the electrode heights, widths and thicknesses if the "cell geometry" option is "pouch" and from the parameters "Cell cooling surface area [m2]" and "Cell volume [m3]", respectively, otherwise. When using the lumped thermal model we recommend using the "arbitrary" cell geometry and specifying the parameters "Cell cooling surface area [m2]", "Cell volume [m3]" and "Total heat transfer coefficient [W.m-2.K-1]" directly. ([#3707](https://github.com/pybamm-team/PyBaMM/pull/3707))
+
 # [v24.1rc0](https://github.com/pybamm-team/PyBaMM/tree/v24.1rc0) - 2024-01-31
 
 ## Features

diff --git a/docs/source/examples/notebooks/models/jelly-roll-model.ipynb b/docs/source/examples/notebooks/models/jelly-roll-model.ipynb
diff --git a/docs/source/examples/notebooks/models/pouch-cell-model.ipynb b/docs/source/examples/notebooks/models/pouch-cell-model.ipynb
diff --git a/docs/source/examples/notebooks/models/thermal-models.ipynb b/docs/source/examples/notebooks/models/thermal-models.ipynb
diff --git a/examples/scripts/thermal_lithium_ion.py b/examples/scripts/thermal_lithium_ion.py
@@ -6,11 +6,15 @@
 pybamm.set_logging_level("INFO")
 
 # load models
+# for the full model we use the "x-full" thermal submodel, which means that we solve
+# the thermal model in the x-direction for a single-layer pouch cell
+# for the lumped model we use the "arbitrary" cell geometry, which means that we can
+# specify the surface area for cooling and total heat transfer coefficient
 full_thermal_model = pybamm.lithium_ion.SPMe(
     {"thermal": "x-full"}, name="full thermal model"
 )
 lumped_thermal_model = pybamm.lithium_ion.SPMe(
-    {"thermal": "lumped"}, name="lumped thermal model"
+    {"cell geometry": "arbitrary", "thermal": "lumped"}, name="lumped thermal model"
 )
 models = [full_thermal_model, lumped_thermal_model]
 
@@ -31,27 +35,43 @@
     }
 )
 # for the lumped model we set the "Total heat transfer coefficient [W.m-2.K-1]"
-# parameter as well as the "Cell cooling surface area [m2]" parameter. Since the "full"
-# model only accounts for cooling from the large surfaces of the pouch, we set the
-# "Surface area for cooling" parameter to the area of the large surfaces of the pouch,
-# and the total heat transfer coefficient to 5 W.m-2.K-1
+# parameter as well as the "Cell cooling surface area [m2]" and "Cell volume [m3]
+# parameters. Since the "full" model only accounts for cooling from the large surfaces
+# of the pouch, we set the "Surface area for cooling [m2]" parameter to the area of the
+# large surfaces of the pouch, and the total heat transfer coefficient to 5 W.m-2.K-1
 A = parameter_values["Electrode width [m]"] * parameter_values["Electrode height [m]"]
+contributing_layers = [
+    "Negative current collector",
+    "Negative electrode",
+    "Separator",
+    "Positive electrode",
+    "Positive current collector",
+]
+total_thickness = sum(
+    [parameter_values[layer + " thickness [m]"] for layer in contributing_layers]
+)
+electrode_volume = (
+    total_thickness
+    * parameter_values["Electrode height [m]"]
+    * parameter_values["Electrode width [m]"]
+)
 lumped_params = parameter_values.copy()
 lumped_params.update(
     {
         "Total heat transfer coefficient [W.m-2.K-1]": 5,
         "Cell cooling surface area [m2]": 2 * A,
+        "Cell volume [m3]": electrode_volume,
     }
 )
-params = [full_params, lumped_params]
+
 # loop over the models and solve
+params = [full_params, lumped_params]
 sols = []
 for model, param in zip(models, params):
     sim = pybamm.Simulation(model, parameter_values=param)
     sim.solve([0, 3600])
     sols.append(sim.solution)
 
-
 # plot
 output_variables = [
     "Voltage [V]",

diff --git a/pybamm/models/full_battery_models/base_battery_model.py b/pybamm/models/full_battery_models/base_battery_model.py
@@ -4,7 +4,6 @@
 
 import pybamm
 from functools import cached_property
-import warnings
 
 from pybamm.expression_tree.operations.serialise import Serialise
 
@@ -683,24 +682,6 @@ def __init__(self, extra_options):
                             f"Possible values are {self.possible_options[option]}"
                         )
 
-        # Issue a warning to let users know that the 'lumped' thermal option (or
-        # equivalently 'x-lumped' with 0D current collectors) now uses the total heat
-        # transfer coefficient, surface area for cooling, and cell volume parameters,
-        # regardless of the 'cell geometry option' chosen.
-        thermal_option = options["thermal"]
-        dimensionality_option = options["dimensionality"]
-        if thermal_option == "lumped" or (
-            thermal_option == "x-lumped" and dimensionality_option == 0
-        ):
-            message = (
-                f"The '{thermal_option}' thermal option with "
-                f"'dimensionality' {dimensionality_option} now uses the parameters "
-                "'Cell cooling surface area [m2]', 'Cell volume [m3]' and "
-                "'Total heat transfer coefficient [W.m-2.K-1]' to compute the cell "
-                "cooling term, regardless of the value of the the 'cell geometry' "
-                "option. Please update your parameters accordingly."
-            )
-            warnings.warn(message, pybamm.OptionWarning, stacklevel=2)
         super().__init__(options.items())
 
     @property

diff --git a/pybamm/models/submodels/thermal/base_thermal.py b/pybamm/models/submodels/thermal/base_thermal.py
@@ -179,32 +179,74 @@ def _get_standard_coupled_variables(self, variables):
         Q = Q_ohm + Q_rxn + Q_rev
 
         # Compute the X-average over the entire cell, including current collectors
+        # Note: this can still be a function of y and z for higher-dimensional pouch
+        # cell models
         Q_ohm_av = self._x_average(Q_ohm, Q_ohm_s_cn, Q_ohm_s_cp)
         Q_rxn_av = self._x_average(Q_rxn, 0, 0)
         Q_rev_av = self._x_average(Q_rev, 0, 0)
         Q_av = self._x_average(Q, Q_ohm_s_cn, Q_ohm_s_cp)
 
-        # Compute volume-averaged heat source terms
-        Q_ohm_vol_av = self._yz_average(Q_ohm_av)
-        Q_rxn_vol_av = self._yz_average(Q_rxn_av)
-        Q_rev_vol_av = self._yz_average(Q_rev_av)
-        Q_vol_av = self._yz_average(Q_av)
+        # Compute the integrated heat source per unit simulated electrode-pair area
+        # in W.m-2. Note: this can still be a function of y and z for
+        # higher-dimensional pouch cell models
+        Q_ohm_Wm2 = Q_ohm_av * param.L
+        Q_rxn_Wm2 = Q_rxn_av * param.L
+        Q_rev_Wm2 = Q_rev_av * param.L
+        Q_Wm2 = Q_av * param.L
+        # Now average over the electrode height and width
+        Q_ohm_Wm2_av = self._yz_average(Q_ohm_Wm2)
+        Q_rxn_Wm2_av = self._yz_average(Q_rxn_Wm2)
+        Q_rev_Wm2_av = self._yz_average(Q_rev_Wm2)
+        Q_Wm2_av = self._yz_average(Q_Wm2)
+
+        # Compute total heat source terms (in W) over the *entire cell volume*, not
+        # the product of electrode height * electrode width * electrode stack thickness
+        # Note: we multiply by the number of electrode pairs, since the Q_xx_Wm2_av
+        # variables are per electrode pair
+        n_elec = param.n_electrodes_parallel
+        A = param.L_y * param.L_z  # *modelled* electrode area
+        Q_ohm_W = Q_ohm_Wm2_av * n_elec * A
+        Q_rxn_W = Q_rxn_Wm2_av * n_elec * A
+        Q_rev_W = Q_rev_Wm2_av * n_elec * A
+        Q_W = Q_Wm2_av * n_elec * A
+
+        # Compute volume-averaged heat source terms over the *entire cell volume*, not
+        # the product of electrode height * electrode width * electrode stack thickness
+        V = param.V_cell  # *actual* cell volume
+        Q_ohm_vol_av = Q_ohm_W / V
+        Q_rxn_vol_av = Q_rxn_W / V
+        Q_rev_vol_av = Q_rev_W / V
+        Q_vol_av = Q_W / V
 
         variables.update(
             {
+                # Ohmic
                 "Ohmic heating [W.m-3]": Q_ohm,
                 "X-averaged Ohmic heating [W.m-3]": Q_ohm_av,
                 "Volume-averaged Ohmic heating [W.m-3]": Q_ohm_vol_av,
+                "Ohmic heating per unit electrode-pair area [W.m-2]": Q_ohm_Wm2,
+                "Ohmic heating [W]": Q_ohm_W,
+                # Irreversible
                 "Irreversible electrochemical heating [W.m-3]": Q_rxn,
                 "X-averaged irreversible electrochemical heating [W.m-3]": Q_rxn_av,
                 "Volume-averaged irreversible electrochemical heating "
                 + "[W.m-3]": Q_rxn_vol_av,
+                "Irreversible electrochemical heating per unit "
+                + "electrode-pair area [W.m-2]": Q_rxn_Wm2,
+                "Irreversible electrochemical heating [W]": Q_rxn_W,
+                # Reversible
                 "Reversible heating [W.m-3]": Q_rev,
                 "X-averaged reversible heating [W.m-3]": Q_rev_av,
                 "Volume-averaged reversible heating [W.m-3]": Q_rev_vol_av,
+                "Reversible heating per unit electrode-pair area " "[W.m-2]": Q_rev_Wm2,
+                "Reversible heating [W]": Q_rev_W,
+                # Total
                 "Total heating [W.m-3]": Q,
                 "X-averaged total heating [W.m-3]": Q_av,
                 "Volume-averaged total heating [W.m-3]": Q_vol_av,
+                "Total heating per unit electrode-pair area [W.m-2]": Q_Wm2,
+                "Total heating [W]": Q_W,
+                # Current collector
                 "Negative current collector Ohmic heating [W.m-3]": Q_ohm_s_cn,
                 "Positive current collector Ohmic heating [W.m-3]": Q_ohm_s_cp,
             }

diff --git a/pybamm/parameters/geometric_parameters.py b/pybamm/parameters/geometric_parameters.py
@@ -42,8 +42,19 @@ def _set_parameters(self):
         self.r_inner = pybamm.Parameter("Inner cell radius [m]")
         self.r_outer = pybamm.Parameter("Outer cell radius [m]")
         self.A_cc = self.L_y * self.L_z  # Current collector cross sectional area
-        self.A_cooling = pybamm.Parameter("Cell cooling surface area [m2]")
-        self.V_cell = pybamm.Parameter("Cell volume [m3]")
+
+        # Cell surface area and volume (for thermal models only)
+        cell_geometry = self.options.get("cell geometry", None)
+        if cell_geometry == "pouch":
+            # assuming a single-layer pouch cell for now, see
+            # https://github.com/pybamm-team/PyBaMM/issues/1777
+            self.A_cooling = 2 * (
+                self.L_y * self.L_z + self.L_z * self.L + self.L_y * self.L
+            )
+            self.V_cell = self.L_y * self.L_z * self.L
+        else:
+            self.A_cooling = pybamm.Parameter("Cell cooling surface area [m2]")
+            self.V_cell = pybamm.Parameter("Cell volume [m3]")
 
 
 class DomainGeometricParameters(BaseParameters):