A systematic study on the applicability and limits of detailed chemistry based NOx models for simulations of the entire engine operating map of spark-ignition engines

Abstract

The scope of the present investigation is the systematic assessment of the applicability and the limitations of a multi-zone engine model combined with a detailed chemistry solver for NOx prediction. The results are evaluated based on an extensive data set from test-bench experiments throughout an entire engine operating map.Modern engine development, as any optimization process, needs to attain different target values as, for example, engine performance and driveability. Additionally, due to strict governmental regulations, reducing pollutant emissions has become a major issue for the engine design process. Therefore, this target value has to be investigated in very early stages of engine development already. Here, a fast yet predictive computational model is applied, retaining reliable system responses simulating a wide range of operating conditions.The numerical model consists of two parts. The first part is a pressure analysis, which post-processes the experimental data of the intake, cylinder and exhaust pressure for use in the second sub-model. The latter is a multi-zone cylinder model coupled to a detailed chemical kinetic solver to investigate NOx emissions. For the chemical kinetic investigation, we compare 5 different iso-octane mechanisms, which describe the fuel oxidation with different levels of complexity, as well as 11 sub-mechanisms for NOx formation. The model shows good agreement with the experimental results for most of the engine map. Areas of the engine map with larger deviations are analyzed based on 3D CFD data.This approach can be considered a first step for fast and reliable transient simulations of the engine operating map, which are not feasible to be carried out using full 3D detailed modelling studies.