Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page
Ask Your Question
2

Radiance simulation detects no daylight illuminance

asked 2017-09-27 10:52:13 -0600

Ward I gravatar image

updated 2017-09-27 19:19:35 -0600

When I simulate light in my OpenStudio models with Radiance it fails to detect any daylight in my zones at all and the illuminance maps show a dark plane with zero lux throughout the year

sensors and illuminance maps were placed as instructed and my models give the expected output when simulating with OpenStudio and EnergyPlus alone, without Radiance (also for lighting)

Global horizontal illuminance also gives the expected results, so the problem is not that there's no sunlight in my model/weather file

This seems to indicate the problem doesnt lie with the model but with the Radiance Software

I've already tried reinstalling the software

I use OpenStudio v2.2

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted
5

answered 2017-09-27 11:32:51 -0600

updated 2017-10-25 17:32:26 -0600

Good times! When you say your models "give the expected output" with EnergyPlus alone, what exactly does that mean, regarding the daylighting? i.e., are you looking at the illuminance map data from the eplusout.sql file for those runs? Do they show daylight?

UPDATE (2017.10.16) Sorry for the delay, I just returned to the office from medical leave. I tried your simple model (linked in your first comment) with OS 2.3 and it worked fine. Since Results Viewer is no longer included with OpenStudio, I loaded the merged_space.ill file in Excel and looked at the Radiance daylight data there. It clearly shows reasonable daylight during daylit hours.

The data is presented as a matrix of data points (columns) and annual hours (rows). So the first four (summer) to eight (winter) columns will always read zero, right down the column. Does this make sense?

UPDATE (2017.10.24) Thanks for posting links to the stdout and stderr. It appears the measure is unhappy with the weather file:

gendaymtx: input weather tape format error
No Instance(s) Available.

...which leads to bad data output that is, unfortunately, poorly reported to you:

fatal - unexpected EOF in header
<stdin>: cannot load matrix
rmtxop: operation failed on '-'
No Instance(s) Available.

output\ts\WG0.ill: cannot load matrix
rmtxop: operation failed on 'output\ts\WG0.ill'

Please try this model with a standard weather file (i.e. one included with EnergyPlus) and if you could post a link to your Brussels wx file I'll take a look at that and try to see what the specific issue is. It appears the Radiance weather translator is pretty particular about format, and we can do a better job of letting the user know about it.

SOLVED (2017.10.25) Thanks again @Ward I, I think we got it. Your advisor's modified file had a few lines with a bunch of trailing commas in them -- most notably, the first line of the header which I'm betting is the only one the Radiance epw data translator really cares about. Specifically, Lines 1, 4, 5, and 7 all had a number of trailing commas, e.g.:

## BEL_Brussels.064510_IWEC_epw_correctedsnowfall.epw
LOCATION,BRUSSELS,-,BEL,IWEC Data,064510,50.9,4.53,1,58,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,

## BEL_Brussels.064510_IWEC.epw
LOCATION,BRUSSELS,-,BEL,IWEC Data,064510,50.9,4.53,1.0,58.0

I removed the trailing commas from L1, L4-5 but left them in L7 (it's a comment line I suspect the Radiance translator skips anyway), applied the new weather file to your test model, and it ran fine. Please let me know if this works on your main model as well; I expect it will. I will enter an issue about this in the OpenStudio issues tracker. I doubt we will get the Radiance weather translator fixed anytime soon, but at least we can notify the user there was an issue with weather translation a little better than we are.

edit flag offensive delete link more

Comments

Resultsviewer gives (less detailed) illuminance maps when using E+ for lighting. On these you can clearly see the position of the windows and such. I didn't analyse or or checked its accuracy but these results seem to be normal as opposed tot the 0 lux throughout the year with Radiance.

I made a very simple model just to test this problem and to make sure the problems weren't with my actual, complexer models: https://www.dropbox.com/s/sxe7wev8bv8...

Ward I gravatar imageWard I ( 2017-09-28 05:15:15 -0600 )edit

Any results? Updating to version 2.3 doesn't seem to help

Ward I gravatar imageWard I ( 2017-10-08 09:23:16 -0600 )edit

Yes, I was able to try your simple model in 2.3; please see my updated answer for additional info.

rpg777 gravatar imagerpg777 ( 2017-10-16 15:14:42 -0600 )edit

That the model works implies that it's a local software problem I already tried reinstalling the software several times The simulation explicitly states that it doesn't get any results when parsing the daylighting results " Info: An error has occurred; no results for space 'Space_1'."

Ward I gravatar imageWard I ( 2017-10-21 10:24:06 -0600 )edit

I regret that Dview doesnt show the illuminance maps. I've taken a look at the merged_space.ill file but found it hard to interpret, despite reading https://unmethours.com/question/22864...) It seems way too short even for the small test model, since I expect a value for every point at least for every hour: https://www.dropbox.com/s/uyh2xog2k4d...

Ward I gravatar imageWard I ( 2017-10-21 10:24:22 -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

1 follower

Stats

Asked: 2017-09-27 10:52:13 -0600

Seen: 247 times

Last updated: Oct 25 '17