LAUSR

Abstract This study aims to accurately analyze the three-dimensional (3-D) mesoscopic permeability of porous asphalt mixture. Industrial CT scanning and 3-D image reconstruction technology were used to analyze the mesoscopic void space in porous asphalt mixture. A mesoscopic permeability numerical model was built based on the Lattice Boltzmann Method (LBM). The vertical permeability of porous asphalt mixture was analyzed with the numerical model and verified by a self-designed permeability coefficient tester that can apply a confining pressure to the specimen side using a rubber membrane. Results indicate that the self-designed permeability tester has a rigorous test theory, simple operation, and reliable results. Results also illustrate the uniformity between the test and numerical model calculations. By applying different boundary conditions, the magnitude and distribution of permeability flow velocity at any point within the porous asphalt mixture sample can be accurately obtained. The principal and deviator vector of the macroscopic permeability coefficient can be acquired. The 3-D mesoscopic permeability was found to strongly correlate with the mesoscopic void structure. The transverse permeability coefficient is about 1.5–2.0 times the vertical permeability coefficient. Samples in different orientations or with different heights had a variable permeability coefficient. The larger the sample’s height, the smaller the permeability coefficient. The findings of this study are expected to provide more precise 3-D permeability parameters for drainage system design using asphalt pavement and satisfy the requirements of drainage function and traffic safety.