Cost effective emissions and minor species predictions via coupling of computational fluid dynamics and chemical reactor network analysis

The progress in recent years and the advent of new powerful computers have allowed experts to simulate combustion-turbulence interaction reasonably well and predict temperature and velocity fields with acceptable accuracy. However, the current technology and available computer power do not suffice in predicting the concentration of minor species such as NO x and CO. As the reduction of these pollutants requires expensive experimentation, much attention has been directed towards more cost effective ways of simulating pollution emission via numerical methods. This research project has been conducted in order to obtain a universal cost effective method for predicting emissions via a system of Chemical Reactor Networks (CRN). This was achieved via coupling of CFD-CRN. While CFD provided temperatures, residence time and major species' concentrations, CRN was able to accurately tackle the complex chemical kinetics for prediction of minor species on a personal computer; A task which would have taken months via CFD on a computer cluster. RANS and LES simulations of an industrial Rolls Royce RB211 combustor were performed with and without Discrete Phase Modeling. CRNs were then extracted from the CFD field based on temperature, composition and geographical location via an efficient coded algorithm. It is demonstrated that the chemical kinetic computation based on the extracted CRNs from CFD provides reasonable results compared with experimental data on some of the CO predictions. It is strongly believed that higher resolutions of reactors will at least provide reasonable trends upon boundary condition variations. The algorithm developed in this study leads to a more universal approach for cost effective prediction (quantitatively or qualitatively) of combustion emissions, which contribute to cardiovascular and respiratory ailments.