ResStock simulation: where to find number of indoor units (ie heads) of MSHP

asked 2020-10-17 17:25:09 -0600

sashadf1
1146 ●1 ●41 ●10

updated 2020-10-22 16:51:21 -0600

__AmirRoth__

4446 ●5 ●17 http://bleedinggreenna...

Hi,

Background: I am using OpenStudio v2.9.0 and EnergyPlus v9.2. I am modelling aggregate residential housing stock using NREL's ResStock v2.2.4.

I am currently trying to determine the number of heads (ie the number of indoor units) for ductless minisplit heat pump modelling, as this affects initial capital cost. In the parametric analysis tool, my input is typically Apply Upgrade = HVAC System Heat Pump|MSHP, SEER 15, 8.5 HSPF to specify a change from a propane furnace + AC baseline building to a building with a SEER 15 MSHP. Nowhere in the PAT GUI do I specify the number of indoor units, so I am assuming that ResStock and PAT are automatically choosing the number of units based off of a default algorithm that takes into account number of rooms, number of living and conditioned spaces, etc.

Is there anywhere in the ResStock output (ie the building characteristics JSON file, the E+ output file, the simulation log file, the results.csv file, or the annual timeseries file (which I save as a pickled dataframe) where I could find the integer number of indoor units for a MSHP simulation?

Thanks!

edit retag flag offensive close merge delete

add a comment

2

answered 2020-10-21 22:55:21 -0600

sashadf1
1146 ●1 ●41 ●10

We have decided to use the number of bedrooms from the ResStock output as a proxy for the number of indoor heads.

edit flag offensive delete link

Comments

That's a great idea. Why not number of bedrooms plus 1 for main living space though?

Eric Wilson ( 2020-10-23 17:10:50 -0600 )edit

I am following what a colleague of mine did, but I will bring this up in the next team meeting to see if we should do this instead.

sashadf1 ( 2020-10-23 17:24:51 -0600 )edit

Is four indoor heads too much to be realistic? Seems like a lot to me but I don't have any experience to back that up.

sashadf1 ( 2020-10-26 14:45:19 -0600 )edit

It really comes down to how the system was designed and if there is backup heat. I've seen some houses with one indoor head per floor, and some with one head per bedrooms, and even one apartment building with two heads in each living room because it was a historic building that couldn't be insulated and they didn't put backup electric resistance heat.

Eric Wilson ( 2020-10-26 14:55:38 -0600 )edit

add a comment

1

answered 2020-10-19 10:03:07 -0600

Eric Wilson

1853 ●12 ●8 https://resstock.nrel.gov/

The number of indoor heads or whether a multi-split or multiple 1 indoor to 1 outdoor units is not explicitly modeled.

If you want to calculate capital costs using discrete system sizes, you can manually calculate for each home using the reported cooling capacity field (reported heating capacity will refer to aux heat sizing) in the results.csv. However, that would require knowing the maximum heating capacity for the modeled MSHP at the design temperature for each home in your run, which is not straightforward to calculate (you can likely get pretty close with the NEEP spreadsheet).

--

This is a minor point, but since unit sizing (e.g., 3x 12 kBtu/h units vs. 4x 9 kBtu/h units does affect performance to some degree, the options in ResStock, because they are primarily based on data from real units, do have implied system sizes in terms of their performance. By looking at the options listed in BEopt, you can see that the SEER/HSPF levels for the "MSHP, SEER 14.5, 8.2 HSPF" and "MSHP, SEER 29.3, 14 HSPF" options both align with 12 kBtu/hr units (1 indoor to 1 outdoor).

image description

15 kBtu/hr units have slightly lower rated SEER/HSPF and 9 kBtu/hr units have slightly higher rated SEER/HSPF, however 15 kBtu/hr units may actually outperform 9 kBtu/hr units since they have higher efficiency at lower compressor speeds.