Computational evaluation of mechanisms affecting radiation in gas- and coal-fired industrial furnaces

A computational evaluation of mechanisms affecting radiation in gas- and coal-fired industrial furnaces was conducted with the purpose of determining the influence of individual mechanisms on the radiation and combustion processes. Computer models were used to gain insight into the behavior of radiation in furnace combustion and to quantitatively identify mechanisms that had the greatest impact on surface heat transfer and radiative flux divergence. Mechanisms considered were surface emittance and temperature, gaseous absorption and emission, particle absorption, emission and scattering, soot, and turbulence-radiation coupling.; The discrete-ordinates method was selected for modeling the radiative transfer equation based on its applicability to three-dimensional absorbing-emitting-scattering systems, its accuracy in predicting anisotropic intensity fields and surface heat transfer in complex geometries, and its relative computational efficiency. The method simulated highly directional local intensities and anisotropic scattering and accurately predicted incident surface fluxes in a gas-fired industrial-scale furnace and a coal-fired utility boiler. The S{dollar}sb4{dollar} approximation order gave the best combination of accuracy and computational efficiency.; Mechanisms that most affected radiation were medium and surface temperatures; effects of wall emittance, nonlinear turbulence coupling, and soot formation in gaseous, fuel rich systems were second order to temperature. Gas absorption and emission and particle absorption, emission and scattering had the least effect on radiation. Use of forward, backward and isotropically scattering phase functions produced little difference in radiative flux results. Spectral effects of gas properties were accounted for using a hybrid model comprised of real gas models and total absorptivity-emissivity models calibrated with spectral data. Particle absorption and scattering coefficients were based on efficiencies calculated from Mie theory using spectral refractive indices from the literature. Time-averaged radiative transfer and flux divergence equations accounted for effects of flow-induced turbulent fluctuations. Nonlinear dependencies of gas properties were accounted for using statistical relations for local instantaneous stoichiometry and enthalpy available from the turbulent combustion model. Mean particle properties were ensemble-averaged based on local instantaneous particle statistics. Nonlinear turbulence effects significantly affected the estimation of mean properties. Overall results suggest many stoichiometric-to-lean-firing coal-fired furnaces can be modeled with good accuracy by including nonlinear turbulence effects and neglecting scattering.