Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page
Ask Your Question
2

Big difference in results between EnergyPlus versions

asked 2018-08-10 05:58:03 -0500

Tobias's avatar

updated 2018-08-13 14:21:50 -0500

I have a question regarding a big change in results from running the same .idf in EnergyPlus version 8.5 vs version 8.9. I noticed the difference when I had to update some studies that I have done previously since I couldn’t recreate the same results. I located the difference to be that I was now running version 8.9 instead of 8.5 previously. Here is an example file, run in version 8.5 and version 8.9, with the same idf: Files

Does anybody know what changed in the EnergyPlus versions that is causing this difference (I guess it has to do with the calculation methods for the objects in the idf file)?

I’m inclined to stick with EnergyPlus version 8.5 because when running a comparison with a different simulation tool (IDA ICE) there is very high correspondence of results, which there is not when comparing IDA ICE to version 8.9

* UPDATE *

Thank you very much for your input @mdahlhausen - here are some more details:

Note - Good catch that HVAC equipment sizing has changed, but since I'm modelleling and ideal system that keeps the temperature between 22 and 24 degC, in my mind the yearly energy consumption should be the same, do you agree?

+++++ Version 8.5 +++++

Cooling peak load (before sizing factor): 975.40212 W

Heating peak load (before sizing factor): 174.39179 W

Zone Ideal Loads Supply Air Total Cooling Energy (Yearly): 1086.916048 kWh

Zone Ideal Loads Supply Air Total Heating Energy (Yearly): 121.428125 kWh

Min Air Temperature: 22 degC

Max Air Temperature: 24 degC

+++++ Version 8.9 +++++

Cooling peak load (before sizing factor): 870.05914 W

Heating peak load (before sizing factor): 132.20619 W

Zone Ideal Loads Supply Air Total Cooling Energy (Yearly): 1016.466383 kWh

Zone Ideal Loads Supply Air Total Heating Energy (Yearly): 75.030207 kWh

Min Air Temperature: 22 degC

Max Air Temperature: 24 degC

+++++ Idf object +++++

HVACTemplate:Zone:IdealLoadsAirSystem,
Fraunhofer, !- Zone Name
Fraunhofer_HVAC,    !- Template Thermostat Name
,  !- Availability Schedule Name
40,  !- Heating Supply Air Temp {C}
,  !- Cooling Supply Air Temp {C}
0.008,  !- Max Heating Supply Air Humidity Ratio {kg-H2O/kg-air}
0.0085,  !- Min Cooling Supply Air Humidity Ratio {kg-H2O/kg-air}
,  !- Heating Limit
,  !- Maximum Heating Air Flow Rate {m3/s}
,  !- Maximum Sensible Heat Capacity
LimitFlowRate,  !- Cooling Limit
autosize,  !- Maximum Cooling Air Flow Rate {m3/s}
,  !- Maximum Total Cooling Capacity
,  !- Heating Availability Schedule
,  !- Cooling Availability Schedule
None,  !- Dehumidification Control Type
,  !- Cooling Sensible Heat Ratio
,  !- Dehumidification Setpoint
None,  !- Humidification Control Type
,  !- Humidification Setpoint
None,  !- Outdoor Air Method
,  !- Outdoor Air Flow Rate Per Person
,  !- Outdoor Air Flow Rate Per Floor Zone Area
,  !- Outdoor Air Flow Rate Per Zone
FraunhoferOutdoorAirCntrl,  !- Design Specification Outdoor Air Object Name
,  !- Demand Controlled Ventilation Type
DifferentialDryBulb,  !- Outdoor Air Economizer Type
,  !- Heat Recovery Type
,  !- Sensible Heat Recovery Effectiveness
;  !- Latent Heat Recovery Effectiveness
edit retag flag offensive close merge delete

Comments

can you add more detail to your post? What end-uses have changed, by how much, and what kind of HVAC system are you modeling? Is your system size the same? Meaning, is the HVAC equipment sizing different, or just the results? There have been quite a few bug fixes in EnergyPlus between 8.5 and 8.9 related to HVAC and sizing - If your old model is showing good agreement, that may be actually a result of overfitting elsewhere.

mdahlhausen's avatar mdahlhausen  ( 2018-08-10 10:40:08 -0500 )edit

run the simulation for the sizing period and export out load variables for your ideal loads objects and zones. That may point to what has changed. There have been several bug fixes related to sizing between 8.5 and 8.9. (original poster of this comment was @mdahlhausen )

Aaron Boranian's avatar Aaron Boranian  ( 2018-08-13 14:23:19 -0500 )edit

1 Answer

Sort by » oldest newest most voted
3

answered 2018-08-13 11:51:51 -0500

As noted by @mdahlhausen every release can introduce a change in results due to bug fixes or other improvements. Every local install of EnergyPlus includes a "changelog.html" file in the install root folder that lists all new features, defects repaired, and other changes for that release. The change lists are also posted on the EnergyPlus github release pages.

For a case like this, determine the specific release where the change appears by incrementing the file one version at time. Then consult the changelog for that release.

In this case, the file is quite simple, so I can guess that the change has to do with a bug fix related to BSDF (complex) fenestration. Stepping through the releases shows that the change occurred with version 8.8. Looking at the v8.8 changelog shows this related to BSDF:

6245 Improved Airflow Calculations for BSDF window types:

This version of EnergyPlus introduces improved version of airflow calculations for BSDF window type. It also include several bugfixes . . . Depending on shading construction and surface properties, results can be significantly different from previous version (especially in cases when front and back surface emissivity of shading layer is different.

Given that, I would trust the v8.9 results more than v8.5.

edit flag offensive delete link more

Comments

Thank you for the elaboration @MJWitte !

Tobias's avatar Tobias  ( 2018-08-14 01:43:13 -0500 )edit

Also, note more fixes coming in v9.0 for BSDF windows, see #6855 "BSDF Venetian Blinds - Add option for vertical blinds and correct surface emissivity calculation."

MJWitte's avatar MJWitte  ( 2018-08-14 08:28:31 -0500 )edit

Your Answer

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

Add Answer

Careers

Question Tools

1 follower

Stats

Asked: 2018-08-10 05:58:03 -0500

Seen: 783 times

Last updated: Aug 13 '18