Data-based symbolic simplification of kinetic mechanisms for surface reaction networks

Detailed kinetic surface reaction mechanisms are used in kinetic models for automotive catalysis applications. Therefore, these models are complex and often numerically stiff. For simulation, optimization and control purposes it is desirable to employ models of lower complexity that are accurate in specific operating regimes. In this contribution, a number of data-based kinetic mechanism reduction methods are combined in order to obtain an automated method for mechanism simplification. The principle of this method is to use time-scale analysis to generate nonlinear lumpings from partial equilibrium and quasi steady-state assumptions These lumpings have to remain as decoupled as possible, so that they can be solved explicitly.