Who is this for: Users of Autodesk Building Performance Analysis tools who want to know how changes to the DOE-2.2 simulation engine is impacting the energy results in the new release of Green Building Studio.
Takeaway: The February 14, 2014 release of Green Building Studio (GBS) web service now uses the newest version of DOE-2.2-48r, which includes Autodesk code enhancements. Revisions to the DOE-2.2 software result in some changes to the GBS energy analysis results (thereby affecting Revit and Vasari energy analysis results).
Autodesk and Intel teamed with JJHirsch and Associates to embark on making DOE-2.2 and eQuest more BIM friendly. This included speed enhancements, better shading polygon support, and increased limits for several objects. The current release of eQuest and DOE-2.2 includes these enhancements and are available at DOE2.com.
Autodesk Green Building Studio (GBS) uses the DOE-2.2 simulation engine for calculating the hourly, whole building energy usage of your project. The DOE-2.2 software was developed by James J. Hirsch & Associates (JJH) in collaboration with Lawrence Berkeley National Laboratory (LBNL). DOE-2.2 has been validated by Lawrence Berkeley National Laboratory and the Los Alamos National Laboratory.
GBS has adopted the newest version of DOE-2.2 (version 48r in 64-bit), for our February 2014 release. The baseline used for the following comparisons is the Green Building Studio October 2013 build, which uses version 44e4 of DOE-2.2. The 64-bit version allows us to utilize more of our cloud-based server’s hardware and memory, which results in an increase in the simulation speed.
Simulation time of a GBS test model (GSA Headquarters, used in the Autodesk Benchtop Commercial Office Project), were reduced from 50 seconds with DOE-2.2 44e4 (used by the previous GBS release) to 32 seconds with DOE-2.2-48r 64-bit (used by the February 14, 2014 GBS release). We expect typical reductions in simulation time with this new release of approximately 30 percent.
The validation results presented here are from 464 test cases covering 4 conceptual models, 16 climate locations, and 29 different building types. The change in annual electricity consumption ranges between -2.1% to 3.0%, with a median change of 0.4% and an average change of 0.4%. The fuel usage differences range from -7.6% to 6.9%, with a median of -1.0% and an average of -1.2%. One of the instances of the largest change in annual fuel usage is the Fargo ND Single Family test model, which is described below.
Changes in Annual Electricity Uses in 464 Test Cases
Changes in Annual Fuel Uses in 464 Test Cases
Some of the main computational changes to DOE-2.2, which impact the GBS energy results, include modifications of window convective calculations per NREL documentation, and the addition of an ASHRAE enhanced infiltration correction.
Other significant changes to DOE-2.2 version 48r include conversion to Intel Fortran and C compilers; replacement of triangular or rectangular shapes for shading surfaces with rich polygons (up to a maximum of 128 vertices); and the increase in the number of limits of many model parameters. This will enable the energy analysis of more complex Revit models that may have previously been exceeding the the previous version’s limits.
|Baseline (DOE-2.2-44e4) limits||New DOE-2.2 version 48r limits|
|Number of spaces||4096||9999|
|Number of exterior walls||8192||64,000|
|Number of windows||8192||64,000|
|Number of interior walls||8192||64,000|
|Number of Underground walls||8192||10,000|
|Number of doors||1024||4096|
|Number of building shades||1024||10,000|
|Number of polygons||32,768||120,000|
Below is a summary of the energy results changes from two of our test models.
One of the office test models is located in Boston MA, with an area of 25,000 square feet, five floors with approximately 40% window-to-wall ratio. The largest end-use change for this test model is a reduction in heating energy (11.5%). The largest heating load component of this test model is window conductivity (approximately 60%). One of the changes to version 48r of DOE-2.2 is the modifications of window convective calculations, therefore it is expected that this model would have changes in heating energy. The window conductivity heating load is reduced an average of 10% from the baseline to the new release.
Boston MA Office Test Model
|Office Test Model Annual Electricity Use (kWh)||HVAC||Lighting||Misc Equipment||Total|
|February 2014 Release||117,446||82,451||124,837||324,734|
|Office Test Model Annual Fuel Use (Therms)||HVAC||DHW||Total|
|February 2014 Release||2,513||476||2,989|
The Fargo ND Single Family test model is 3000 square feet, including a conditioned underground level. The largest change is a 7% increase in heating energy, with the largest heating load component being infiltration. One of the changes to version 48r of DOE-2.2 is the infiltration correction, therefore it is expected that this model would have changes in heating energy.
Fargo ND Single Family Test Model
|Single Family Test Model Annual Electricity Use (kWh)||HVAC||Lighting||Misc Equipment||Total|
|February 2014 Release||3,486||4,512||5,246||13,244|
|Single Family Test Model Annual Fuel Use (Therms)||HVAC||DHW||Total|
|February 2014 Release||1161||149||1310|
We’d like to hear from you: Is this information helpful, and what other information pertinent to GBS energy analysis calculations would you like to see? Contact us at email@example.com, or through the Autodesk Green Building Studio Community Forum.