Kinetic Parameters Estimation using Vehicle Data for Exhaust After-treatment Devices

Karthik  Ramanathan, General Motors, Global Research and Development Center

One of the main challenges in modeling exhaust after-treatment devices (catalytic converters) is in obtaining the right kinetic mechanisms and expressions for the various heterogeneous reactions occurring in the device. Laboratory experiments give detailed insights to the reaction mechanism and analytical forms of the rate expressions as they are well-controlled and well-behaved as compared to vehicle tests. However, to predict emissions on a vehicle test, the laboratory-estimated kinetic parameters are not entirely capable because of the various uncertainties in the vehicle tests.

In this work, six different vehicle datasets are used to calibrate and validate the Three Way Catalyst (TWC) kinetic model. Our emphasis in this work is restricted to predicting the light-off (cold-start emissions) in TWC. The kinetic model is calibrated using 4 vehicle datasets (which use the FTP drive cycle) using iSIGHT software package. The kinetic parameters of the various reactions occurring in the TWC are estimated to match the experimental data through exploratory and local optimization methods. A systematic approach (with increasing complexity) is used to estimate the kinetic parameters. The estimated parameters are then used to validate the model on two different vehicle datasets (one NEDC drive cycle and one FTP drive cycle) with different catalyst compositions, and engine power (and hence different engine out exhaust compositions). Though the parameter estimation process is illustrated through the TWC, it is kept as generic as possible to exhaust aftertreatment devices and a set of guidelines for parameter estimation (specifically for use in exhaust aftertreatment devices) will be presented.

Download Presentation: