A Model for the Prediction of Thermal, Prompt, and Fuel NOx Emissions From Combustion Turbines

Abstract

A model has been developed for the prediction of NOx emissions from combustion turbines. Thermal, prompt, and fuel NO are all treated and are all assumed to be formed at a stoichiometric equivalence ratio. Prompt and fuel NO are assumed to be fast with respect to thermal NO and establish a finite concentration of NO at the beginning of the thermal NO formation process. Thermal NO is calculated via the extended Zeldovich mechanism; a thermal NO formation time is determined from the ratio of flame length to convective velocity within the combustor. Prompt NO is assumed to be formed from the hydrocarbon chemistry and is related to the equilibrium concentration of NO rather than to O-atom overshoot. Fuel NO is calculated assuming an indispensible intermediate in the formation mechanism and a constant Fenimore α parameter for combustion turbine flames. The model employs equilibrium hydrocarbon chemistry; the flame temperature and concentrations of key species are determined in an equilibrium subroutine. The effects of water or steam injection and ambient humidity are included through their impact on the flame temperature and species concentrations. The model has been applied to can-type combustors and its accuracy has been verified by data on high and low nitrogen fuels with and without water injection, and on combustors of different geometry. The treatment of all three mechanisms of NO formation is unique to this model and permits prediction of emissions for the range of conventional and alternative fuels encountered in industrial combustion turbines.