Abstract

Each year fires destroy millions of acres of woodland, lives, and property, and significantly contribute to air pollution. Increased knowledge of the physics and properties of the flame propagation is necessary to broaden the fundamental understanding and modeling capabilities of fires. Modeling flame propagation in fires is challenging because of the various modes of heat transfer with diverse fuels, multi-scale turbulence, and complex chemical kinetics. Standard physical models of turbulence like RANS and LES have been used to understand the flame behavior, but these models are limited by computational cost and their inability to resolve sub-grid scales. Application of several other models and empirical studies in fire modeling are usually limited to fire spread rate only. In some fires, flame propagation often occurs through convective heating by direct flame contact as opposed to radiative preheating alone. Under these conditions, resolution of the flame front can provide the detailed physics and insights into the flame propagation. The One Dimensional Turbulence (ODT) model is extended to turbulent flame propagation in biomass fuel beds representative of those in wild land fires. ODT is a stochastic model that is computationally affordable and can resolve both large and fine scales. ODT has been widely applied to many reacting and non-reacting flows like jet flames and pool fires. A detailed particle combustion model has been developed and implemented in the ODT model to investigate the fluctuating flame-fuel interface and to study flame propagation properties. The particle reaction is modeled as a single global decomposition reaction model. Radiative, convective, and internal particle conductive heat transfer are included. Gaseous combustion is modeled with a lookup table parameterized by mixture fraction and fractional heat loss using steady laminar flame let solutions. Results are presented from simulations of flame propagation in buoyantly driven flows. Particle size and loading are varied to study their effects in flame spread. A timescale analysis is performed to compare radiative, convective, conductive, and reactive particle time scales to the turbulent fluctuations. The flame propagation in homogeneous turbulence is also studied which better represents the wildland fire. The time scales involved in the wildland fire are overlapped using LEM model to study their effects on the flame properties and flame spread.

Degree

College and Department

Ira A. Fulton College of Engineering and Technology; Chemical Engineering

Rights

http://lib.byu.edu/about/copyright/