Discussion Forum

Hello,

the problem I'm currently trying to solve consists of a coaxial cable (cylinder) arrangement buried in soil, and I'm hoping to get a good estimate of the temperatures of the inner and outer conductor. Only the inner conductor generates thermal losses, the insulation material between the inner and outer conductor is mainly gas, therefore I have to take convective heat transfer into account.
Since the surrounding soil can be seen as a big thermal capacitance, the system wont' be thermally stable beneath 300 days. Solving for Non-Isothermal Flow (predefined coupling of heat transfer and laminar flow) then results in very long simulation times.

What would be a good way to optimize my problem?
I'm primarily interested in the steady state result. Is it possible to tune the simulation to achieve a convergent solution with a stationary solver for convection problems?
Another option might be an adjustment of the time stepping: the absolute change in the fluid velocity between to consecutive days is very small. Is there a way to just solve for the heat transfer physic for a day, then recalculate the coupled physics, then back to step one and so on...?
(Additional information: the flow is generally not very fast, < 0.15 m/s.)
Since the difference seems to be small, the transported heat energy shouldn't depend on correct flow velocity values for every second.

Thanks in advance!

Regards
Maximilian