NO_xPrediction in 3-D Turbulent Diffusion Flames by Using Implicit Multigrid Methods

Abstract

Modeling of the nitric oxide formation in the turbulent non premixed methane and syngas (CO/H,/N,) diffusion flame is studied using an implicit time-stepping and multigrid technique. The chemical kinetic model for both methane-air in a sudden-expansion combustor and syngasair combustion in a laboratory combustor as wellas in a gas turbine combustor is assumed to have 49 species and 229 finite-rate, reversible reaction steps. The standard k-£ turbulence model and the algebraic correlation closure model are applied to close the time-averaged Na vier-Stokes and species equations (Liao et aI., 1995)respectively. The computation requires about 250 time steps to reduce the residual by 3 orders of magnitude for the 3-D turbulent methane-air diffusion flame case on a 34 x 18 x 18 grid, which shows convergence rate is much faster than conventional iterative methods. Computational results with detailed chemistry are exhibited and some of them are compared with experimental data. Qualitative agreement between the computational results and experimental results is observed. A three-dimensional calculation for a gas turbine combustor shows the potential of this method for modeling the pollution formation in practical flows.

Kinetic Capillary Electrophoresis (KCE): A Conceptual Platform for Kinetic Homogeneous Affinity Methods

“…As a result, the species, which migrate with velocities different from v , are simulated “out of the grid”. This leads to rounding errors that are severely aggravated by sharp fronts of electrophoretic peaks. , We addressed this problem by adopting the multigrid approach, which was successfully used in other areas of physical sciences. − We designed a multigrid algorithm for solving system 2 with an individual Δ x for every one of the three components: …”