A Heat Release Model Of Gasoline CAI Combustion Based On Group Effect Of Flame Kernels

Gasoline CAI (Controlled Auto-Ignition, CAI) combustion is a promising combustion mode with advantages of high efficiency and low emissions Compared to traditional gasoline spark ignition combustion, CAI is characterized with multi spontaneous auto-ignition and will be significantly affected by temperature and fuel stratification. The current CAI heat release models are limited due to the absence of physical interpretation on the CAI combustion process. Hence, in this research, a model has been built to predict the heat release of gasoline CAI combustion, which is more practical for its potential to characterize the physical process of CAI combustion.Due to the stratification of temperature and composition, the auto-ignition sites occur and develop sequentially, leading to rapid spontaneous ignition. During the combustion process, strong interactions exist between auto-ignition sites. The phenomenon above matches well with what describes in the Swarm Intelligence Theory (SIT). The main idea of SIT is that the development of simple individual behavior can be altered by the interactions, which lead to group phenomenon eventually. SIT can be used to build phenomenological models. In this research, auto-ignition sites are defined as the flame kernels. CAI combustion is described as the flame kernels’collective behavior. As a consequence, the idea is proposed to build a CAI heat release model based on group effect of flame kernels.Based on the three-dimensional simulation, the CAI combustion process can be associated with the flame kernel group development with no chemical reaction kinetics taken into consideration. A function that reasonably describes the multi spontaneous combustion has been developed to represent the development of the number of flame kernels. The function together with an ignition criterion constitute the new CAI heat release model which is named as CAI combustion model based on flame kernel group effect. The test data with relative air/fuel ratio between1-4and residual gas fraction below70%was used to determine the model coefficients. The simulation results matches well with the experimental data which confirms that the model is able to predict heat release of gasoline CAI combustion under different operation conditions. Meanwhile, the function coefficients are related to the combustion process parameters, so the model is more practical.The model is then improved by taking the effect of the stratification of temperature on heat release into consideration. The analysis of CAI combustion process based on CFD simulation shows that less number of initial flame kernel as well as the average growth rate of flame kernel is obtained with more uniform temperature distribution. The two parameters mentioned above are crucial to determine the skeleton of the model. By set up correlations between the two parameters and temperature inhomogeneity, the heat release model is then capable to describe the effect of temperature stratification on CAI combustion. The method has been proved to be able to improve the prediction accuracy of CA50and combustion duration by0.47℃A and0.8℃A respectively.