DATE:

Friday, October 30, 2020

TIME:

3:30 PM

LOCATION:

Virtual Zoom Conference

SPEAKER:

Dr. Pavel Popov, Aerospace Engineering, San Diego State University

ABSTRACT:

Due to their low computational cost, assumed probability density function (PDF) methods, such as the flamelet-progress variable (FPV) method, are widely used for combustion simulations. To model the subgrid variation of mixture fraction, assumed PDF methods generally use the beta distribution or one of its multivariate extensions. This work addresses a shortcoming of the beta distribution when applied to this setting. Regardless of parameter values, the beta distribution contains the entire sample space of a conserved scalar, which can be shown to violate the maximum principle of elliptic partial differential equations. To remedy this, a new class of PDF is introduced and tested on a conserved scalar mixing problem. The new PDF do not violate the maximum principle, and are in better quantitative agreement than the beta distribution with a transported PDF solution, which is inherently more accurate than assumed PDF solutions. The effects of residence time are examined, and extensions to multiple mixture fractions are outlined.

HOST:

Jose Castillo

VIDEO: