Review of simulating four classes of window materials for daylighting with non-standard BSDF using the simulation program Radiance

From a practical point of view, daylighting in architecture uses the visible spectrum of solar energy directly and efficiently. It offers potential to decrease the energy consumption of a building and enhance the aesthetics of indoor spaces. Optimisation of daylighting can be roughly divided into two approaches, or a combination of the two in one project: First, the shape of the building can be optimised for a specific location and climate, using traditional materials. This is the classical factor of climate and local conditions on architectural shapes [Lec91]. Secondly, the use of new window materials with complex light-scattering and -redirection offers the potential to enhance daylighting with fewer requirements on the building shape. This has been the subject of research over a number of years [Wit90], [Com93], [jou94], [JHG11]. From a physics point of view, the core concept of the physics and numerics of light-scattering and -redirection is the bidirectional scatter distribution function (BSDF) [NRH77], [Sto90]. Lighting properties of window materials, whether structured on a macroscopic scale (embedded mini-louvers between panes) or microscopic scale (e.g. holografic elements), can be described with the BSDF concept fittingly. However, a detailed description of the relations between the physics of the BSDF, the possible advantages of a material for daylighting and the numerical details of algorithms to simulate it would exceed this text and is left to a specific further paper. Tools for simulation of daylighting have to predict accurate lighting levels in real-world applications, including materials with complex scattering and redirecting properties. This sets them apart from general rendering in computer graphics, where materials predominantly have simpler optical properties and ”just have to look right”. Daylight simulations are frequently using the software Radiance , [GR08] found a market share greater 50% market in their 2006 survey. Radiance has been developed since 1990 by Greg Ward, then at Lawrence Berkeley Lab, CA, USA [WH92]. A number of other raytracing programs are established in the optics industry (e.g. for lens design) and other programs, some based on less general Radiosity algorithms, have been commercially available for daylighting. Radiance is still in active use, either a stand-alone tool, in open-source front-ends [Ope13] or in commercial programs [sup11], [sup08]. The core calculation engine of Radiance , having been designed for dayand artificial lighting from the start, offers a number of advantages, for example: support for non-diffuse surfaces, efficient algorithms with comparatively low memory footprint and source code availability. Historically, not all ”legacy” window materials in Radiance are supported in all lighting situations, due to limits of the built-in algorithms. Even recent papers are not specifically detailed about this [RA06].