A Methodology for Studying the Relationship Between Heat Release Profile and Fuel Stratification in Advanced Compression Ignition Engines

Low temperature combustion strategies have demonstrated high thermal efficiency with low pollutant emissions (e. g., oxides of nitrogen and particulate matter), resulting from reduced heat transfer losses and lean air-fuel mixtures. One such advanced compression ignition combustion strategy, Reactivity Controlled Compression Ignition (RCCI), has demonstrated improved control over the heat release event due to the introduction of in-cylinder stratification of equivalence ratio and chemical reactivity via direct injection of a high-reactivity fuel into a premixed low-reactivity fuel/air mixture. The nature of the RCCI strategy provides inherent fuel flexibility, however, the direct injection strategy must be tailored to the combination of premixed and direct injected fuel chemistry and engine operating conditions to optimize efficiency and emissions. In this work, a 0-D methodology for predicting the required fuel stratification for a desired heat release rate profile for kinetically controlled stratified-charge combustion strategies is proposed. The methodology, referred to as Fuel Stratification Analysis (FSA), was inspired by a similar approach which utilized ignition predictions calculated via a Livengood-Wu integral approach correlated with experimental heat release profiles to determine in-cylinder temperature stratification in homogeneous charge compression ignition (HCCI) combustion. The methodology proposed in this work expands upon this method to include strategies involving fuel stratification (such as RCCI). Reacting and non-reacting CFD simulations were performed with the KIVA3V release 2 code to validate the CFD. Reacting simulations were validated against published experimental HCCI and RCCI data, and non-reacting simulations were used to generate fuel distribution profiles to compare to the FSA results. The results of this validation showed that the FSA method was able to provide good overall agreement in the predicted fuel distribution compared to the actual fuel distributions from CFD simulations within the range of injection timings of interest in RCCI combustion (-140 degrees to about -35 degrees after top-dead-center). For later injection timings, FSA predictions are not able to capture the actual fuel distributions present at the start of combustion, likely due to a transition into a mixing dominated, as opposed to a kinetically dominated, combustion regime, thereby violating one or more inherent method assumptions.