Robust Filtering for Ammonia Coverage Estimation in Diesel Engine Selective Catalytic Reduction Systems
In this paper, we investigate the nonlinear observer designs to estimate the ammonia coverage ratio in the diesel engine selective catalytic reduction (SCR) systems. The ammonia coverage ratio is an important variable due to its critical role in the SCR NOx conversion and the ammonia slip. However, the ammonia coverage ratio cannot be directly measured by onboard sensors. Therefore, it is necessary to develop effective observers to estimate the ammonia coverage ratio online. Based on a three-state SCR model, we develop two nonlinear observers. The first one only employs the dynamics of the ammonia concentration. The structure and the algorithm are simple. But it is sensitive to the measurement noises and the uncertainties in the system parameters. The second one is a discrete-time smooth variable structure estimator which is robust to the measurement noises, the approximation error, and the system uncertainties. Both estimators are implemented on a full-vehicle simulation of the FTP75 test cycle. The simulation results have verified the theoretical analysis.