Glare Simulation Questions
Patrick Keeney
J. Alstan Jakubiec
Thanks for your well-wishes. I admit, its a little odd to respond to
questions about a software not released yet, but I am sure that some might
find this interesting nonetheless. Plus, your questions are very good
ones. By the way, the 1.9 version of DIVA will always be free to download.
> First, in order to calculate an Annual Glare Map, am I to assume that a
> "viewport" must me set in the Rhino file. To elaborate....this
> "viewport" must be a view from a typical place where the occupant will
> be sitting (its an office building). Therefore, if you were to run a
> glare simulation
> while the standard four screens are showing (ie, plan, an exterior
> perspective, and 2 elevation views), you would have an incorrect
> calculation of glare?
This works much the same as when you run a visualization in DIVA. It will
by default use the current selected viewport (but will only work when it
is a perspective type, not one of the parallel projections). You can also
save viewports and simulate all saved viewports at once. This is explained
somewhat on the page:
http://diva4rhino.com/user-guide/simulation-types/annual-glare Running an
annual glare calculation using an exterior view would present meaningless
results, as the evaluation is based on occupant surveys in interior,
office-like conditions.
> Second, following a correct run of an Annual Glare Map, I'd like to move
> to a Climate Based Metric simulation to calculate Daylight Autonomy. I
> was
> going to start with the Conceptual Dynamic Shading method. After
> choosing the workplane sensor nodes, am I to assume that if the DGP
> level exceeds
> the allowable level at *just one* of these nodes, it will tell the blinds
> to close? (I suppose this makes sense because if one person
> is uncomfortable, then they will pull the blinds.) Likewise, if the
> blinds close, will they reopen the next hour when the nodes don't exceed
> that
> allowable DGP level?
This is a great question. The distinction between sensor nodes and views
is very important. DGP is calculated per view (one view, or multiple if
you use the saved views option), and sensor nodes are just daylight
sensors which can be set to be the location of a shading-control sensor or
a lighting-control sensor. First, let me cover the example where you DO
NOT calculate DGP. Without a DGP run, a sensor will trigger the blinds
when direct sunlight hits the sensor or based on lux thresholds as defined
in the automatic shading control modes. This is described more in detail
at http://diva4rhino.com/user-guide/rhino/advanced-shading under the
"Dynamic Shading Control Systems" heading.
If you simulate DGP first and specify a manual control, the DGP value
itself (comfort) is supposed to control the shading device. We have a
hidden GUI option that allows you to set a separate DGP view for each
shading control system, but DAYSIM cannot support this yet, hence it is
disabled. In fact, in our early tests it seems that the beta version of
DAYSIM that ships with the beta version of DIVA-2.0 isn't currently
respecting the DGP simulation results for blind control when it should.
This is something we are working with the DAYSIM developers to remedy
before our final release. Because a specific DGP view cannot be associated
with a specific shading system, all shading systems will be controlled by
the worst-case DGP view using the "occupants cannot adapt" value under
"adaptive comfort." Read a little more on adaptive comfort at this page:
http://diva4rhino.com/user-guide/simulation-types/climate-based -- its
mostly experimental for now.
> Also, how does DIVA know where to place these blinds.....if you see the
> attached image, I am simulating two buildings (the left without glare
> calculated & the right with glare calculated) to see the effects that
> including glare in the calculation will do.....how will DIVA know to pull
> the blinds on glazing on the building on the right and not the building
> on the left. ie, how does it know to associate certain glass .rad
> materials
> with a certain node group and not with another node group?
I recommend to read the limitations of conceptual and detailed shading
here, http://diva4rhino.com/user-guide/rhino/advanced-shading. Let me know
if you have any questions after that!
Best,
Alstan
> the allowable level at *just one* of these nodes, it will tell the blinds
Patrick Keeney
--
You received this message because you are subscribed to the Google Groups "DIVA for Rhino" group.
To post to this group, send email to diva-for-rhino@googlegroups.com.
To unsubscribe from this group, send email to diva-for-rhino+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/diva-for-rhino?hl=en.
Max D
> A few more questions. What can I use to open an OCT file or an RAD file. I
> am trying to double check the results from the Annual Glare Map DGP
> results. I was hoping for something similar to the green and red image on
> this page: http://diva4rhino.com/user-guide/simulation-types/annual-glare
> I've tried wxfalsecolor and a text reader but no luck.
how would you check the results by looking at the octree? it contains
the scene geometry, sky etc., so there is no way opening it would
allow you to "see" what the dgp calc has given you. the annual glare
images should be in the results/filename folder and pop up in html
page that opens after a calc.
> Also, how will DIVA know to use the DGP results....the results from that
> simulation go to the C:\DIVA\Temp\Daylighting folder, while my actual Diva
> simulation results go to the Daylighting Analysis_Rhino file name - Results
> folder. Hopefully, this is not too much of a dumb question, but since there
> is no place to specify the input file for the DGP results, I guess I'm
> confused as to how it makes the linkage.
As Alstan recently told me, they are not being used for the conceptual
shading mode at this moment.. and skimming his earlier reply, it
appears they are not being used for advanced shading control right
now, either. Instead, advanced shading uses specified sensor nodes.
> This is assuming that DIVA can read the DGP results to begin with because
> you had mentioned the bug that you are working thru for the new release?
In principle Daysim can read the DGP results and use them for
abovementioned operations, but moving to the new release there simply
seem to be some quirks right now ;)
Best wishes,
Max
J. Alstan Jakubiec
Indeed what Max says about rad and oct files is correct -- they contain
only geometric information. The actual output file for an annual glare
(eDGPs) simulation will have the extension *.dgp. The output you reference
(http://diva4rhino.com/user-guide/simulation-types/annual-glare) should be
automatically generated when you run an annual glare simulation (from the
Metrics command, Daylight Images > Annual Glare). You will need to be in
an active perspective view in the Rhino file when you run it. Be warned,
it can take a few hours to complete successfully.
>> Also, how will DIVA know to use the DGP results....the results from that
>> simulation go to the C:\DIVA\Temp\Daylighting folder, while my actual
>> Diva simulation results go to the Daylighting Analysis_Rhino file name
>> - Results folder. Hopefully, this is not too much of a dumb question,
>> but since there is no place to specify the input file for the DGP
>> results, I guess I'm
>> confused as to how it makes the linkage.
Not a dumb question at all. Eventually, once DAYSIM provides for it, we
will make the linkage more explicit. The current behavior operates like
this:
1. Hourly illuminance files are generated by DAYSIM for every possible
shading state.
2. ds_electric_lighting decides how to set the shading states (and thus
which illuminance value to use at each hour).
A. If a dgp file exists for each shading state and the control mode is
manual or conceptual shading is used, it will use DGP to control the
shading state.
B. If dgp files are not present, direct sunlight is used as a correlate
for glare in manual control modes and conceptual shading modes.
C. If automatic controls are specified, those are used in all cases.
3. DA files are generated based on the decided shading states at each hour
and the occupancy schedule.
In summary, the program always looks for DGP files, but if they don't
exist, then direct lighting on defined workplane sensors is used instead.
As I mentioned before though, DGP isn't being properly considered for
controls in the current beta release.
Best,
Alstan
Patrick Keeney
--
You received this message because you are subscribed to the Google Groups "DIVA for Rhino" group.
To post to this group, send email to diva-for-rhino@googlegroups.com.
To unsubscribe from this group, send email to diva-for-rhino+unsubscribe@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/diva-for-rhino?hl=en.
Max D
no problem, we're here to help. inline as usual--
> Image 1).....it shows that blinds are down pretty much the entire time
> during the winter months, but are raised pretty much the entire summer. I
> could possilby justify this because the winter months have lower sun angles
> which means direct light will penetrate further into the floor plate, thus
> having direct contact with the workplane sensors. Could someone please
> expound on this for me?
I think your reasoning on this is sound. It is interesting to see the
direct comparison between a "worst glare / direct light hits sensor"
result and the schedule you would get from a dgp profile. Of course
they are different, as they use individual decision schemes as to why
blinds get pulled and when.
> Test 3 - Conceptual Shading device using DGP results.....This shading
> shedule chart shows that the blinds are pretty much pulled the entire
> year......however after double checking the .dgp file in excel, it showed
> that there were only 327 hours out of the year where DGP was greater than
> .4. Doesn't the software automatically use .4 as the switch between blinds
> pulled and blinds not pulled?
As the DGP results do not currently get used for blinds actuation, you
would see no difference in the conceptual shading CBM results, whether
you have generated a DGP profile or not. That will change, of course!
The .4 threshold is correct:
http://diva4rhino.com/user-guide/rhino/advanced-shading
Best wishes,
Max
J. Alstan Jakubiec
> Test 2- conceptual shading device with workplane sensors.....I can
> rationalize the decrease in Daylight Autonomy (they go from about 45%
> without shading to 35% with a shading device...this makes sense).
> However, I am concerned about the chart that shows when the blinds are
> down (see
> Image 1).....it shows that blinds are down pretty much the entire time
> during the winter months, but are raised pretty much the entire summer.
> I could possilby justify this because the winter months have lower sun
> angles which means direct light will penetrate further into the floor
> plate, thus having direct contact with the workplane sensors. Could
> someone please
> expound on this for me?
as they experience direct sunlight. Although it is likely to be bad for
thermal loads (blocking solar during winter, allowing it during summer),
the decision framework is based on assumed comfort at the workplane
sensors (or based on DGP, but as already established, this is not the case
in the beta release).
> Test 3 - Conceptual Shading device using DGP results.....This shading
> shedule chart shows that the blinds are pretty much pulled the entire
> year......however after double checking the .dgp file in excel, it showed
> that there were only 327 hours out of the year where DGP was greater than
> .4. Doesn't the software automatically use .4 as the switch between
> blinds pulled and blinds not pulled?
This is an example where the beta is not currently working as designed,
due to what I mentioned previously with the DAYSIM-beta version not making
shading decisions based on DGP, even when it should do so. So in the final
version what will happen is that when a user experiences a DGP>0.4, the
blinds will be closed until the next lunch break or the beginning of the
next day. What probably CURRENTLY is happening is that no workplane
sensors are explicitly set. In the case of not setting workplane sensors,
EVERY sensor functions as a workplane sensor. In this case, if any
directly light comes into the space at all, the blinds will close. Which
is what you see in the second shading schedule you attached.
> Sorry again for these long winded questions.....I'm trying to work
> throught this on my own but when I get pondering results, I could use
> some help.
Not at all. This is useful for me too, as it shows I need to at the least
write an article with more explanation on the control schema. Its half
based on a 2004 paper ("Lightswitch-2002: a model for manual and automated
control of electric lighting and blinds", CF Reinhart, 2004, Solar Energy)
which accounts for user behavior. The automatic control methods using
detailed shading are new but based on existing real-world systems.
Best!
Alstan