Abstract
If non-uniform lighting installations are to become accepted design practice it is important for the designer to be able to predict the lighting distribution accurately. This is only practicable by computer, and so predictions of illuminance distributions within a simulated office were made using a sophisticated suite of commercially available programs running on a desk-top computer. These predictions were compared with detailed illuminance measurements. Uniform and non-uniform arrays of recessed and surface-mounted luminaires with a wide range of light output distributions, and also uplighters were considered. Predictions for non-uniform arrays of ceiling-mounted luminaires generally agreed less well with measured values than predictions for uniform arrays. However, in most cases agreement at the majority of points within the room was better than ± 10%, sufficiently accurate for most design purposes. For the uplighter cases agreement was not so good, although almost all points were within ±20%. Care is necessary when modelling upward light from ceiling-mounted luminaires and uplighters; consideration of the adequate subdivision of surfaces is especially important. The need for detailed luminaire data was also demonstrated. The desirability of some form of quality assurance for lighting prediction software is highlighted.
Get full access to this article
View all access options for this article.
