Efficient Engineering CFD Modeling of Oxygen-Enriched Staged Ammonia Swirl Combustion Using a Nonadiabatic Dynamic Adaptive FGM Framework

Abstract Existing numerical methods for staged ammonia combustion incur prohibitively high computational costs, limiting their applicability to engineering calculations. To address this limitation, this work develops an engineering-oriented, nonadiabatic dynamic adaptive flamelet-generated manifold (DA-FGM) method within a Reynolds-averaged Navier–Stokes (RANS) framework to enable computationally efficient analysis of staged ammonia combustion. By incorporating localized heat-loss effects through enthalpy mapping and validating against experimental and higher-fidelity numerical results, the framework is applied to investigate oxygen-enriched ammonia swirl combustion under an axially staged configuration. Across oxygen enrichment levels of 21%–35%, at a fixed primary equivalence ratio of Φpri = 1.2 and global equivalence ratio of Φglob = 0.4, oxygen enrichment is found to enhance overall flame stabilization and modify multistage combustion behavior through coupled thermo-aerodynamic effects. Thermally, increased oxygen concentration (XO2) promotes fuel-rich pyrolysis and accelerates the formation of H2 intermediates that dominate subsequent stages. Aerodynamically, reduced secondary jet momentum weakens local shear levels and alters the turbulent strain rate distribution, thereby reshaping flame stabilization characteristics. Under the present protocol with constant power and fixed Φglob, increasing XO2 from 21% to 35% leads to a monotonic reduction in NO emissions while increasing NH3 and H2 slip. This behavior arises from a competition between enhanced downstream thermal DeNOx activity in the core flow, driven by transported NH3 and elevated temperatures, and incomplete conversion in near-wall regions where heat loss suppresses local reactivity. The results highlight the critical role of thermal management in sustaining high-efficiency NH3 combustion under oxygen-enriched conditions.