A Chemical Reactor Network methodology for estimating NOx emissions under non-premixed lean hydrogen combustion with water injection

This study investigates the ability of Chemical Reactor Network (CRN) methodologies in predicting NOx reduction through water injection in a lean non-premixed hydrogen combustor for a given outlet flame temperature. A new methodology for creating a chemical reactor network to analyze emissions is proposed and validated. The novelty lies in the definition of the connections between the different reactors composing the network using an optimization algorithm. This approach allows the prediction of emissions without the need for complete knowledge of the flow field of the combustor through PIV or high-fidelity CFD simulations. The proposed method proves to be a reliable and computationally efficient framework for predicting hydrogen combustion emissions, achieving thousand-fold computational savings over CFD while maintaining prediction accuracy on the NOx emissions within experimental uncertainty. The results are validated through comparisons with CFD simulations, which are first assessed against experimental data without water injection, demonstrating CFD’s effectiveness in this scenario. The CRN, trained without water injection, successfully applies to the water injection cases, demonstrating its predictive capability beyond the training conditions.