Assign a spacial temperature vector as initial temperatrue field for DFN model

[xyyrly]

Description

Hi team,
I try to set an initial temperature distribution in x dimension for DFN but now it looks only a scalar value can be set for the whole domain. Could it be fixed? Thanks!

Motivation

No response

Possible Implementation

No response

Additional context

No response

[tommaull]

I'll take this one

[xyyrly]

Thank you very much!

[tommaull]

xyyrly - assuming you were using the xfull temperature model?

[xyyrly]

Hi tommaull,

I am using a new thermal model to calculate electrolyte temperature along x dimension and particle temperature along r dimension, kind of like DFN model.

[tommaull]

Can you write out the steps to replicate with existing models?

[xyyrly]

Oh sorry I was confused. I think here what I hope to realize is to assign a spatical temperature distribution to DFN model without subthermal model. Is it possible?

[tommaull]

I think a thermal model is required. @tinosulzer - what are your thoughts? My approach was going to be to add something like this to xfull:

def set_initial_conditions(self, variables):
T = variables["Cell temperature [K]"]

    if isinstance(self.param.T, (list, np.ndarray)):
        if len(self.param.T) == 3:
            # Get sizes of negative, separator, and positive domains
            sizes = {
                "negative electrode": self.mesh["negative electrode"].npts,
                "separator": self.mesh["separator"].npts,
                "positive electrode": self.mesh["positive electrode"].npts,
            }

            # Check if sizes match the length of T
            if all(size == len(T) for size in sizes.values()):
                self.initial_conditions = {
                    "negative electrode": self.param.T[
                        : sizes["negative electrode"]
                    ],
                    "separator": self.param.T[
                        sizes["negative electrode"] : sizes["negative electrode"]
                        + sizes["separator"]
                    ],
                    "positive electrode": self.param.T[
                        sizes["negative electrode"] + sizes["separator"] :
                    ],
                }
            else:
                raise ValueError("T vector size does not match domain sizes")
        else:
            raise ValueError("T should be a vector of length 3")
    else:
        self.initial_conditions = {T: np.full_like(T, self.param.T)}

[rtimms]

@tommaull take a look at e.g. how we define initial concentration, which can depend on x and r, and then how it is used in the model.

You'll need to make initial temperature a FunctionParameter, then you can pass T(x) as an initial condition.

[tommaull]

Put in a draft PR to show progress but having an issue I don't understand in the expression tree.

set T_init as a function in the x domain
#3151