LAUSR

For the sake of reducing the computational load and enhancing the implementation possibility, a dual time-scale extended Kalman filter is developed in this paper. First, a conventional multistate mathematical selective catalytic reduction (SCR) model is simplified into a single-state SCR model. This simplification would reduce the computational cost a lot. Inspired by the fact that the variation of NH$_{3}$ coverage ratio between two cells changes relatively slow, then, the dual time-scale extended Kalman filter is proposed. The dual time-scales include one fast time-scale for the average NH $_{3}$ coverage ratio and a slow time-scale for the difference of NH$_{3}$ coverage ratios between the actual cell and the average value. In the dual time-scale scenario, it is unnecessary to estimate the NH $_{3}$ coverage ratio of each SCR cell every time. Finally, simulation results for a medium-duty diesel engine through the MATLAB/SIMULINK show the proposed method has good performance and lower computational load.