Construction:AirBoundary - "Solar Enclosures" for Grouped Zones Solar Distribution

asked 2019-11-11 09:17:51 -0600

jmwoolley
77 ●4

updated 2020-01-26 11:38:48 -0600

EnergyPlus v9.2.0 introduced a new object class - Construction:AirBoundary - intended to model boundaries between thermal-zones that are not associated with physical surfaces. A thermal-zone boundary surface that uses Construciton:AirBoundary allows transmission of solar radiation and longwave infrared radiation between adjacent thermal-zones. The feature can also model balanced air mixing between adjacent thermal-zones without explicit definition of ZoneCrossMixing objects. Prior to this feature, one could either model such thermal-zone boundaries: (1) as windows (which allow transmission of solar radiation but not infrared radiation), or (2) as Material:InfraredTransparent surfaces (which allow transmission of longwave infrared radiation but not solar radiation), or (3) as lightweight opaque surfaces (which absorb both solar radiation and longwave infrared radiation).

I've recently implemented Construction:AirBoundary in the model of a building that has many contiguous open-plan interior spaces. I receive the following fatal error several times:

** Severe ** SetupSolarEnclosuresAndAirBoundaries: Solar enclosure grouping error for Surface="L1_S32_WALL_325_A".This surface enclosure num=1. Other side enclosure num=2

** Severe  ** SetupSolarEnclosuresAndAirBoundaries: Solar enclosure grouping error for Surface="L1_S11_WALL_000_A".This surface enclosure num=1. Other side enclosure num=2

... etc

According to the Input Output Reference, if two or more contiguous thermal-zones are separated by Construction:AirBoundary surfaces, these thermal-zones are grouped together when calculating the distribution of solar heat gains. The error appears to be related to the method used for grouping thermal-zones. My model has many contiguous thermal-zones, most of which have multiple Construction:AirBoundary surfaces. In some cases, two thermal zones are connected to each other by multiple Construction:AirBoundary surfaces. Presumably, this method would group all contiguous thermal-zones into one big zone, but the error suggests that at least two "solar enclosure groups" have been assigned.

In case it is not clear, the surfaces named in the error are using Construction:AirBoundary, so they should not separate "solar enclosure groups".

I have the following questions:

Is there any way to determine what thermal-zones have been assigned to different "solar enclosure groups"? Better yet, might there be a way to choose which zones get assigned to different "solar enclosure groups"?
When thermal-zones are grouped together for solar distribution calculations, the resulting "solar enclosure groups" will very likely be non-convex (this is certainly true for my model). How does the Construction:AirBoundary feature deal with models that use FullInteriorAndExterior solar distribution? Does it revert to FullExterior and simply distribute solar heat gains across all floor surfaces in the "solar enclosure group"?
What is the cause of this error? (Why is EnergyPlus attempting to create two "solar enclosure groups" for groups of zones that are separated by Construction:AirBoundary surfaces?)
What can be done to avoid the error I've encountered?

edit retag flag offensive close merge delete

Comments

Oh boy, I will be following this question. Exciting topic!

Luis Lara ( 2019-11-12 11:07:37 -0600 )edit

@jmwoolley et al Adding this comment at the top for good visibility. Several issues with Construction:AirBoundary have been fixed for the v9.3 release (end of March 2020). The interior window is now working, the enclosure grouping error is fixed, and beam solar now penetrates to other zones with FullInteriorExteror. See 7867.

MJWitte ( 2020-03-23 12:15:19 -0600 )edit

One remaining question noted below was 1-2% solar imbalance between transmitted solar, absorbed solar, and outgoing solar. This appears to happen even in zones without air boundaries. I don't have an explanation for that.

MJWitte ( 2020-03-23 12:17:03 -0600 )edit

add a comment

2 Answers

Sort by » oldest newest most voted

answered 2019-11-15 12:31:55 -0600

MJWitte

9740 ●6 ●10 http://www.gard.com/

updated 2019-11-15 12:57:20 -0600

The enclosures may be reported in the eio output with Output:Surfaces:List,ViewFactorInfo;. This lists every enclosure for both solar and radiant exchange (they can be different depending on the user inputs in the air boundary surfaces) along with surface-to-surface view factors.
Grouped solar enclosures use the global solar distribution option. If the resulting zone is non-convex, then FullInteriorAndExterior may have problems. Nothing special has been implemented to handle this.
(and 4.) The error message implies that some pair of interzone surfaces are not using the same Construction:AirBoundary object. Need to see the offending idf file to answer with certainty.

I'll also post some comments on the questioner's self-answer.

edit flag offensive delete link

add a comment

answered 2019-11-14 15:37:21 -0600

jmwoolley
77 ●4

After some further investigation, I have partially answered some of my questions.

First, I've discovered that I can manually force which thermal-zones are grouped into "solar-enclosures" by defining Material:InfraredTransparent surfaces as the inter-zone boundaries wherever I want to separate "solar-enclosures". I can thereby avoid the error I encountered, but the downside is that Material:InfraredTransparent surfaces are not transparent to solar radiation. Regardless, the work-around does not explain why the error would occur.

Second, by reviewing the outputs Surface Inside Face Solar Radiation Heat Gain Rate and Surface Inside Face Beam Solar Radiation Heat Gain Rate it appears that Construction:AirBoundary surfaces convert beam solar radiation to diffuse solar radiation. Surfaces in zones beyond a Construction:AirBoundary surface receive "Solar Heat Gain" but receive zero "Beam Solar Radiation". This would explain how the feature deals with non-convex solar enclosures - it seems that when using "FullInteriorAndExterior" solar distribution, shadowing calculations are only performed for the thermal-zone with windows, not for the "solar-enclosure group" as a whole.

However, I am confused by this behavior because it is not consistent with documentation for Construction:AirBoundary Solar and Daylighting Method = GroupedZones (EP v9.2.0 IOM - 9/27/2019). Interestingly, the behavior observed is exactly what I would expect for calculation of solar distribution through interior windows, but when I choose Solar and Daylighting Method = InteriorWindow I receive the following warning:

** Warning ** CreateAirBoundaryConstructions:Construction:AirBoundary Solar and Daylighting Method=InteriorWindow is not functional.  

**   ~~~   ** Using GroupedZones method instead for Construction:AirBoundary = AIR BOUNDARY.

Can anyone provide further clarification about what to expect from this new (very important) feature?

edit flag offensive delete link

Comments

Without Construction:AirBoundary surfaces using the grouped solar option, solar enclosures will be one-to-one with zones. No grouping. The zeros for inside face beam solar may indicate a problem when using FullInteriorAndExterior with grouped zones. Example file 5ZoneAirCooled_AirBoundaries shows values for beam incident on floor F5-1 (in the core zone) with FullExterior but zeros with FullInteriorAndExterior. Will look into that. Regarding the disable InteriorWindow option - well - it's exactly what the error message says - it isn't working right yet.

MJWitte ( 2019-11-15 12:56:31 -0600 )edit

@MJWitte Another detail that may be helpful. When I use Solar and Daylighting Method = GroupedZones together with Solar Distribution = InteriorAndExterior, I find that for a solar enclosure the sum of outputs Surface Inside Face Solar Radiation Heat Gain Rate aligns with the output Zone Windows Total Transmitted Solar Radiation Rate, even though the output Surface Inside Face Beam Solar Radiation Heat Gain Rate is zero for all surfaces in interior zones. So, it seems that the energy balance is correct, and as if solar radiation is transferred to interior zones as diffuse. You agree?

jmwoolley ( 2019-11-21 11:05:22 -0600 )edit

Not sure. It's encouraging that the energy balances. Is the sum of Surface Inside Face Solar Radiation Heat Gain Rate the same (or close) using FullExterior for the same file? That would be good, but the ultimate goal here was to project the beam solar into the connected spaces, and that appears to be broken.

MJWitte ( 2019-11-25 10:13:21 -0600 )edit

@MJWitte Using either FullExterior or InteriorAndExterior, the sum of Surface Inside Face Solar Radiation Heat Gain Rate plus Surface Window Shortwave from Zone Back Out Window Heat Transfer Rate is nearly equal to Zone Windows Total Transmitted Solar Radiation Rate. For both solar distribution methods, the former is always ~1-2% smaller than the later. Why?

jmwoolley ( 2019-12-08 16:27:12 -0600 )edit

@MJWitte Using FullExterior all floors in a solar enclosure receive Surface Inside Face Beam Solar Radiation Heat Gain Rate, and no other surfaces receive beam solar. Surface Inside Face Solar Radiation Heat Gain Rate per Area is equal for all floor surfaces, confirming that beam solar is distributed equally across all floor surfaces.
Using InteriorAndExteriorSurface Inside Face Beam Solar Radiation Heat Gain Rate reports reasonable values for appropriate surfaces in exterior zones, but no surfaces in the interior zones receive beam solar.

jmwoolley ( 2019-12-08 16:40:40 -0600 )edit

see more comments