Can 5 phase method model dynamic blinds？

1

Recently, I was studying control strategy optimization of dynamical blind slats’ tilt angle. 5 phase method was used to model the annual indoor illuminance. However, I run into a problem that is how to model the dynamic blinds using 5 phase method. Since the slat tilt angle varied with the solar angle of incidence, it seems that 5 phase method can’t model it because the BSDF can’t change hourly in the 5-phase method.

2

Just as in the 3-phase portion, you need to compute the relevant images and matrices with a number of blind slat positions, then pick and choose between them for the final result. What part are you concerned about?

4

Hi Greg, My thoughts is shown as following:

In 3 pm portion:
1.eveytime use gendaylit to generate a sky verctor for a specific hour, and then input this skyvector into dctiemstep to calculate the illuminance for the specific hour.
2.use python to repeat the procedure 1 with varied hours and BSDF file.Below is the pseudo code:

for i in range(8760):
    directIradiance, diffuseIradiance = readepw(hours=i)
    sky = executegendaylit( directIradiance, diffuseIradiance)
    genskyvec -m 1 < sky > skyvector
    ...
    dctimestep View BSDF[i] D skyvector > illum.txt

And in direct sun coefficient and direct solar part, similar methods were used, everytime I used xform to rotate the blind slats, and then use for… loop to repeat.

But I think this method is very complicated, is there any more simple methods?

Thank you!

5

Sorry for the slow response – our group is in the middle of completing a paper submission, with many moving parts. @Taoning_Wang has written a Python system called frads that automates N-phase calculations and their analysis, which is complicated as you point out. Hopefully, he’ll have time in the next days to offer some direction. Meanwhile, feel free to explore the github project page.

Cheers,
-Greg