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

I've been working with the project_singlefamilydetached directory, and I'm wondering exactly how to determine the heating load of a home.

My objective is to compare heating with a GSHP to a propane furnace. Some context:

1. I'm running two simulations: one with the GSHP upgrade ("HVAC System Heat Pump|GSHP") and one with the propane upgrade ("HVAC System Propane Heating|Propane Furnace, AFUE 60%"). Other than the two changes in the Apply Upgrade field, every other parameter is the same.
2. The location is fixed to be MI_Alpena.County.Rgnl.AP.726390
3. I'm using a constant seed value of 10 in the Algorithm Settings

After running the two simulations, I download the respective timeseries.csv files, and compare outputs of buildings with the same building_id that has had the respective upgrades applied. What I've found so far:

1. Converting the values in electricity_heating_kwh from the GSHP upgrade scenario to kJ, and converting the values in propane_heating_mbtu from the propane upgrade scenario to kJ, do NOT equate each other. For the GSHP, I multiplied the values by the COP (3.6), and for the Propane, I multiplied the values by the AFUE (0.6)--after converting both to kJ, of course.
2. I compared values in the "ZONE PREDICTED SENSIBLE LOAD TO HEATING SETPOINT HEAT TRANSFER RATE (LIVING ZONE) [W]" and "ZONE PREDICTED SENSIBLE LOAD TO SETPOINT HEAT TRANSFER RATE (LIVING ZONE) [W]" columns for both upgrade scenarios, and found that they do NOT equate each other. Even after converting the negative values to equal 0, since I ONLY wish to determine heating load, the values still don't equate each other. The two column labels I mentioned were obtained by referencing the eplusout.rdd file.

I'm unsure what to do at this point. Is there something I'm not accounting for? Am I not looking at the right variables? If I'm comparing two buildings, with all variables remaining constant except for the HVAC upgrade, shouldn't there be something that lets me know the heating load of the home, independent of the type of the heating system?

edit retag close merge delete

Sort by » oldest newest most voted

Besides what answered by Eric which is very helpful, I'll suggest also running an extra simulation with IdealLoadsAirSystems to eliminate any impact brought by different hvac systems (Mostly I'm lean to trusting this number to be loads as my understanding).

The predicted sensible load variables you're looking at are still dependent on hourly zone temperature, so if your hvac systems are not conditioning zones into exactly the same temperatures every hour, the numbers might vary. So if in your case even annually aggregated load numbers are very much varying, you might want to check hourly zone air temperatures for both systems to see how well the setpoint is maintained.

Hope this helps.

more

How would I simulate with IdealLoadsAirSystems using Parametric Analysis Tool + ResStock directory? There is no "IdealLoadsAirSystems" variable in the eplusout.rdd file, and neither is there one in the options.tsv file either.

Thank you for your response by the way, very much appreciated.

( 2019-10-22 20:19:58 -0500 )edit

There's no easy way to do this through PAT/ResStock. You'd have to take the OpenStudio/EnergyPlus models, update them outside PAT/ResStock, and then re-run the simulations manually.

( 2019-10-23 09:36:32 -0500 )edit

@y.tanaka It won't be hard if you're using OpenStudio app to create your PAT seed model, there's an option of "Turn On Ideal Air Loads" under "Thermal Zones", I recommend to make a copy of your model, remove all your other hvac components and check that for all the conditioned zones. Hopefully it will run successfully and give you the objective load numbers.

( 2019-10-24 17:33:51 -0500 )edit

Method #1 won't work because the GSHP COP changes with indoor and outdoor conditions.

Method #2 could work, but there are nuances depending on HVAC sizing and setpoints. You could compare coil heat output variables, though there are also nuances there (e.g., duct losses). I think that's what we did to calculate the BEopt "delivered energy" outputs.

more

Thank you for your insight Eric. I tried comparing coil heat output variables, and there are still times when the values between the two differ dramatically. Values that differ from each other by >5% occurs in approximately 2% of the dataset; sometimes, GSHP will have a heating value whereas Propane has a heating value of 0, and vice-versa.

Is there a way to fix the COP of the GSHP, regardless of indoor/outdoor conditions? Also, what do you think about observing all heat losses&gains instead?

( 2019-10-23 14:04:40 -0500 )edit

It sounds like you just want the hourly heating load for all homes. You could run all homes with electric baseboard heat, which would remove the effects of fans, pumps, and duct losses. This would perfectly condition all heated basements, which would give a higher heating load than typical operation.

( 2019-10-24 15:32:17 -0500 )edit

How would I compare electric baseboard heat with the propane furnace upgrade? Wouldn't I just run into the same problems that I have with the GSHP upgrade?

Some further context, I'm trying to model a dual-fuel, GSHP+propane furnace home, using a GSHP that outputs 18 kBtu/hr. Times when heating demand is >18 kBtu, propane furnace kicks in to supplement heating. Since there's no option for that in ResStock, I'm attempting to do this manually. Are you saying that I can take the electricity_heating_kwh of the baseboard scenario and that's the objective heating load?

( 2019-10-24 16:44:26 -0500 )edit