How does blind states in daylight coefficient based simulation work?


#1

Hey Oliver, by any chance that you tested the “states” in this case. for some reasons, the results from different state doesn’t make sense to me.


3-phase method example file - Direction of windows
#2

Hi Mingbo, why is that? Which part doesn’t make sense to you?


#3

Hi @mostapha I was testing daylight coefficient workflows. Regardless the hour mismatching problem on direct illuminance data (I mentioned in the other post), switching states (VT-0.9, and VT-0.1) doesn’t do anything to the direct or diffuse result (minor +/- 50-100 lux) .

Is this “state” meaning the same as “state” in Daysim? this state setting is for Radiance to run several simulations with each state, correct?


#4

Yes and yes. If you see the folder you check the folder you will see 4 files for each state. For instance for north_facing window group and dark_glass_0.25 state you will find these files:

  1. total..north_facing..dark_glass_0.25.ill: This file includes the total contribution from Tregenza sky. (gendaymtx)
  2. direct..north_facing..dark_glass_0.25.ill: This file includes the contribution from direct only Tregenza sky (gendaymtx -d). The −d option may be used to produce a sun-only matrix, with no sky contributions.
  3. sun..north_facing..dark_glass_0.25.ill: This file includes the contribution of direct sun from hourly analemma. The values in this file will be used to correct the values inside direct file.
  4. north_facing..dark_glass_0.25.ill: This is the final results which is = total - direct + sun.

You can add/remove the contribution of each source by using blind_states input. You can also open the files and check the values from two different source if that is easier for comparison.

For future reference:

  1. This naming convention is changing to a new naming convention:
  1. Soon there will also be a radout.db file will include all these results in one place. You can query the values for every combination from Results table.

#5

ok, thanks for clarify. now I think I was understanding “states” correctly.
So here is problem:
I am just trying to test if states work correctly in DC simulation. The model here is a simple box model only with one facade has window.
I ran two VT (5%, 60%) individually without any state setting, and here are illuminance results at 2pm of Dec21st:
image
image

#################################################################################
###########################Below are simulations with state settings######################
#################################################################################

Here is a screenshot of how I set states:


and reading data from each state:

State (0,-1) without any window (light source)
image

State (0,0) normal state (VT 0.6)
image

State (0,1) normal state (VT 0.05)
image

So clearly it didn’t run with VT0.05.


#6

Thanks @MingboPeng! Yes. Something is not working. I’m in the middle of making several changes to honeybee[+] right now and can’t test it for myself but if you zip the folder and send me a link to it I can take a closer look and see what’s going on. Also does the material with 5% transmittance created correctly? That is the first thing that I would check.


#7

Hi @MingboPeng, Which .xml files did you use?


#8

Hi @OlivierDambron, I am not testing 3 phase yet. this was just daylight coefficient.