Approximate Deconvolution and Explicit Filtering For LES of a Premixed Turbulent Jet Flame

Abstract

A novel approach is discussed for Large Eddy Simulation (LES) of premixed turbulent combustion, which is valid with high order numerics along with grid resolution providing quite resolved scalar signals. It is based on an approximate and discrete deconvolution of density weighted scalars so that unclosed terms can be estimated on the LES mesh, to be subsequently explicitly filtered. The method is first tested for a one-dimensional laminar filtered flame, using a single-step Arrhenius reaction rate. Two options are proposed and compared for the fluxes, a full deconvolution, or, the application of a dynamically computed corrective factor to the diffusive flux computed from the resolved scalar signal. Then, the approximate deconvolution/filtering of the burning rate is applied to three-dimensional LES of a turbulent Bunsen flame (Chen et al. 1996), using tabulated detailed chemistry. Most of the statistical properties of both velocities and scalars are reproduced. The modeling procedure is free from any adjustable parameter, aside from numerical interpolation and/or high-order damping of scalar gradients, and can be readily applied to any chemical scheme, opening new perspective for turbulent combustion modeling, specifically to address intermediate species and pollution with LES.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Sep 19, 2014
Accession Number
AD1006441

Entities

People

  • L. Vervisch
  • P. Domingo

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Burning Rate
  • Chemical Reactions
  • Chemistry
  • Combustion
  • Computational Fluid Dynamics
  • Diffusion
  • Equations
  • Flow
  • Jet Flames
  • Kinetic Energy
  • Large Eddy Simulation
  • Measurement
  • Navier Stokes Equations
  • Payload
  • Simulations
  • Three Dimensional
  • Viscous Flow

Fields of Study

  • Physics

Readers

  • Combustion science or combustion engineering.
  • Finite Element Method (FEM) for solving Partial Differential Equations (PDEs)
  • Ocean-Atmosphere Mesoscale Modeling, Data Assimilation, and Flux Boundary Layers