ILDM Reaction Modeling in Turbulent Combustion Simulations

Diem-Phuong Nguyen, University of Utah

In numerical simulation of turbulent pool fires, the purpose of reaction modeling is to computationally link large chemical kinetic mechanisms to turbulent combustion computations. In pool fires, radiation is the dominant mode of heat transfer. Radiation is dependent on mixture absorption, emission, and scattering; hence, radiation is strongly affected by soot. Soot formation and destruction involve large chemical kinetic mechanisms, making chemistry an integral part of fire simulation. Chemical reactions in a fire are comprised of thousands of elementary steps and hundreds of species with time scales ranging from picoseconds to seconds.

Small-scale, realistic chemistry is incorporated into large-scale components of the fire simulation through a subgrid scale (sgs) Intrinsic Lower Dimensional Manifold (ILDM) model. ILDM is based on the concept that, for a given state space (fixed mixture fraction, f, and enthalpy, h), reaction trajectories quickly converge onto a 1-D manifold that can be represented by three parameters: f, h, and an extent of reaction parameter. The reaction pathways are calculated from the detailed chemical kinetic mechanism. Thus, both gas phase and solid phase kinetics are integrated into the computational fluid dynamics (cfd) code.

The benefits of implementing the ILDM reaction model are both computational and scientific. Computationally, the sgs framework allows widespread parallelization of the reaction model calculation because it requires as input only the mean values of the reaction model parameters, computed on the mesh, for a single grid cell. Therefore, the chemistry of each cell can be calculated independently of what goes on in neighboring mesh cells.

Scientifically, ILDM bridges microscopic details to the macroscopic domain by providing information about the molecular reactions at a scale below the level of the mesh. Soot predictions based on detailed gas phase kinetics of high molecular weight hydrocarbons are a first in large-scale turbulent pool fire simulations.

Abstract Author(s): Diem-Phuong N. Nguyen<br />Advisor: Philip J. Smith