CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   Simulating Heat Transfer in a Parabolic Trough concentrating Solar collector receiver (

Ash_Mech_Eng December 8, 2012 13:21

Simulating Heat Transfer in a Parabolic Trough concentrating Solar collector receiver
Hi all,
I am currently trying to simulate radiative heat transfer in a solar collector receiver using Fluent. The collector consists of multiple solid and fluid zones, the outer glass tube, the evacuated annulus region, an inner glass tube and the fluid medium through which the heat transfer fluid is circulated.

The heat transfer fluid inlet temperature is 50 degrees celsius and I am hoping to gain an outlet temperature of at least 250 degrees celsius.

I would like to apply a direct solar irradiance of 32000 W/m2 to the outer glass tube and allow the solver to compute the intensity attenuation throughout the domain to find the heat transfer due to this concentrated solar insolation.

Currently i am experiencing a number of problems

- Trying to represent an evacuated region using a created vacuum material causes an error in the AMG solver

- using small values for the density and near infite values for the specific heating capacity to resemble a vacuum results in all of the heat being dissipated in the vacuum and the inner tube and fluid remain at 50 degrees celsius

-Although i have applied direct radiation of 32000 W/m2 in the radiation tab on boundary conditions, its higher than this in regions within the domain although radiation should be scattered and absorbed by the glass as it travels through the domain

Is this simulation possible using this software? Is there anything fundamental that i may be missing? Can anyone help me?

If you require any further model description please let me know

Kind Regards

dsking December 10, 2012 11:02

I have also tried to use radiation models in fluent to no avail. This problem sounds similar to simulating a greenhouse, maybe you could search for greenhouse simulations with radiation models and get some hints from those.

Also you could try to model the vacuum as a semi-transparent solid with a very large Cp and small k. Sounds like a fun problem though, good luck.

All times are GMT -4. The time now is 03:02.