Assessment of the CSP Solver Performance in Accelerated Reacting flow Simulations with Stiff Detailed Chemistry

Numerical simulations of chemically reacting flows pose significant computational challenges due to the stiffness and large dimensionality of the governing equations. A recent study showed that the computational singular perturbation (CSP) solver enables the acceleration of detailed-chemistry direct numerical simulations (DNS) of premixed ammonia–air flames. By applying the operator-splitting technique, the CSP solver removes stiffness on-the-fly and numerically integrates a non-stiff reduced order model (ROM) of the thermo-chemical operator. This allows to advance the thermo-chemical state with an explicit numerical scheme making use of larger integration time steps. In this work, the approach is applied in the direct numerical simulation of a two-dimensional ethanol–air flame–vortex interaction, achieving a tenfold speedup of the chemistry integration compared with conventional backward differentiation formula (BDF) methods and demonstrating that the CSP solver yields greater acceleration when stiffer chemical kinetics are involved.