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

Engine Feature Comparison

asked 2015-11-13 15:17:20 -0600

updated 2016-01-09 15:27:50 -0600

Update: it would be good if this table (see Google Spreadsheet linked from @JulienMarrec answer) could be completed soon so that I can use it in a DOE document. Thanks everyone. Any DOE-2.1E (@JoeHuang) and HAP experts out there?

Also note 10 additional rows added to spreadsheet.

At the risk of starting a war on a Friday, I want to try to use UH to crowd-source some information about features available in various engines. For each of the following features, I am looking for one answer per engine and each answer to consist of a list of yes/no/partial/user-hook to the ability to model each of the following. Feel free to append to the list if you want to expand scope.

  1. Dedicated outdoor air systems (DOAS)
  2. Variable refrigerant flow (VRF) systems
  3. Variable-speed drives
  4. Commercial refrigeration
  5. Air-source heat-pump
  6. Water-source heat-pump
  7. Ground-source heat-pump
  8. Multi-function heat-pump
  9. Radiant heating
  10. Radiant cooling
  11. Evaporative cooling
  12. Absorption chiller
  13. Condensing boiler
  14. Heat recovery
  15. Ice storage
  16. Equipment cycling
  17. Optimal start/stop
  18. Static pressure reset
  19. User defined control
  20. Photo-voltaics
  21. Building integrated photo-voltaics
  22. Solar thermal
  23. Natural ventilation
  24. Displacement ventilation
  25. Underfloor air distribution
  26. Illuminance calculations (i.e., virtual sensors)
  27. Illuminance calculation-driven lighting control
  28. Illuminance calculation-driven shade/blind control
  29. Thermal mass
  30. Thermal bridges
  31. Phase-change-material
edit retag flag offensive close merge delete


Adding to the list: 32. Electrical Storage (e.g. electric battery model)
33. Complex Electric Power calculations (ie. magnitude and phase)

Ralph Muehleisen gravatar image Ralph Muehleisen  ( 2015-11-13 22:15:37 -0600 )edit

Is #33 really a BEM feature?

__AmirRoth__ gravatar image __AmirRoth__  ( 2015-11-14 10:39:27 -0600 )edit

It should be. Complex electrical building loads are needed for microgrid, distribution, and transmission network design and control. NERC just made a requirement that transmission operators needed new complex dynamic load models for buildings for better reliability.

Ralph Muehleisen gravatar image Ralph Muehleisen  ( 2015-11-14 10:59:25 -0600 )edit

I added columns for EnergyPro and Carrier HAP in the spreadsheet, if someone wants to go after those. Thanks to everyone who has contributed so far. This is valuable.

__AmirRoth__ gravatar image __AmirRoth__  ( 2015-11-16 10:41:32 -0600 )edit

Added additional rows based on my understanding of capabilities of E+ v8.4 and eQUEST v3.65 build 7163. Would be useful to have a consistent definition of Yes and No applied to this document, as many times a creative work-around can be used to abstract a physical phenomena into restrictive model inputs. In that spirit - does 'No' mean 'without a work-around requiring pre-processing or post-processing' to evaluate? And does "Yes' mean that the data model supports direct modeling of the physical phenomena..? That leaves the 3rd option partial/user hook even more unclear to me.

ChrisBalbach gravatar image ChrisBalbach  ( 2016-01-10 10:34:28 -0600 )edit

7 Answers

Sort by ยป oldest newest most voted

answered 2015-11-14 03:23:04 -0600

I think it's going to be messy trying to do this here.

I'd like to propose we try to fill a collaborative Google spreadsheet located here.

Feel free to add your answers, or even engines or features.

If you are disagreeing with a previous rating, please leave a comment.

Also, I think the "Partial / User-Hook" is a bit confusing, since with a stretch everything can be a "partial" if you're creative enough (and/or not very concerned about accuracy).

I would draw the limit as to being stuff you do only in the modeling software. Using outputs of the model to calculate heat recovery for example shouldn't be qualified as "partial". Neither should calculating the performance of a facade using another software (e.g. THERM) to create a derated U-value.

But using a "dummy zone" in eQuest to control the temperature of a DOAS is a "partial" to me.

edit flag offensive delete link more


OK. I just thought that the comment feature here could be used for discussion, while the answers are refined. But Google Docs works also. And there will be a pointer to it from here.

I intended "user-hook" to cover things like user-functions and EMS. I'm not exactly sure what I meant by "partial". Anyways, I agree with your refinement.

__AmirRoth__ gravatar image __AmirRoth__  ( 2015-11-14 08:44:41 -0600 )edit

@ __AmirRoth__ how do you want to handle daylighting stuff in this document? @rpg777 would probably say that E+ does a terrible job at the illuminance metrics, but gets the job done when paired w/ Radiance via OS. Should this be marked as a Yes or Partial? I'm really surprised that Trace can claim any sort of reasonable capability in that area.

Also, should IES VE be considered an engine per the original question?

ljbrackney gravatar image ljbrackney  ( 2015-11-16 09:13:41 -0600 )edit

answered 2015-11-16 10:59:35 -0600

updated 2015-11-16 13:41:37 -0600

If we're talking engines-only here, I believe there are a few columns on the list that do not belong there. I also agree with @Julien Marrec that this "Partial / User Hook" business is confusing; it's the basis for tools that are not purely engines already appearing on this list, and is the basis I'd use to lobby for other tools -- engines and interfaces alike -- to be added to it.

Many items on the feature list can only be evaluated in a BEM context with a multi-engine approach, which begs the question: what are we really trying to enumerate, here? Because we've got a (growing) list of... "things", that covers physics, mechatronics, physiology, psychology, economics, and voodoo, many of which don't lend themselves to a simple yes or no without some explanation.

Sorry to be the one to whiz in the Wheaties this fine Monday morning, but I'd just like to see this effort end up being useful (not saying that this is easy, either).

edit flag offensive delete link more


You forgot hermeneutics. Which rows would you delete from a comparison of BEM engines?

__AmirRoth__ gravatar image __AmirRoth__  ( 2015-11-16 12:19:32 -0600 )edit

No, I didn't, and I wouldn't delete any rows. I would delete a column or two if this is truly to be a comparison of just pure simulation engines. If it's inclusive of middleware, then I'd add OpenStudio, Honeybee, and SPOT Pro to the columns. And then we'd also have to add Radiance and AGi32 to the list, as engines capable of modeling daylighting, and with the help of the aforementioned middleware, checking half a dozen of the row's boxes, there.

rpg777 gravatar image rpg777  ( 2015-11-16 13:36:33 -0600 )edit

answered 2015-11-14 08:02:17 -0600

SEI_Associates gravatar image

I have filled in the Trane Trace column in Julien's Google spreadsheet.

edit flag offensive delete link more

answered 2015-11-16 10:03:03 -0600

crduggin gravatar image

I went ahead and put in answers for IESVE.

edit flag offensive delete link more

answered 2015-11-15 21:42:07 -0600

hongtz gravatar image

Others to consider:

  1. occupant behavior
  2. demand response
  3. operational faults
  4. dynamic glazing
  5. district cooling and heating systems
  6. CHP
  7. simplified and detailed air flow models
  8. non-vapor compression cooling technologies, e.g. thermoelectric
  9. other renewable, e.g. wind turbine
  10. DC power
edit flag offensive delete link more


@hongtz, thanks. Do 'occupant behavior' and 'demand response' fall under 'user-defined control'? Otherwise, what is the operational/accepted definition of these features?

__AmirRoth__ gravatar image __AmirRoth__  ( 2015-11-16 10:22:07 -0600 )edit

Yes both relate to interaction with or control of energy services in buildings in response to occupants needs or demand response signals. The 'user-defined control' feature can include them as long as it enables adequate coding to implement these types of controls.

hongtz gravatar image hongtz  ( 2015-11-17 11:27:47 -0600 )edit

answered 2015-11-22 19:57:44 -0600

updated 2015-11-22 22:15:49 -0600

eQuest claims the ability to model a waterside economizer but it can't without many work-arounds. I added this to the list.

edit flag offensive delete link more

answered 2015-11-16 14:31:32 -0600

Chris Jones gravatar image

I don't believe you should group DOE2.1e and EnergyPro together. EnergyPro includes a bunch of functions that enable modeling such things as VRV that DOE2.1e doesn't model directly.

edit flag offensive delete link more


EnergyPro also uses some other engines or calculations beyond DOE-2.1E, so it is better to list DOE-2.1E.

hongtz gravatar image hongtz  ( 2015-11-17 11:31:07 -0600 )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



Asked: 2015-11-13 15:17:20 -0600

Seen: 1,166 times

Last updated: Jan 09 '16