Nonlinear Disturbance Observer design for estimation of ammonia storage ratio in selected catalytic reduction systems

Urea based selected catalytic reduction (SCR) system is a promising devise to achieve high NOx reduction and widely used in diesel engine after-treatment systems. The ammonia storage ratio is critical in SCR feedback control for high NOx reduction and low ammonia slip. However, it cannot be directly measured by production sensors. To effectively estimate the ammonia storage ratio on line and reduce the cost by using exhaust gas sensors, a cost-effective Nonlinear Disturbance Observer (NDO) was designed based on part of the three-state SCR model by using NOx sensors only. In this approach, the ammonia storage ratio is treated as an external disturbance and estimated. The stability of the estimation was also analyzed in the paper. The simulation results based on the full-vehicle simulation of FTP-75 test show that the NDO can effectively estimate the ammonia storage ratio when NOx sensor with/without measurement noise or ammonia cross-sensitivity.