Direct numerical simulation (DNS) is a productive research tool in combustion science used to provide high-fidelity computer-based observations of the micro-physics in turbulent reacting flows. It is also a unique tool for the development and validation of reduced model descriptions used in macro-scale simulations of engineering-level systems. Because of its high demand of computational power, current state-of-the-art DNS remains limited to small computational domains, small Reynolds numbers, and simplified problems corresponding to adiabatic, non-sooting, gaseous flames in simple geometries. This Ph.D. study is part of a multi-institution collaborative research project aimed at using terascale technology to overcome many of the current DNS limitations.

Two different tracks are followed in the present work: a DNS development track, and a DNS production track corresponding to a study of flame-wall interactions. Due to project management issues, the two tracks remain separate in this work. In the first track, we develop numerical and physical models to enhance the capability of our fully compressible DNS solver for turbulent combustion. The Acoustic Speed Reduction (ASR) method is a new perturbation method designed to reduce the stiffness associated with acoustic waves found in slow flow simulations and to thereby enhance computational efficiency. The Navier-Stokes Characteristic Boundary Conditions (NSCBC) are modified to allow for successful simulations of turbulent counterflow flames. In addition, a semi-empirical soot model and a parallel thermal radiation model based on a ray-tracing method are developed and implemented into our DNS code. All the models are validated, showing that the capability of our DNS tool is greatly enhanced.

In the second track, we perform a DNS study of non-premixed flame-wall interactions. The structure of the simulated wall flames is studied in terms of a classical fuel-air-based mixture fraction and a new variable, called the excess enthalpy variable, which characterizes deviations from adiabatic behavior. Using the excess enthalpy variable, a modified flame extinction criterion is proposed and tested against DNS data.

While beyond the scope of this Ph.D. thesis, it is expected that follow-up studies of flame-wall interactions will take advantage of the new DNS software features developed in the first track of the present work.