Multi-channel physics-based modeling and experimental validation of an uncoated gasoline particulate filter in clean operating conditions

This paper presents the development of a two-dimensional (2-D), multi-channel, Gasoline Particulate Filter (GPF) system model. The GPF is an aftertreatment device designed to reduce soot particulate emissions in gasoline direct injection engines. A first-principle modeling approach using mass, energy and momentum balance equations is used to describe the clean filter dynamics. The 2D model is simulated by finite element analysis and the unknown model parameters are identified using a Particle Swarm Optimization (PSO) algorithm over experimental data from a GPF instrumented with 15 thermocouples. A validation study is presented that shows the effectiveness of the model proposed. The novel 2D GPF dynamics constitute the modeling foundation needed for the development of control-oriented models that will be explored in future works.