A review of remote-control strategies for reactivity controlled compression ignition combustion

Over the past years, the increasingly stringent emission regulations for Internal Combustion Engines (ICE) have led to the development of non-conventional combustion strategies like Low Temperature Combustions (LTC) characterized by high efficiency and low pollutant emissions. One of the most relevant LTC strategies is the Reactivity Controlled Compression Ignition (RCCI), characterized by the combustion of a lean mixture composed by air and a low reactivity fuel (LRF, gasoline in the case under study) ignited by a high reactivity fuel (HRF, Diesel in this study), directly injected in the combustion chamber, due to the high pressure and temperature in the cylinder. The proper management of this combustion strategy results in high efficiency and low engine-out emissions, with the simultaneous mitigation of both nitrogen oxides (NOx) and particulate matter (Soot). On the other hand, this combustion methodology is affected by high instability and high sensitivity to slight variations of the in-cylinder thermal conditions. Previous works demonstrate that combustion stability can be guaranteed through closed-loop control strategies that vary the injection parameters to keep the center of combustion (CA50, i.e. the angular position when the 50% of fuel burned within the engine cycle) at a proper target value. Although the center of combustion can be directly evaluated from in-cylinder pressure measurement, the on-board installation of in-cylinder sensors is still uncommon, mainly because they would increase the cost of the whole engine management system. Due to the above considerations, two different closed-loop control strategies have been developed by the authors of this paper to evaluate combustion characteristics using low-cost sensor, that are already present on-board for other management purposes. The current work summarizes these different strategies and demonstrate how a feed-forward estimation of the ignition delay, based on the estimation of the in-cylinder temperature, can improve the transient behavior of the control and how it is possible to replace all the information from in-cylinder pressure signal.