[esp-r] Re: Calculation of solar gains through complex shading devices

Tue May 7 14:51:30 BST 2013

BSDFs can be generated using WINDOW, and I evaluated using Radiance's
genBSDF to generate Solar Bidirectional Information (compared the results
with the analytical models done inside WINDOW 6.0 and over). Nowadays, the
BSDFs come in a standard (probably still evolving) .xml file, and I thinkg
E+ format is very simple (I saw one once, but very quickly).

On the other hand, how does ESPr use Radiance or the Lighting calculations?
Making an annual simulation may take a really long time.

The reason why I chose the Three-phase method is because it allows "easy"
control of shading, and a very fast simulation. Also, it allows to change
and evaluate different shading systems. So it is very good for optimization.

2013/5/7 Geissler Achim <achim.geissler at fhnw.ch>

>  Hi Germán****
>
> ** **
>
> some comments below, inline.****
>
> ** **
>
> Best****
>
> Achim****
>
> ** **
>
>
> ---------------------------------------------------------------------
> T +41 61 467 44 51
>
> achim.geissler at fhnw.ch
> www.fhnw.ch/iebau
> ---------------------------------------------------------------------
>
> ****
>
> ** **
>
> *Von:* germolinal at gmail.com [mailto:germolinal at gmail.com] *Im Auftrag von
> *Germán Molina Larrain
> *Gesendet:* Dienstag, 7. Mai 2013 15:33
> *An:* Geissler Achim
> *Cc:* esp-r at lists.strath.ac.uk
> *Betreff:* Re: [esp-r] Calculation of solar gains through complex shading
> devices****
>
> ** **
>
> Hi Achim, ****
>
> ** **
>
> my original idea of workflow was thought on the basis of a straightforward
> finide difference simulation (as I have done in Matlab)... so it would be
> to make the solar simulations (calculation of solar gains and Daylight),
> and then, to make the thermal simulation; knowing the flows to each node
> (calculated before).****
>
> ** **
>
> Now, since you say that ESPr supports BSDF input, maybe the whole thermal
> simulation should be done in ESPr, leaving Radiance just for lighting. I
> have been working on the generation of the BSDFs, and I think the problem
> is practically solved right now, using tools that already exist (I expect I
> can publish something related to this). How long do those (BSDFed)
> simulations take? I have heard that in E+ they take considerably more time
> than the other ones.****
>
> ** **
>
> I'm afraid I can't tell you, as I have never personally used that
> particular feature of ESP-r. Also, I am not sure if the format is any
> "general" BSDF type. How do you generate your BSDFs?****
>
> ** **
>
> As far as I am aware, the three-phase-method and the "coupling" with
> EnergyPlus requires two runs? Could you describe the necessary workflow /
> calculation steps? Where / in which program is the setting of a Venetian
> type blind set, for example? Based on what?****
>
> ** **
>
> The direct coupling, I was informed, is complicated. But, just as in ESPr,
> E+ was provided with the capability of using BSDF information in a very
> similar way as the Three-phase method does. So everything is actually made
> on E+.****
>
> ** **
>
> ** **
>
> Anyway, I guess the best should be to make the lighting simulation,
> calculate the electric lighting consumption and transform it into internal
> gains for a posterior ESPr simulation, considering BSDFs? Please feel free
> to comment.****
>
> ** **
>
> Well, basically ESP-r and Radiance can be used fully coupled. I.e., if you
> want to have lighting control, you can couple a Radiance model, let that do
> the lighting calculations and give back which percentage of the pre-set set
> point are covered by daylight - ESP-r will then add the rest as electric
> internal gain based on the defined electric light installation. The gain
> due to daylight is taken into account by the solar gain. This can also be
> used with the CFCs - there is an experimental feature which generates the
> Radiance Model with a Venetian type blind during run-time with the slat
> angle as given by solar needs (e.g. "cut off" control). Computing times
> are, of course, somewhat longer than usual.****
>
> ** **
>
> ** **
>
> THANKS VERY MUCH****
>
> ** **
>
> German****
>
> ** **
>
> 2013/5/7 Geissler Achim <achim.geissler at fhnw.ch>****
>
> Dear German****
>
>  ****
>
> basically, that is a very interesting idea. However, I would not expect
> this to be easy and straightforward, as one strength of ESP-r is that it
> solves all the physics simultaneously. Currently, there are two possible
> ways to include complex shading: The "CFC" model for multi-layer
> constructions (to date, this covers venetian type blinds, roller blinds and
> meshes are "in the works" as far as I am aware) and the input of
> bi-directional reflection / transmission input data ("BRTD") - here the
> problem is basically generating the necessary input data, however this may
> be possible via Window 7 (LBNL), now.****
>
>  ****
>
> ESP-r can basically take the solar distribution on the surfaces of the
> zone behind the window into account. In the mainstream code, the CFCs lack
> daylighting, however. I am not sure what the situation in this regard for
> the "BRTD"-type of models is.****
>
>  ****
>
> As far as I am aware, the three-phase-method and the "coupling" with
> EnergyPlus requires two runs? Could you describe the necessary workflow /
> calculation steps? Where / in which program is the setting of a Venetian
> type blind set, for example? Based on what?****
>
>  ****
>
> Best****
>
> Achim****
>
>  ****
>
>  ****
>
>  ****
>
> *Von:* esp-r-bounces at lists.strath.ac.uk [mailto:
> esp-r-bounces at lists.strath.ac.uk] *Im Auftrag von *Germán Molina Larrain
> *Gesendet:* Dienstag, 7. Mai 2013 01:07
> *An:* esp-r at lists.strath.ac.uk
> *Betreff:* [esp-r] Calculation of solar gains through complex shading
> devices****
>
>  ****
>
> Hello, ****
>
>  ****
>
> I am German Molina, and I am doing my MSc. on complex shading systems, and
> how to treat them for simulation purposes (Lighting and Solar Heat Gains).
> I do not really use ESPr or any other thermal simulation tools, just
> Radiance for now.****
>
>  ****
>
> Finished my literature review, I realized that one method used for
> lighting (the Three-phase method) might be good to be a good estimator of
> the Solar Heat Gains through complex fenestration systems... later I found
> out that some people did this, and actually nowadays something very similar
> can be implemented in EnergyPlus.****
>
>  ****
>
> Long story short, I think (it is waiting for a validation) that I can
> accurately calculate the Solar Heat Gains through virtually any complex
> fenestration system of known BSDF matrices; and not only net solar heat
> gains, but I think I can also say where the solar radiation is going
> (calculate the fraction that goes to the ceiling, to the floor, etc.). How
> difficult would it be to use this output as an input for ESPr? For what I
> understand, ESPr does not handle very well these kind of systems?****
>
>  ****
>
> What I am proposing can be actually reduced to just one phrase: Lets use
> Radiance to what it does best (radiation transport) and ESPr to what I
> think it does best (Thermal simulation).****
>
>  ****
>
> THANKS VERY MUCH****
>
>  ****
>
> German Molina****
>
> ** **
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.strath.ac.uk/archives/esp-r/attachments/20130507/fb681b46/attachment.html 