Flameless combustion technology has been developed over the past twenty years achieving low-NOx emissions and high energy efficiency for industrial applications. In the present work, three aspects of flameless combustion were examined based on a burner employing the Strong-Jet/Weak-Jet (SJ/WJ) configuration. In the first part of the work, a 3-D SJ/WJ physical model was developed in the Lagrangian perspective for an isothermal pair of free jets. The model was used to predict the WJ trajectory, identify important design/operation factors, and estimate the extent of mixing in the main combustion region (confluence region). The model was also validated with experimental data and showed excellent agreement over a wide range of flow conditions. In the second part of the work, a simplified chemical kinetic model was developed for the flameless combustion of natural gas. A detailed chemical reaction mechanism (GRI Mech 3.0) was successfully reduced to a skeletal chemical reaction mechanism under flameless combustion conditions by Principal Component Analysis, sensitivity analysis and reaction flow analysis. The skeletal mechanism was further simplified to a set of 2-D manifolds by Trajectory-Generated Low-Dimensional Manifolds (TGLDM) method. The set of 2-D manifolds was tested by the Batch Reactor (BR) and Perfect Stirred Reactor (PSR) models. From the BR model test, it was found that the chemical reaction rates were well represented by the 2-D manifolds. The effect of the physical perturbation, tested by PSR model, could be handled by the perpendicular projection instead of the orthogonal projection because both showed similar discrepancies with the skeletal mechanism. In the final part of the work, the steady-state Reynolds-Averaged Navier-Stokes (RANS) simulation was conducted for the turbulent flameless combustion in the SJ/WJ furnace, based on the Probability Density Function (PDF)/Mixing approach. The set of 2-D manifolds and Conditional Source-term Estimate (CSE) method were used for the combustion reaction and the estimation of the mean production/destruction rate, respectively. This CSE-TGLDM model provided good predictions of major species concentrations. However, the gas temperatures and CO concentrations were highly over-predicted. / Thesis (Ph.D, Chemical Engineering) -- Queen's University, 2010-09-23 11:05:21.884