Ceiling vs IntFlr and Adiabatic surfaces

[GoogleCodeExporter]

What SDD Object:Property(s) are relevant to this issue?
Ceiling
IntFlr

Explanation of addition, revision, or clarification:

Hoping to get some clarifications on ACM or SDD modeling guidelines (if any) 
for the following scenarios:

a) Is the floor of a space with no plenum the Ceiling of the space below, or 
the IntFlr of the space above?  
b) Is the floor of an attic (conditioned or unconditioned) space an IntFlr or 
Ceiling?
c) Is the "floor" of a plenum space a Ceiling or an IntFlr?
d) If a tree falls in a forest and no one is around to hear it, does it make a 
sound? (sorry, couldn't help it; these questions seem more philosophical than 
modeling related, but I believe are important to confirm for rules/third-party 
translation) 

This question was spurred by a Severe error in the 00300 prototype that Kyle 
identified where a zone did not have a Floor; here is the error: 

   ** Severe  ** ComputeIntSolarAbsorpFactors: Solar distribution model is set to place solar gains on the zone floor,
   **   ~~~   ** but Zone ="PERIMETER_MID_ZN_2 THERMAL ZONE" does not appear to have any floor surfaces.
   **   ~~~   ** Solar gains will be spread evenly on all surfaces in the zone, and the simulation continues...

I dug into this model a bit and there are some other serious issues with 
geometry definitions; which probably stem both the original E+ file definitions 
and the state of the Forward translator back when these models were created.

Before these models are fixed (requires re-exporting geometry or some 
substantial OS/text editor work), I thought we should make clear the modeling 
conventions for these surfaces.  Then there is the task of going through the 
00200, 00300, 00400, and 00600 prototypes to fix the geometry....

Finally, how are adiabatic interior surfaces defined in the SDD?  Is it simply 
that they do not have an AdjacentSpcRef?

Proposed resolution:

Please provide any additional information below.

Original issue reported on code.google.com by da...@360-analytics.com on 1 May 2013 at 1:40

[GoogleCodeExporter]

I'm not sure I know the exact answer to all this.  I understood that a surface 
between two zones was part of both of them, i.e. it might be both a floor and a 
ceiling, depending on the reference.  The other side of each interior surface 
is defined as other zone, adiabatic, or some other options, which determine the 
boundary conditions for calculating heat transfer through the surface.  If it 
is adiabatic, then any specification of the adjacent space is not used, there 
is simply no heat transfer.  

Original comment by rhedr...@archenergy.com on 1 May 2013 at 6:18

[GoogleCodeExporter]

I mostly agree with Roger.  My understanding of the SDD modeling convention for 
surfaces is as follows:

1. Only a single "side" of each interior surface in the building is 
instantiated in an SDD XML instance file.

2. Each instantiated interior surface is a child of a single host Space 
instance.

3. Each interior surface type (Ceiling, InteriorFloor, InteriorWall) has an 
AdjacentSpaceReference property that specifies the Space on the "other side" of 
the surface.

Therefore, for all of David's a, b, and c scenarios I would say the following: 
the InteriorFloor child of a Space is one side of the Ceiling of the adjacent 
Space below.

It is acceptable for a Space to not have an explicit InteriorFloor child 
instance as long as the adjacent Space below has an explicit Ceiling child 
instance.  Either way the OpenStudio transformation would create the implicit 
other side surface in the adjacent space, or set the EnergyPlus flag that 
causes EnergyPlus to auto-create the other side.

I believe this is true whether the Space has a plenum or not, is a plenum, or 
is an attic.  My assumption here is that plenum and attic spaces are 
instantiated as Space objects in the SDD.  I'm not sure how the SDD identifies 
an "attic" space.  Space.ConditioningType seems to identify plenum spaces.

And I do not know how adiabatic surfaces are identified in the SDD.  How are 
adiabatic surfaces used in the Standards?

Trees should probably have a tilt angle property so we can tell if they've 
fallen without having to detect a sound.

Original comment by RJHitchc...@gmail.com on 1 May 2013 at 9:08

[GoogleCodeExporter]

Revising owner to Roger, as he is working on envelope rules.  We may need some 
CHECKCODE/CHECKSIM rules to ensure spaces have IntFlr objects when there is not 
Ceiling that references the space as "AdjacentSpcRef".

Original comment by da...@360-analytics.com on 15 Feb 2014 at 7:36

• Changed state: Discussion
• Added labels: Priority-Low
• Removed labels: Priority-High

[GoogleCodeExporter]

This is a translation issue, and I believe it is operating correctly.  The 
.cibd file only specifies a single surface between two zones.  EnergyPlus is 
converting that one surface to a pair of surfaces, either a floor/ceiling pair 
or a pair of interior walls. Closing issue. 

Original comment by rhedr...@archenergy.com on 12 Mar 2014 at 11:46

• Changed state: DoneVerified