| In order to predict radiative heat transfer accurately in many practical turbulent flows, it is necessary to consider the effects of turbulent fluctuations on the radiation calculations. However, to date virtually models of radiating reactive flows have ignored turbulence radiation interactions (TRI), i.e., radiation calculations have been based on mean temperature and concentration fields, although experimental work has suggested that mean radiative quantities may differ significantly from those based on mean scalar values. In this study, it is demonstrated that probability density function (PDF) methods can be successfully used in the study of TRI. The first half of the thesis aims at improving the current solution approach of the PDF transport equation. An effective particle tracing scheme on structured/unstructured grids in hybrid finite volume/PDF Monte Carlo methods is proposed, in which an adaptive time step splitting method is introduced. The new scheme is able to reduce statistical errors and save significant amounts of cpu time. In addition, it allows the PDF/Monte Carlo code to use any grid that is constructed in the finite volume code. The second half of work aims at studying turbulence-radiation interactions. Firstly, radiation and chemical reaction submodels are discussed. A state-of-the-art gas spectral radiation model as well as a finite-rate reaction mechanism is used to capture the complex physics of turbulent reactive flows. The code is then carefully validated before it is used to study turbulence-radiation interactions. Many different simulations are carried out to study different effects of radiation and turbulence-radiation interactions. The importance of different TRI-related correlations and the importance of TRI in the prediction of NO emission levels is also investigated in this study. |
