Scalar dissipation rate based multi-zone model for early-injected and conventional diesel engine combustion

Abstract

Low-temperature combustion (LTC) concepts, such as Homogeneous Charge Compression Ignition (HCCI) or Premixed-Charge Compression Ignition (PCCI), have the potential of simultaneous reducing nitrogen oxides (NOx), soot, and unburned hydrocarbons (uHC). However, the successful implementation for internal combustion engines is difficult. Control strategies need to be employed to ensure appropriate combustion phasing over wide ranges of operating conditions. Model-based engine control is particularly successful when physics-based models are employed. Multi-zone combustion models represent potential candidates for efficiently computing combustion in PCCI diesel engines. Multi-zone models were originally developed for HCCI gasoline engine combustion and did not account for mixing between zones due to the relatively homogeneous mixture while later developments considered small inhomogeneities. However, for diesel or PCCI combustion, this is not justified due to noticeable fuel stratification. Therefore, a novel mixing model for multi-zone modeling accounting for mass and energy exchange between zones is presented in this work. The model is derived from the representative interactive flamelet (RIF) model and thus depends on the scalar dissipation rate as well as the mixture fraction in each zone. The multi-zone model can be used as a stand-alone model after performing a number of non-reactive computational fluid dynamics (CFD) simulations to train an empirical, engine-specific model for the scalar dissipation rate. With the stand-alone model, cost-efficient parameter studies can be performed, with further model reduction, the use in model-based control algorithms is also possible. For validation of the stand-alone multi-zone model, experiments were conducted with a four-cylinder diesel engine. 105 operation conditions including variations in start of injection (SOI), injected fuel mass (FMI), and external exhaust gas recirculation (EGR) were selected to assess the model performance. CFD simulations applying the RIF model were carried out for representative cases to further assist in validating and analyzing the new multi-zone model. Predictions of the multi-zone model regarding indicated mean effective pressure (IMEP), combustion phasing (CA50), and emissions of nitrogen monoxide as well as unburned hydrocarbons are compared against experimental data and results from numerical simulations. Overall good agreement over various operating conditions was found demonstrating the capability of the multi-zone model to adequately capture PCCI diesel engine combustion.