The use of efficient analytically reduced chemistry for industrial-scale LES of gas turbine combustors

Accurate yet computationally efficient treatment of combustion chemistry remains a major challenge in large-scale reactive-flow simulations, particularly in industrial applications where long transient calculations and high spatial resolution are required. Analytically Reduced Chemistry (ARC) addresses this challenge through a multi-step reduction strategy combining skeletal reduction, species/reaction lumping, and Quasi-Steady-State Assumption (QSSA) in order to reduce both the number of transported species and the stiffness of the chemical source-term system while retaining kinetic fidelity. This study presents the development, validation, and first integration of an ARC mechanism within the ANSYS® Fluent commercial solver, which is widely used as a Large-Eddy Simulation (LES) reference tool for gas turbine combustor design and analysis. The mechanism is specifically developed for industrial natural-gas combustion applications involving elevated-pressure operation, hydrogen enrichment, and diluted-reactant conditions. The reduction workflow, supported by intermediate validation in Cantera, is outlined, followed by the implementation of the ARC formulation within Fluent’s species-transport framework. ARC accuracy is assessed against experimental laboratory flame data, while stiffness and computational cost are evaluated against skeletal mechanisms on an industrial test rig. Results show that ARC maintains strong predictive capability while markedly reducing the dimensionality and stiffness of the transported chemistry system, yielding substantial gains in numerical robustness and computational efficiency that make it suitable for high-fidelity industrial reactive LES workflows.