Application of the explicitly iterative scheme to simulating subsonic reacting gas flows

This paper is devoted to the study of the possibility of applying an explicitly iterative (local iterative modified–LI-M) scheme for calculating dissipative terms in the solution of problems of subsonic reacting flows with radical chain reactions, active diffusion processes, significant heat transfer, and energy absorption. Simulation of such flows is characterized by a restriction on the integration time step, primarily due to the predominance of diffusion processes over convective ones and the presence of rapid chemical reactions. The mathematical model is described using the multicomponent Navier–Stokes equations. The combination of nonuniformy scaled processes in the model led to the use splitting by physical processes—chemical kinetics is integrated by the Radau method with adaptive time step, the convective flow is calculated using the Rusanov flux and WENO scheme, and dissipative flows are calculated using the explicitly iterative LI-M scheme. As a result, a numerical algorithm and computer code for studying subsonic reacting flows are developed and some computational experiments are carried out. A one-dimensional nonstationary inhomogeneous equation was solved to test the implemented algorithm. It is shown that the application of the LI-M scheme to the calculation of the dissipative part makes it possible to get rid of the diffusion restriction on the integration time step. Numerical simulation of the process of methane high-temperature conversion in axisymmetric geometry is carried out. This process is characterized by rapid chemical reactions, significant local changes in temperature, gas density, and thermophysical characteristics, which imposes significant restrictions on the integration time step. It is shown that the proposed algorithm makes it possible to perform calculations with a step exceeding the diffusion restrictions on the time step. The calculations are compared with calculations using a previously verified algorithm, and a good coincidence of the results with a significant gain in program execution time is demonstrated. Numerical simulation of the gas flow in a cylindrical pipe is carried out, and the results are verified by demonstrating grid convergence.