Question-and-Answer Resource for the Building Energy Modeling Community
Get s tarted with the Help page
Ask Your Question

Best Ground Temperature Variable for Conductive Heat Transfer Between Building and Ground

asked 2021-03-03 12:01:23 -0500

sashadf1 gravatar image

updated 2021-04-14 10:24:17 -0500


This is for an applied research project on Transactive Energy.

I am trying to create a simple box model of a building, with wall and roof surfaces touching outdoor ambient air, and the floor touching the ground.

For outdoor air temperature I am using the variable:

Output:Variable,*,Site Outdoor Air Drybulb Temperature,hourly; !- Zone Average [C]

The heat transfer between the building and the ground is represented with Q_ground = -sigma_ground(T_indoor-Tg), where T_indoor is the interior building temperature, Tg is the exterior ground temperature, and sigma_ground is the bulk conductivity between the building and the ground. If Q_ground is negative, heat flows from the building to the ground. If Q_ground is positive, heat flows from the ground to the building. (Conduction means heat flows from hot to cold).

Indoor air temp is approximated with the zone air temp of the living zone.

For Tg, I have a few different options: Output:Variable,,Site Ground Temperature,hourly; !- Zone Average [C] Output:Variable,,Site Surface Ground Temperature,hourly; !- Zone Average [C] Output:Variable,*,Site Deep Ground Temperature,hourly; !- Zone Average [C]

Which ground temperature is the best? I am assuming either Site Ground Temperature or Site Surface Ground Temperature would be appropriate.

Edit 4/14/2021: My models use Foundation Kiva to calculate ground heat transfer. All my floor and wall (bc crawl zone) contact surfaces with the ground have a boundary condition of "Foundation" and a boundary condition object of Foundation Kiva.

Because of this, neither of the 3 Output Variables listed above are actually used to calculate ground heat transfer; rather, this is done separately by a call to Kiva when EnergyPlus is running. As far as I can tell, Kiva models its own ground temperatures using soil data and data from the epw file.

edit retag flag offensive close merge delete



In reality, there is no single ground temperature for heat loss. Ground temperatures vary by depth below grade, vary from center of building outward, and are impacted by the presence of the building itself. You might read some of the information in this question.

shorowit gravatar image shorowit  ( 2021-03-03 16:38:17 -0500 )edit

For our purposes, which is a really simple model, we just need a proxy value, not a heat profile that is detailed horizontally and vertically.

sashadf1 gravatar image sashadf1  ( 2021-03-05 03:29:30 -0500 )edit

1 Answer

Sort by ยป oldest newest most voted

answered 2021-03-05 10:55:16 -0500

The first you mentioned if "ground" is your floor boundary condition. But these values simply will be what you put in monthly gound temperature object (so, these temperatures are your inputs).

This is the simpler way to model ground heat transfer, but really I do not know if this fit with your purposes. You may want to looking for some more detailed model

edit flag offensive delete link more


I found out that our models use Foundation Kiva as the default ground heat transfer model, which is more complicated than using a constant ground temperature to calculate heat flux from the ground into the building.

Foundation Kiva's algorithm varies ground temperature both vertically and horizontally around the foundation, and specifies core and perimeter temperatures for the ground as well as heat transfer through the ground. In short, it is much more complicated than just one single ground temperature.

sashadf1 gravatar image sashadf1  ( 2021-04-22 07:19:44 -0500 )edit

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer


Question Tools

1 follower


Asked: 2021-03-03 12:01:23 -0500

Seen: 192 times

Last updated: Apr 14