Abstract
KEYWORDS – Diesel engine, Emission regulation, After-treatment system, NO/NOx ratio, NOx estimation model
ABSTRACT – To overcome the stringent NOx emission regulations, car manufacturers have been actively adopting after-treatment devices, such as selective catalytic reduction (SCR) or lean NOx trap (LNT). However, the NOx reduction efficiency after the NOx passes through the after-treatment device changes with different NOx/NO ratios in engine-out emissions. For this reason, a precise NO/NOx ratio engine-out estimation is critical for maximizing NOx reductions. In this paper, a NOx prediction method based on an empirical relationship was established and validated during steady-state experimental conditions and the WLTC homologation cycle.
The prediction model was verified on a 1.6-L EURO5 light-duty diesel engine. The engine was equipped with a single-stage turbocharger and a common rail direct injection system with solenoid injectors. The representative steady-state operating points were selected not only to cover the entire WLTC but also to test the swept conditions, such as the EGR, main injection timing, rail pressure and boost pressure, to determine the model’s robustness. Various NOx estimation models, such as the 1-D table model, exponential model and polynomial model, have been considered and applied to previous real-time NO prediction model by the authors. The real-time predictions for both NO and NOx were conducted in the ES1000 with the software ASCET. In this paper, the accuracies of the results from the NOx prediction models are presented. The prediction models were verified through steady-state experiments, where the predicted and measured values during WLTC were found to be highly consistent.