Implementation of Reduced Mechanisms With ISAT Into CFD Simulations of Full-Scale Combustion Systems

Abstract

The prohibitive CPU cost and memory requirements associated with detailed chemistry prompted the use of reduced mechanisms for turbulent reacting flow simulations. In this study, an In Situ Adaptive Tabulation (ISAT) algorithm, which was originally developed in the context of PDF methods of turbulent combustion, was applied to reduce the cost of the reduced mechanisms. Reduced mechanisms and ISAT have been implemented into VULCAN, a NASA CFD code, to simulate high-speed hydrocarbon combustion. A supersonic diffusion flame case was used to test the efficiency of ISAT. For a 20-specie C2H4 mechanism, the total CPU cost was cut by a factor of 26. ISAT has also been implemented into REI’s proprietary multiphase, turbulent reacting CFD code, GLACIER. It was tested with a 10-specie reduced mechanism to model the selective non-catalytic reduction process for NOx control in a coal fired boiler. CPU savings of approximately 40 percent was observed, which is much lower than that was achieved in VULCAN. The difference in CPU savings was attributed to the different numerical methods used in the two codes. The re-use of Jacobian matrices of the chemical source terms in VULCAN resulted in significant reduction of CPU time.