Fbes：Fraction of body exposed to sun

[mercury1123]

Hello，everyone!I have some questions about calculating solar adjusted MRT.In my calculations, the direct normal radiation were already high at some moments, and there was no abnormality in the air temperature and surface temperature, but the radiant temperature at this time did not reach the expected value.So I wonder that，is there something wrong with Fbes(the fraction of body exposed to sun)?And,in the computation of solar adjusted MRT, how can we define or find information about Fbes?
When uploading my GH file, I also uploaded the EPW file because my EPW file was modified based on the measurement data.Hope to get your answers and suggestions.CHN_Tibet.Lhasa.555910_CSWD_DRYT_1.epw (1.5 MB)NewModel.gh (938.5 KB)

[chris]

@mercury1123 ,
I think I addressed your question here:

Indoor Radiant Temperature

@mercury1123 , Excellent question. The Honeybee microclimate maps and Ladybug Solar Temperature Adjustor are using the SolarCal model. The Fsvv is computed by shooting out a hemisphere of rays from the point of interest (where the occupant is located) and seeing how many rays get through geometry to see the sky. This is the case for both the microclimate maps and the solar temperature adjustor. For the Fbes, the sun vector for the given hour is shot through the point(s) of interest to see …

If it would help, I can post an example of how to use the solar temperature adjustor to build a microclimate map that has the fractional Fbes (rather than a 0 or a 1) if you think that this is the issue you are expereinceing here.

[chris]

I’ll also add that another (faster) means of getting a fractional fBes (that isn’t a 0 or a 1) is to build a microclimate map with 3 planes (one at the feet, one at the mid section, and one at the head). Averaging the MRT results across the 3 planes should give you MRT vales that are comparable to that achieved by using a fBes derived from 3 points. I can post an example for this one if you’re running a longitudinal section across a lot of time and need the faster runtime.

[mercury1123]

@chris

use the solar temperature adjustor to build a microclimate map

Thank you for your reply. Now that MicroclimateMap has calculated Fsvv and Fbes, is it possible to see the result in a component? I am very interested in using of the solar temperature adjustor to build a indoor microclimate map that you mentioned.I greatly appreciate that If you can provide a example.

[chris]

I know that I am several months late in posting the sample file that I mentioned here but better late than never!

Here you can see a hydra example that shows how the microclimate map calculations are happening under the hood:
http://hydrashare.github.io/hydra/viewer?owner=chriswmackey&fork=hydra_2&id=Deconstructed_Microclimate_Map

Additionally, unlike the microclimate maps, you will see that you get a finer level of control over the shortwave calculation because the definition uses the “solar temperature adjustor” component. This allows you to get Fbes values that are different than 0 or 1. It also allows you to change things like the orientation of the body in relation to the sun.

[AbrahamYezioro]

Hi @chris,
Thanks for this new example.
I have a couple of questions that i was in doubt if asking here or reporting as a bug in github. I decided that here is the place:

1. You are asking for an analysis period of typicalAutumnWeek. Though the whole year is simulated instead just this week. As a result you get here a 8935 list. I remember in previous versions that the simulation was performed just for the requested period.
2. In the example the ColorSurfaces component is complaining about:

1. Not all of the connected _srfData branches are of the same length or there are more than 8760 values in the list.

in a way that the rest of the script fails to continue the calculation. All branches (21) are 8935 long. Tried to connect the weekly analysisPeriod but that didn’t work either.
3. Also noticed that all components are dated 20 november … a new version is coming up?

Thanks,
-A.

[AbrahamYezioro]

Update:
Running with the E+ component indeed simulates only the requested analysis period.

And another one. Checking the idf produced by the OS component it simulates the whole year + a week (I’m using OS 2.7 and it may be related to other issues reported for this version):

-A.

[AbrahamYezioro]

Another update:
Uninstalled 2.7 and installed 2.4: It is working fine. So, probably, another bug for 2.7.
-A.

[chris]

@AbrahamYezioro ,

Glad you figured it out. There is a bug in OpenStudio 2.7 for RunPeriods less than a month so we currently recommend that everyone stick to using OpenStudio 2.5 or 2.4 for now if you want full compatibility with Honeybee. OpenStudio team is aware of the bug and they have told me that they have fixed it in the version that they are working on now. So (fingers crossed) that OpenStudio 2.8 won’t have any major Honeybee bugs when it is released.