Virtual Sensor Modelling for Ammonia Slip Estimation to Improve SCR Performance in Real Drive Emissions

Urea-NH3 dosed Selective Catalytic Reduction is a powerful reaction system to ensure NOx reduction in the exhaust gases by minimizing ammonia slip. When the dosed ammonia exceeds the actual request than the required, NH3 to NOx ratio is potentially high, the unused ammonia is limited to 10ppm corresponding to experimental result of every World Harmonic Transient Cycle. The dosage estimation depends on the NOx sensors which has this drawback of high cross-sensitivity to ammonia that can affect the measurement of NOx and compromise the SCR-ASC closed loop strategies. This paper aims to resolve the complexity in prediction of ammonia slip to resolve the cross-sensitivity of tailpipe NOx sensor in the SCR system by a closed loop estimation of NOx and ammonia slip to ensure high NOx conversion efficiency. The focus is to develop a simplified model-based solution for estimating ammonia slip, because of the limitations in the real drive conditions in SCR system. This model approach is designed in such a way that it predicts the NOx and NH3 emissions after the SCR catalyst and calculates even in the failure conditions of closed loop feedback of urea dosing. A Filtering solution that combines the signals from inlet NOx sensor, tailpipe NOX sensor and the total system efficiency to provide a reliable estimation of NH3 slip.