Dear all,
I am modeling PCM in heat transfer solids and fluids using COMSOL 5.6, but I am facing some problems with the solution in the initial time steps (within 20 seconds).
The domain temperature goes below the initial temperature of the domain.
I am using a time-dependent PARDISO solver and time-stepping BDF. I am attaching an image for your reference.
Your answer will be highly obliged.
Thanking you all
Could you be more explicit on the kind of boundarie conditions on the interface fin-PCM....?
Can you please check the boundary condition? I think some problem is coming from your boundary condition, especially at the interface between the PCM and fin, as can be seen, the fin temp is lower than the bulk of the regions. Have you checked with interfacial jump boundary conditions on the interface (such as Robin boundary)?
These are what I would try:
1. Check your mesh size, change it and see if you still see the issue.
2. Use a higher order discretization for temperature (quadratic, etc).
3. In your results, look for the regions that are below the initial temperature. They are usually at the singular points (i.e. corners). If this is the case, try to refine the mesh in those regions.
4. Sometimes, the time step that the software selects automatically is not small enough. Try to set it manually and see if the issue is resolved.
- Badges
- Science topic
- Similar topics
- Agricultural Science
- Soil Science
- Soil Analysis