[bdgmm]

Dear all,

To conduct urban microclimate and building energy studies, I developed a coupled scheme between OpenFOAM and EnergyPlus. The OpenFOAM model is in charge of assessing outdoor conditions (i.e. air motion and temperature) around a specific building at the urban microscale (i.e. with in a range of 1 km). For this purpose, I am using an incompressible solver: the buoyantBoussinesqSimpleFoam.

Currently, I am trying to compare estimates of the outdoor temperature with measurements. I find that the outdoor temperature provided by the OpenFOAM solver underestimates the measurements. I tried to increase the resolution of my meshing, but the underestimation even becomes larger. Somehow, I suspect that the heat transfer through the air is underestimated by the OpenFOAM solver. For instance, when I use the function wallHeatTransferCoeff to calculate the convective heat transfer coefficients, I obtain values between 0.01 and 0.5 W/m2-K while they should be above 5 W/m2-K.

The OpenFOAM documentation does not say much how the heat transfer or convective heat transfer coefficients are calculated by the buoyantBoussinesqSimpleFoam and integrated into energy conservation equation. Does anyone know how the heat transfer or convective heat transfer coefficients are calculated by the buoyantBoussinesqSimpleFoam solver? If yes, what would you recommend to improve the estimates of outdoor temperature with OpenFOAM when coupled with a building energy model?

Regards,
Miguel

[kerim]

bdgmm


Dear bdgmm,


It is very interesting research. Would you like to share with your case files?
Kerim

[bdgmm]

Hi Kerim,

Thank you very much for your post.

Unfortunately, I am not allowed to share the case files. I was just wondering whether someone in the CFD community has already tried to validate the outdoor temperature with one of the OF solvers.

I recently tried to get in touch with the OF support, but the do not provide any documentation about how the solvers evaluate the heat transfer or the convective heat transfer coefficient. From what I could experience, they seem to be underestimated.

Did you experience something similar in your study? The OF support told me that the buoyantBoussinesqSimpleFoam solver has been replaced by the more general buoyantSimpleFoam solver which can used be with the Boussinesq equation of state. Do you know how to do that?

Regards,
Miguel