[bldg-sim] EnergyPlus - DOE

Griffith, Brent Brent_Griffith at nrel.gov
Thu Jan 19 10:55:43 PST 2006


Here is some more information on EnergyPlus simulation run times.  
 
I recently did a study with 5,374 EnergyPlus baseline models covering virtually every kind and size of commercial building out there (as reflected in 1999 CBECS public use data).  These are all full-blown, annual building models with 5-zone-per-floor thermal zoning, 15-minute timesteps, PSZ and VAV HVAC systems, etc, etc.  
 
Here is a summary of the run times I encountered:
 
Average:  4.53 minutes
Standard Deviation: 3.15 minutes
Minimum: 0.92 minutes
Maximum: 16.6 minutes
 
These were run on 3GHz, 64 bit AMD processors running Linux with a 64-bit-compile of Version 1.2.3.  
 
I have learned the hard way that it is entirely possible (and quite tempting) to create overly-complicated EnergyPlus models that will lead to excessive run times.  To speed things up, use Zone multipliers wherever practical and keep the number of windows and surfaces low.   I also set Minimum System Timestep to 5 minutes, and Maximum HVAC Iterations to 5. 
 
Brent Griffith, NREL
 


 -----Original Message-----
From: bldg-sim at gard.com [mailto:bldg-sim at gard.com]On Behalf Of Mahabir Bhandari
Sent: Wednesday, January 18, 2006 11:41 AM
To: bldg-sim at gard.com
Subject: [bldg-sim] EnergyPlus - DOE



Brian,

 

I think one additional difference from uses' point of view, which is not reported in this report, is simulation run time. It seems EnergyPlus takes 5-10 times longer to complete a whole year simulations than DOE2. I was running a simple 5 story building with VAV system and it took me more than 40 minutes for annual hourly simulations (1 hr time step) in EnergyPlus while eQuest could run the same building in less than 5 minutes. I am interested in hearing from others on the run time issue.

 

Mahabir Bhandari

 

 


  _____  


From: Brian Fountain [mailto:bfountain at greensim.com] 
Sent: Wednesday, January 18, 2006 12:12 PM
To: bldg-sim at gard.com
Subject: [bldg-sim] EnergyPlus - DOE

 

That is a big question.

 

For a comparison of current energy simulation software, you could look at Dru Crawley's paper comparing 20 tools:

http://www.energytoolsdirectory.gov/pdfs/contrasting_the_capabilities_of_building_energy_performance_simulation_programs_v1.0.pdf

 

Main differences between DOE and EnergyPlus:  

1)      DOE calculates the space loads for each hour then calculates the system and plant response for each hour.  EnergyPlus simultaneously solves the loads the system/plant response for each timestep.

2)      DOE has a fixed 1 hour timestep, EnergyPlus has a variable timestep (down to 1 minute?).

3)      There are also key issues with the weighting factor assumptions, the wall convective coefficients.

 

I haven't used EnergyPlus much yet.  What did I miss guys?

 

Brian Fountain

 

-----Original Message-----
From: bldg-sim at gard.com [mailto:bldg-sim at gard.com] On Behalf Of Renato Sousa
Sent: January 18, 2006 10:56 AM
To: bldg-sim at gard.com
Subject: [bldg-sim] EnergyPlus - DOE

 

Hello everybody.

 

I would like to know if anybody has experience on using both EnergyPlus and DOE.

Currently I'm using an interface for EnergyPlus. From the software description I couldn't find out the main differences between EnergyPlus and DOE. Does anyone know the main differences or where can I find them?

 


More information about the Bldg-sim mailing list