SCR Catalytic Converter State And Parameter Estimation For Diesel Engines Based On Dual Extended Kalman Filter

With the rapid development of automobile industry, the number of motor vehicles and ownership, the annual production and sales of motor vehicles in China have ranked the first one in the world. The development of the automotive industry has made a huge contribution to the economic development of China. However, the growth of vehicle ownership also brings lots of environmental problems, especially the particulate matter and nitrogen oxides emissions from diesel engines in the motor vehicles.In the phase of China State V and State VI, Selective Catalytic Reduction (SCR),which is the most important technology for reducing NOx emission of diesel engines, has been studied in-depth and widely used. The thesis is focused on the following five aspects:the SCR catalytic converter system modeling, the design of state estimation and parameter estimator of Dual Extended Kalman Filter (DEKF) study of NH3 storage capacity, and experimental results validation of state estimatorFirst, the establishment of real-time Simulink simulation model of catalytic converter is completed based on chemical mechanism of SCR catalytic converter, including the catalyst temperature model, NH3 selective catalytic reduction reaction model, NH3 oxidation reaction model and NH3 adsorption as well as desorption reaction model.Second, it is proposed theammonia storage quantity model of NH3 storage capacity under the interaction of temperature and space velocity from the study of temperature and space velocity effects on the NH3 storage capacity. The Dual Extended Kalman Filter is designed based on the mathematical model of the SCR catalyst and the ammonia quantity model. The estimation algorithm combined state and parameter Extended Kalman Filter which are at mutual coupling condition to run in parallel. The state estimator is embedded into the parameter estimator, and state estimation and parameter estimation also predict and update together so that the state estimates are more close to the true value. The downstream concentration of NOx and NH3, NH3 coverage and NH3 storage capacity of catalytic converter are estimated by DEKF state estimator, respectively. Moreover, the four parameters in the nonlinear state model of SCR catalyst are estimated by DEKF parameter estimator, including NH3 selective catalytic reduction reaction rate factor, NH3 oxidation rate factor, the rate of NH3 adsorption reaction factor and NH3 desorption reaction factor. The estimation process utilizes the inputs of the engine transient ETC operating conditions, SCR catalytic inlet NOx concentration, SCR catalytic converter inlet temperature and urea injection amount and outputs of the concentrations of NOx and NH3 as well as NH3 coverage of a SCR catalytic converter model to test the accuracy of state estimator. By applying the the four parameters obtained from the parameter estimator, the comparison is made for the concentration of NOx and NH3 and NH3 coverage of SCR catalytic converter model under the ETC test conditions.Third, utilizing the upstream and downstream of the concentration of NOx and NH3 under the working condition of the engine ETC as the inputs of the Extended Kalman Filter (EKF) state estimator and four parameters obtained from the parameter estimator, a comparison is made between the concentration of NOx and NH3 estimated by state estimator and experimental measurement results. It shows that the state estimator can track the experimental measurement of downstream NOX and NH3 concentration of SCR catalytic converter in real-time and match the estimated states, which may replace expensive NOx or NH3 sensors, optimize control strategies of urea injection, and prevent NH3 leaking. The algorithm developed can also be applied to the study of OBD for SCR system.