Prediction of the Spontaneous Ignition in a GCI Engine using an Extended Physical Model of the Ignition Delay 2023-32-0019

With the aim to further reduce and limit pollutant emissions and fuel consumption towards carbon neutrality, researchers and automotive manufacturers have been studying new combustion technologies, such as low temperature combustions, which provide an efficient combustion with low pollutant emissions. Despite innovative combustion techniques, such as Homogeneous charge compression ignition (HCCI) and Gasoline compression ignition (GCI), proved to reduce pollutant emissions and increase efficiency of internal combustion engines, their large-scale deployment has been limited by problems in combustion management and stability. In fact, the challenge related to these innovative combustion techniques consists in the development of new control strategies and new calibration methodologies, which allow to limit their combustion instability.

By relying on the natural phenomenon of autoignition, researchers showed that, to optimize GCI combustion, it is necessary to adopt a multiple injections strategy. The rise in temperature and pressure produced by the premixed combustion of the first injections (long ignition delay) results in a controllable ignition of the main injection, responsible for torque delivery, which burns with an extremely short ignition delay. In the discussed process, the most critical aspect to be managed is the spontaneous ignition of the fuel introduced with the pre-injections, since its long ignition delay is affected by many engine control parameters, because it is very sensitive to the cylinder thermal conditions. As a result, the ignition delay might be strongly affected by a slight variation in the control parameters and, consequently, lead to misfire or knocking. The goal of this work was to develop an extended control-oriented ignition delay model, suitable to improve GCI combustion stability through the proper management of the pilot injections over the whole engine operating range.