Question-and-Answer Resource for the Building Energy Modeling Community
 | 1 | initial version |
Quantifying glare through simulation is very difficult. Hell, it's hard to do in any empirical manner even in Real Life[TM]. The OpenStudio Radiance measure attempts to tease out glare potential by reporting a variety of so-called dynamic daylight metrics. One, Useful Daylight Illuminance (UDI), is a metric based on horizontal illuminance that awards for illuminance between 100 and 3,000 lux. The idea being that illuminance over 3Klux is going to potentially cause glare (and thermal comfort) issues. THe other way you could do this in OpenStudio is to place glare sensors in your model, and aim them at the facade you're concerned about. These sensors will report the "Simplified Daylight Glare Probability" (DGPs) for each sensor. This metric is also based on illuminance, this time it's vertical illuminance (and the sensor should be placed at the prevailing eye level).
Both of these methods are crude proxies for glare, but they can be calculated in a reasonable time frame and so you can evaluate for glare over the course of a whole year, and with the built-in window treatments in the OpenStudio app. Speaking of, the shade in OpenStudio is based on a light colored 5% openness shadecloth by a well-known manufacturer of French origin, now owned by a larger US conglomerate. ;)
Next up, frit. To do this correctly for glare evaluation, you should apply the frit pattern as an actual pattern in Radiance (typically using the mixfunc to apply the opaque or translucent frit over the clear glass). Alternately you could use the glaze script that comes with Radiance to compute relative trans values for a "blended" performance descritpion of your glazing, with frit. The latter option will not model the effect of direct sun coming through the clear glass areas around the frit. (OpenStudio does not support either of these methods directly.)
Last, the shade. The shade modeling options are similar to the frit. You could model the shadecloth as a pattern and apply via mixfunc, or you could generate a BSDF using genBSDF or using LBL's Optics 6 program. These methods would allow you to model arbitrary materials. Unfortunately OpenStudio does not allow you to specify your own shade for use in the annual Radiance simulation. There is a single generic shadecloth that is used when you select "shade" when creating a new shading control in the plugin or the app.
By the way, if interested in learning the details of rolling your own glazing materials for Radiance, I highly recommend Jack Devalpine's excellent blog post, here.
