You are here

EvalDRC: A tool for annual characterisation of daylight redirecting components with photon mapping

TitleEvalDRC: A tool for annual characterisation of daylight redirecting components with photon mapping
Publication TypeConference Paper
Year of Publication2015
AuthorsSchregle R, Bauer C, Grobe LO, Wittkopf SK
Conference NameCISBAT 2015 International Conference on Future Buildings and Districts
Date Published09/2015
Conference LocationLausanne
ISBN Number978-2-9701052-2-0
Keywordsphoton mapping, raytracing, redirection, spatial daylight autonomy

Annual simulation is a significant indicator of a daylight redirecting component’s per- formance, since it accounts for seasonal variations in daylight availability as well as the system’s response under such conditions. This study details the simulation of a representative redirecting component using a 3D forward raytracing technique to assess its annual daylighting performance. We streamline and largely automate this workflow with the EvalDRC tool, a Python script which implements a simulation frontend based on the Radiance photon map, coupled to a postprocessing and evaluation backend. The redirecting component selected for our case study combines retroreflection with redirection and is designed for optimal daylight availability over the entire year without the need for adjustment. The lamella profile can be mounted in a forward and reversed configuration to combine retroreflection with redirection in the lower resp. upper portions of the fenestration. We evaluate our simulations visually and numerically as high dynamic range (HDR) renderings and a spatial daylight autonomy (sDA) metric based on climate based sky distributions for Geneva, Switzerland. Our case study satisfies the sDA requirement that 55% of the workplane receives an illuminance exceeding 300 lux during 50% or more of the occupancy hours for a whole year. In addition, we propose the msDA, a detailed monthly breakdown of the sDA, for which the criteria are specifically met in the months March–September, while a minimum of 32% is predicted for December. Our results demonstrate the effectiveness of photon mapping for this application, and that the simulation accurately predicts the redirecting component’s expected seasonal behaviour for multiple solar angles and sky configurations. This applies in particular to complex redirecting systems which cannot be reliably simulated with a backward raytracer at reasonable computational cost.