Adaptive scalar transport combustion modeling with finite rate chemistry and flamelet progress variable

This paper proposes an adaptive scalar transport (AST) model for simulating general reacting flows. AST dynamically groups chemical scalars, assigning them to be either solved with detailed transport equations or represented by low-order manifolds. The grouping criterion is tested with two examinations: a drift-difference metric, which gauges the error between scalar transport and manifold representation, and an assessment of each scalar's strain sensitivity. The model also incorporates relaxation procedures for species reassignment and corrections for element conservation. A detailed hybrid algorithm is implemented with treatments ensuring numerical stability during the switch between transported and manifold-represented groups. The model is validated on a multi-modal laminar triple flame, using finite rate chemistry (FRC) and flamelet progress variable (FPV) as sub-models. A sensitivity analysis of the drift-difference parameter demonstrates that predictive accuracy for key species is maintained over a considerable threshold range, provided that the number of transported scalars is intermediate between pure FPV and FRC. The results show that AST significantly reduces computational cost compared to FRC while maintaining the fidelity comparable to transporting scalars, especially for species with long chemical time scales. Throughout the simulation, AST dynamically adapts its species division until the flame stabilizes. The developed AST algorithm provides a multi-modal combustion model to achieve both accurate and efficient reacting flow simulations and is potential to extend to turbulent combustion modeling.