The Comparison of Experimental and Predicted Flame Temperature of Natural Gas Combustion

This present work was focused on the application of the probability density function method in conjunction with the k-epsilon turbulence model in the goal to predict temperature fields of a non-premixed confined natural gas diffusion flame. A combined experimental and computational fluid dynamics modeling study of the turbulent non-premixed natural gas on a laboratory scale has been performed. An experimental study has been carried out on a fire-tube water heater. The flame and gas temperature in the combustion chamber were measured. The real geometry of the fire-tube water heater and the burner were modeled and analyzed by FLUENT code numerically. Turbulent diffusion flames were investigated numerically using a finite volume method for the solution of the conservation and reaction equations governing the problem. The boundary conditions were specified as the same as the measured values. The elemental analysis of the natural gas was taken as the mixture of hydrocarbon, and the air was the oxidizer. The standard k-epsilon model was used for the modeling of the turbulence phenomena in the combustor. The non-premixed combustion model was chosen. In the conserved scalar approach, turbulence effects are accounted for with the help of an assumed shape probability density function or probability density function. The discrete ordinates radiation model is used for modeling of the radiative heat transfer in the combustion room. The model results were compared with the experimental results, in good agreement with the measurements.