Gasoline particulate filters (GPFs) are the most promising and practically applicable devices to reduce Particulate Matter (PM) and Particulate Number (PN) emissions from gasoline direct ignition engines. A model that can predict internal GPF temperature dynamics during regeneration events can then be implemented online to maintain GPF health and aide in exotherm control algorithms without the associated instrumentation costs. This work demonstrates a control-oriented model, which captures the thermal dynamics in a catalyzed, ceria-coated GPF in the axial direction. The model utilizes soot oxidation reaction kinetics to predict internal GPF temperature dynamics during regeneration events using three finite volume cells.
A model methodology initially proposed by Arunachalam et al  is utilized with the GPF of this work, validating the broad applicability of that methodology. Then, the model’s temperature prediction fidelity is improved through axial discretization. The zonal model parameters are identified via a Particle Swarm Optimization using experimental results from the instrumented GPF. Identified parameters from the various data sets are used to develop a linear parameter varying model for prediction of the axial temperature distribution within the GPF. The resulting model is then validated against an experimental data set utilizing the exhaust temperature entering the GPF. The spatial discretization methodology employed both enables the prediction of spatial temperature variation within the GPF and improves the accuracy of the peak temperature prediction by a factor ranging from 2–10x.