Real-time NOx estimation in light duty diesel engine with in-cylinder pressure prediction

Recently, there have been numerous efforts to cope with automotive emission regulations. Various strategies to reduce engine-out NOx emissions and proper after-treatment systems, such as selective catalytic reduction (SCR) and lean NOx trap (LNT), have been taken into account in the engine research field. In this study, real-time engine-out NOx prediction model was established where zero-dimensional NO and NO₂ models were combined with in-cylinder pressure model. During the procedure for estimating NO and NO₂ (NOx), a real-time prediction model of in-cylinder pressure was applied so that the inputs to the NOx prediction model could be provided only by the data acquired from the engine control unit (ECU). This implies that an in-cylinder pressure sensor is not necessarily required to properly predict the engine-out NOx in real time. The real-time NOx estimation model was validated through the worldwide harmonized light-duty vehicle test cycle (WLTC) without a pressure sensor, and the total NOx error during the mode was comparable with the total NOx error of the portable NOx sensor. This real-time NOx estimation model can ultimately contribute to minimizing tail-pipe NOx emissions by influencing both emission calibration at the engine design stage and the management of NOx after-treatment systems where NOx conversion efficiency is heavily affected by the NO₂/NO ratio.