1D engine modeling with detailed reaction kinetics

Karin Fröjd, Cathleen Perlman, Adina Tunér, Fabian Mauss

Abstract

Numerical investigations of the response of fuel consumption and exhaust emissions on
internal-combustion engine parameters or transient-load cycles are needed for the
analysis and optimization of the complete powertrain system. For this purpose, simple
and numerically fast but still accurate models are needed for in-cylinder combustion and
for exhaust after-treatment. The chemical processes must be represented in sufficient
detail to predict emission levels of soot, NOx, CO and unburned hydrocarbons. This work
reports on a network of transient one-dimensional reactors invoking detailed chemistry
for the accurate emission prediction throughout the engine powertrain.
Our approach is based on the stochastic reactor model (SRM) for the simulation
of all in-cylinder processes. Stochastic reactor models assume statistical homogeneity of
the gas charge. The SRM model is suitable to predict combustion and pollutant formation
in SI and Diesel engines including unwanted combustion, such as engine knock. The
network makes further use of stochastic perfectly stirred reactor models for pipe sections.
This allows modeling inhomogeneities in the exhaust gas. Models for catalysts and for
particulate filters are available, invoking detailed surface chemistry or simplified global
chemistry.