Reconstruction of Porous Microstructure Models for Validation and Estimation of Phenomenological Acoustic Parameters

Porous materials are widely used for sound absorption, yet accurately modeling their acoustic properties remains challenging. The widely adopted phenomenological models calculate equivalent density and bulk modulus based on macro material parameters (porosity, tortuosity, resistivity, etc.), usually obtained from estimated physical measurements or computer tomography (CT) images of samples. However, a fully computational simulation approach (i.e., no experiment is needed to estimate parameters) is desired, where the phenomenological macro parameters can be obtained and validated via micro-scale fluid dynamics models. In this work, we present an efficient procedure to construct detailed 3D geometry of porous materials based on CT images of a material sample. The microstructure information is extracted mainly using a watershed-based image segmentation technique. These high-fidelity geometries enable micro-scale fluid dynamics simulations, providing a virtual testbed. By comparing these micro-scale simulation results against predictions from equivalent continuum models, we can rigorously validate the phenomenological parameters. This multi-scale framework facilitates the fundamental understanding of material microstructures, enabling more accurate design and optimization of acoustic treatments for noise control applications in transportation and building systems.