A Sensitivity Study of Thermal and Soot Oxidation Dynamics of Gasoline Particulate Filters 2019-01-0990
Gasoline particulate filters (GPF) are devices derived from diesel particulate filters (DPF) to prevent soot emission from direct injection (DI) technology. Arunchalam et al.  recently presented a numerical model for a ceria-coated GPF that predicts internal GPF temperature and soot amount combusted during GPF regeneration events. Being that such quantities cannot be directly measured during real-time operation, and owing to their critical importance for GPF health monitoring as well as regeneration scheduling, the above model is a good candidate for model-based observer design. Model parameter identifiability and observability analysis are critical steps in the design of a model-based observer.
In this paper, first of all, we propose a reduced-complexity model along with its validation that accurately predicts GPF temperature and soot amount during regeneration events. Sensitivity study is conducted over the above model to understand the model’s identifiability with respect to the model parameters. The effects of each parameter, as well as the initial temperature and amount of soot, on the predicted model output were studied, and changes in the maximum temperature were monitored as well to predict critical temperatures of the cordierite material composing the GPF. GPFs rely on regeneration events triggered by high internal temperatures to maintain high filtration efficiency.
Analysis indicated that the model is most sensitive to the activation energy of GPF regeneration. The sensitivity analysis and reduced model can be used simultaneously for future observer-based design.
1. H. Arunachalam, G. Pozzato, M. A. Hoffman and S. Onori, "Modeling the thermal dynamics inside a ceria-coated Gasoline Particulate Filter," 2017 IEEE Conference on Control Technology and Applications (CCTA), Mauna Lani, HI, 2017, pp. 99-105
Aki Takahashi, Slava Korneev, Simona Onori