Is it possible to solve transient anisotropic heat transfer equation using py-pde?

[louis10643]

Hi,

Is it possible to solve transient anisotropic heat transfer equation using py-pde?

where kx, ky, kz are all different.

I tried to use diffusion module but its D only takes constant as parameter.

[david-zwicker]

This is definitely possible, albeit not with the simple DiffusionPDE class. Instead, you might want to use the powerful PDE class, similar to this example.

[david-zwicker]

A few comments:

• Instead of using gradient_x you can simply write d_dx and you don't need to register the operator beforehand.
• You can use constant fields in the expressions. You simply need to pass them using the consts argument of the PDE class.

[louis10643]

Thank you for your quick reply again! I used d_dx, d_dy, d_dz earlier but it shows errors. This is my PDE:

eq=PDE({"T": d_dx(0.00001d_dx(T))+d_dy(0.00001d_dy(T))+d_dz(0.0000005*d_dz(T))+C"}, bc={"value": 0})

[david-zwicker]

You need multiplication between numbers and operators in python/sympy.

[louis10643]

I see, thank you!

[david-zwicker]

If this doesn't work, it would be good to open a bug report.