Application of null-collision backward Monte Carlo algorithm to digital image rendering of sooting flames in the visible range

The objective of this work is to present a new tool to produce a digital image computation inside a sooting flame using Monte Carlo methods. The studied case consists of a laser emission corresponding to a spatially localized source of collimated radiation, that partially illuminates the environment. The laser emission consequently causes problems of convergence with a Monte Carlo method due to the difficulty in statistically linking sensors to sources. A new algorithm has been proposed to calculate the radiative intensity at a given probe position. An open source tool has been developed based on a combination of Null Collision Algorithms (NCA) with hierarchical grids technics from the computer graphics community, allowing to perform efficient radiative transfer (and possibly produce images). These tools are generic and are able to accept any input data such as, for instance, temperature and concentration fields produced by a Large Eddy Simulation (LES) approach over a tetrahedral grid. We show in this paper a first monochromatic digital image of a laminar sooting flame in the visible range. Such a new tool can for instance provide a useful support for experimental characterization of the radiative behavior of soot aggregates in flames, such as tracking effects of scattering.