A reactor network model for predicting NOx emissions in an industrial natural gas burner

A chemical reactor network (CRN) is developed and applied to the modeling of a natural gas burner. The CRN development is based on experimental and CFD simulation results of the BERL 300 kW test. The CRN element arrangement, reactor volumes, and flow splits are adjusted based on the best agreement with characteristic temperatures of the reactive flow, aiming to reproduce the experimental NOx emissions data. A sensitivity analysis and a parametric study of the CRN are carried out to evaluate its sensitivity to the flow splits adjustments and its ability to predict emission with air preheat and turndown operation, as well as the influence of the reaction mechanism employed. The best agreement was obtained in the air preheat simulations using Konnov 0.4 mechanism, while the GRI-Mech 3.0 is accurate only within 110 K preheating. The turndown operation results are relatively accurate for turndown ratio between 1.0 and 1.3.