An extended Kalman filter for input estimations in diesel-engine selective catalytic reduction applications

Nowadays the legislative regulation on emissions of Diesel engines is stringent such that an aftertreatment system is necessary. To reduce vehicle NO x emission, a selective catalytic reduction (SCR) system is widely used for Diesel-engine applications. In the SCR system, the gaseous ammonia plays a significant role which is utilized as the reduction to eliminate the NO x emission. To facilitate the NO x reduction, a NO x sensor and an ammonia sensor placed before the SCR catalyst are a good strategy. However, physical sensors would increase the system cost and diagnosis challenge. To reduce the number of physical sensors, in this paper, observers are designed with the assist of extended Kalman filter (EKF) to estimate the NO x or ammonia concentrations before the SCR catalyst. Simulation results show that the designed observers based on EKF can achieve the prescribed objectives quite well. HighlightsEKF algorithm is applied to SCR system.Inputs and states are estimated simultaneously.Proposed method is verified via a well-developed simulator.