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

Importance of ground heat transfer for building simulation

I read in various articles that the correct ground heat transfer is very important for building simulation. At the same time many simulation programmes use simplified algorithm because multidimension heat transfer is computationally intensive. Can anybody give me some reasons why it is so important and which simulation engines do a good job? Is this issues also relevant for small buildings with an slab area of less than 12x12m? For which kind of buildings it is particularly important (residential/comercial/etc.)?

edit retag close merge delete

Sort by ยป oldest newest most voted

The importance of ground heat transfer has a lot to do with the size of the building relative to the ground transfer area/perimeter. I've done a lot of work with high rise buildings in Chicago and I can honestly state that ignoring ground heat transfer is justified in most cases because for buildings more than 10 or so stories, the ground heat transfer is such a small fraction of the overall building energy use (well under 5%) that it can be ignored - especially when uncertainties in all the other energy uses is considered.

For smaller commercial buildings and residential buildings, ground heat transfer can be a far more important part of the building heat transfer, especially if the basement or foundation are not well insulated. If they are well insulated (meaning good details to avoid any thermal bridges), then, unless the building is very efficient, it is far less important and can be ignored for rough building estimates.

If the building is short and the ground connection is not well insulated, then you really do need to worry about ground heat transfer.

So, here is my checklist:

1. Is it a high rise? Unless you are designing a VERY high efficiency building, you can ignore it.

2. Is it a low rise but the ground connection is well insulated? You can ignore it for rough calculations of normal efficiency buildings but should include it for high efficiency buildings or more accurate calcs

3. Is it a low rise building without good basement/foundation insulation? You should try to include it.

Because the importance is generally fairly low, I think that using a "closest equivalent" design approach is usually close enough. By that I mean, pick out the closest equivalent to the actual slab/basement/foundation construction that the software you are using might include and use that rather than trying to get the exact details.

One problem with uninsulated basements is you really need the ground temp and not just the outside air temps. EnergyPlus has utility tools to estimate that from the weather files, but adding the ground heat transfer to the models is not particularly easy or streamlined. For insulated slab on grade floors, heat transfer is related to the perimeter of the basement/foundation and the difference between interior and exterior air temps.

One way to "fake" ground heat transfer is by adding an extra exterior wall that has the right overall U value and area such that the UAdelta T is equal to the FP delta T that you actually get from the perimeter heat transfer. You can look up F values from the ASHRAE handbook and if you know the perimeter, create a wall of unit height (1m or 1 ft depending on your units) around the perimeter and model as a simple U value that matches the F value for your foundation design. This will give a decent approximation to the heat transfer you would expect from a more complete analysis and ...

more

The importance of ground heat transfer will depend on the relative efficiency of the rest of your building and how many surfaces are in contact with the ground.

ASHRAE perceives that enough heat is lost through foundations to warrant insulation levels for slabs and below-grade walls in Standard 90.1. In my understanding (correct me if I'm wrong), few engineers actually simulate the foundations of buildings to claim energy savings for higher levels of insulation. Two reasons for this:

1. People assume that foundation heat loss is negligible (without providing much evidence for the claim), and therefore higher levels of insulation won't significantly impact energy savings.
2. As you said, the multi-dimensional calculations are too computationally intensive to include in standard energy analyses.

People claiming that ground heat transfer is important are usually (like me) trying to justify the time they spend studying the problem, while those who think it's negligible are looking for an excuse to avoid the calculations. But the truth is, without the tools capable of performing these calculations, we have a hard time estimating how important the problem actually is.

There are few tools which actually simulate the ground domain coupled to the rest of the building simulation. Take a look at this question on modeling slabs for more information.

As to your question about when ground heat transfer is important: Foundation heat loss is stronger near the slab perimeter, so the problem becomes more important for smaller slabs.

more

Sorry to disagree with you Neal but I do not believe that all those of us who think ground heat transfer are trying to avoid the calculations. I, for one, really like differential equations and would love and opportunity to apply neat methods like conformal mapping to solve the problems but with some problem types, such as high-rise buildings, the fraction of heat transfer through basements/floors is just too small to worry about.

( 2014-11-24 17:02:48 -0500 )edit

"with some problem types, such as high-rise buildings, the fraction of heat transfer through basements/floors is just too small to worry about." [citation needed] : )

So then the foundations should go uninsulated? Just because it's a small fraction of the overall energy use doesn't mean that there are not cost-effective ways to save energy by increasing the insulation level.

My point is that no one ever does the analysis to say one way or the other and people just fall back on 90.1. If you can track down a reference for the analysis that went into the 90.1 F/C factors, let me know.

( 2014-11-24 17:44:31 -0500 )edit

I'll see what I can track down for you Neal. I know that Moncef Krarti of U. of Colorado has several papers in ASHRAE transactions and in Energy and Buildings that you can find in Google Scholar (but which won't give you access to the papers). My first introduction was in Kreider's book "Heating and Cooling of Buildings". I don't have access to it right now, but I'm sure there will be some references to the original work in there.

( 2014-11-24 22:39:25 -0500 )edit