In order to describe the microstructure of the porous wall of a gasoline particulate filter (GPF), a pore size distribution based on a probability density function (PDF) and a non-uniform porosity distribution are introduced. The dynamic process of deep-bed filtration in GPF with inhomogeneous wall structure is studied, considering different particle size distributions (PSDs). The results show that most of the particles are captured in the top region of the porous wall, in which the porosity and permeability reduce more obviously during dynamic filtration, and the bottom of the porous wall contributes little to the overall filtration. As time increases, the filtration efficiency of the porous wall for each particle size increases, and the most penetrating particle’s diameter becomes smaller gradually. The dynamic evolution of characteristic parameters of the porous wall, the most efficient filtration region, the pressure drop and the duration of deep-bed filtration are strongly influenced by PSD. This research illustrates the necessity to consider difference of PSDs when working on the filtration process of GPFs.

为了描述汽油颗粒过滤器（GPF）的多孔壁的微观结构，引入了基于概率密度函数（PDF）的孔径分布和非均匀孔隙率分布。考虑不同的粒径分布（PSD），研究了壁结构不均匀的GPF中深床过滤的动态过程。结果表明，大多数颗粒被捕获在多孔壁的顶部区域，在动态过滤过程中，孔隙率和渗透率降低更为明显，而多孔壁的底部对整体过滤的贡献很小。随着时间的增加，每种粒径的多孔壁的过滤效率都会提高，并且最渗透的粒径逐渐变小。PSD对多孔壁特征参数的动态演变，最有效的过滤区域，压降和深床过滤的持续时间有很大影响。这项研究表明在进行GPF过滤过程中必须考虑PSD的差异。