Hydrothermal Aging Factor Estimation for Two-Cell Diesel-Engine SCR Systems via a Dual Time-Scale Unscented Kalman Filter

Abstract

Selective catalytic reduction (SCR) systems have been widely used to meet the emission regulations and two-cell SCR systems have shown the advantages of high NO $_{x}$ conversion efficiency and low NH $_{3}$ slip simultaneously. However, it is noteworthy that the catalyst performance of SCR device would degrade over the service time gradually. If the performance degrading is not well-compensated, the NO $_{x}$ conversion efficiency would reduce significantly. Different from detecting the aging issue in the laboratory, a practical method is to design an observer to estimate the aging factor online such that the urea injection can be modified accordingly. In this paper, we aim to construct an aging-factor observer for two-cell SCR systems. In order to reduce the algorithm computational load and guarantee the implementation performance, we propose a dual time-scale algorithm based on two-cell SCR model and unscented Kalman filter. There is a fast time scale and a slow time scale in the algorithm. Two simulation studies of constant aging factor and time-varying aging factor are investigated in the simulation environment of MATLAB/SIMULINK. The simulation results indicate that the proposed dual time-scale observer works well under different conditions and the calculational load is reduced significantly.