A Stochastic Simulation of an HCCI Engine Using an Automatically Reduced Mechanism

Abstract

Abstract Homogeneous Charge Compression Ignition (HCCI) Engines are a promising alternative to the existing Spark Ignition Engines and Compression Ignition Engines. In an HCCI engine, the premixed fuel/air mixture ignites when sufficiently high temperature and pressure is reached. The entire bulk will auto-ignite at almost the same time because the physical conditions are similar throughout the combustion chamber. Therefore it is a justified assumption to consider the chemical reactions to be the rate-determining step for the ignition process. This gives us the opportunity to formulate a simple zero-dimensional model with detailed chemical kinetics for the calculations of the ignition process. Ignition calculations using this model have predicted a high sensitivity to fluctuations in temperature and fuel compositions. These predictions have later been confirmed by experiments. Partially stirred plug flow reactor (PaSPFR) can be used to conquer the assumption of homogeneity. The assumption is replaced by that of statistical homogeneity and thus statistical fluctuations caused by inhomogeneities can be studied. However, the CPU-time needed for this approach is increased considerably and the usage of mechanism reduction becomes evident. In this paper, we demonstrate how a reduced mechanism for natural gas as fuel is derived automatically. The original mechanism by Warnatz (589 reactions, 53 species) is first reduced to a skeletal mechanism (481 reactions, 43 species). By introduction of the quasi steady state assumption, the skeletal mechanism is reduced further to 23 species and 20 global reactions. The accuracy of the final mechanism is demonstrated using the stochastic reactor tool for an HCCI engine.