A New Paradigm in Modeling and Simulations of Complex Oxidation Chemistry Using a Statistical Approach
Abstract
The computationally accurate and efficient prediction of turbulent reactive flows remains an important research topic for several reasons, one of which is that the modeling of chemistry is still cumbersome, relatively inaccurate, and cannot portray the complex reactions associated with real fuels. To reduce computational costs and enable the description of concurrent reactions from complex fuels containing a multitude a species, a new concept is here proposed that relies on statistical instead of deterministic concepts for modeling chemical kinetics of oxidation reactions. In this new concept, each species has a time scale associated with it, and the goal is to solve only for the significant scales of the problem and model those scales that are not significant. A set of reaction-species mole fractions is considered a vector and the objective is to map this vector into a much smaller multidimensional vector space of base chemical fragments. The goal is to inquire whether such a base can be found for oxidation reactions, and whether one can derive corresponding reaction-coordinate rates for the base that portray the energetics of the system. In essence, there is a functional mapping representing the transformation from the mole fraction PDF of the species to the mole fraction PDF of the base and the interest is not in calculating this functional but its moments (e.g. statistical means, variances, etc.) needed to adequately describe the fragment vector.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 01, 2006
- Accession Number
- ADP023628
Entities
People
- Josette Bellan
Organizations
- California Institute of Technology