Small scale simulation of artificial smoke optical properties

Abstract

Abstract The use of numerical simulations in Fire Safety Engineering provides faster and more affordable ways of research and development, with the drawback that their use is often conditioned by available expertise and thorough validation. In this paper, the results of two small scale numerical simulations of artificially generated smoke are being validated by comparison of the smoke’s light extinction coefficient values and those of the visibility levels in the testing enclosures with the results obtained through experimental testing. The influence of grid size on simulation accuracy and computational time is evaluated by presenting the small gain in accuracy and the important increase in computational time for different sized meshes, showing that at least in numerical simulations dealing with artificial smoke, a larger grid size can produce fairly accurate results. A comparative evaluation of the effect that the positioning of the measuring devices inside the testing enclosure has on the results is also performed, both for the experimental setup and the numerical simulations.