Speaker:
Xinfeng
Gao CSU Mech. Engrg.
Title:
A Fourth-order Finite Volume Scheme with Parallel Adaptive Mesh
Refinement for Computational Combustion
Abstract:
This talk focuses on a fourth-order finite-volume scheme for
solving the compressible Navier-Stokes equations and a set of
species transport equations on a patch-based, structured adaptive
mesh refinement grid. For spatial discretization,
finite-volume stencils are derived for the viscous stress tensor
operator. Fourier error analysis and stability analysis are
performed for the fourth-order elliptic operator. For time
integration, we use the fourth-order additive Runge-Kutta method to
cope with the stiff
system due to the viscous diffusion and the
chemical kinetics. At each intermediate stage of the
Runge-Kutta method, we also solve a linear system of Helmholtz-type
equations. This fourth-order finite-volume scheme is for
direct numerical simulation of some combustion problems focusing on
fundamentals and can be extended to perform large eddy
simulations. For application to the practical combustion
devices, we will incorporate the embedded boundary method for
complex geometry. Moreover, we will attempt to maximize
performance on current heterogeneous architectures featuring CPUs
and GPUs. Doing so ensures that the detailed simulations of
combustion remains affordable.
Short Bio:
Xinfeng Gao is an assistant professor in the Department of
Mechanical Engineering at CSU. She leads the CFD and
Propulsion Laboratory and her research focuses on the development
and application of advanced CFD algorithms and the computational
combustion modeling. Together these research eefforts converge
toward the design, analysis, and optimization of practical
low-emission and high-efficiency chemical propulsion systems.
Xinfeng Gao earned her Ph.D. in Aerospace Sciences and Engineering
from the University of Toronto Institute for Aerospace Studies in
2008. Prior to joining Colorado State University, she was a
postdoctoral scientist at Lawrence Berkeley National Laboratory
working with two groups: the Applied Numerical Algorithms Group and
the Center for Computational Sciences and Engineering.