Bohan Cao

Broadband sound absorption has long been a concern in noise control engineering, but the inverse design of multi-parallel microperforated panels (MPPs) for broadband sound absorption remains challenging. To address this issue, we propose a deep learning model that combines a variational autoencoder (VAE) with a physics-informed neural network (PINN) to accelerate the inverse design process of a multi-parallel MPP. Following Maa’s theory, we generated a dataset of 500,000 samples to train the model. By incorporating the PINN, we added an acoustic physical constraint to the loss function, promoting model convergence and the derivation of stable, unified parameters. The efficacy of the inverse design model was validated through theoretical analysis, finite element simulations, and impedance tube experiments. The experimental results show that the average sound absorption coefficient of multi-parallel MPPs within the frequency range of 500–1200 Hz is 0.85. Our work contributes to accelerating the inverse design of multi-parallel acoustic metamaterials.

Nanofibers, with their lightweight structure and superior sound absorption, are promising materials for noise control in automotive and architectural applications. However, due to the complex porous structure of nanofibers, established acoustic models often fail to accurately quantify the microstructure’s influence on sound absorption characteristics, resulting in substantial prediction errors. To determine the sound absorption characteristics of nanofibers, an equivalent fiber network model was developed using the multiscale finite element analysis (MFEA) method based on SEM images of nanofibers. The slip boundary condition (SBC) was then applied to calculate the microstructural parameters necessary for macroscopic characterization. The sound absorption coefficients of nanofibers were characterized using three acoustic models, and the results were compared with the experimental data. The predictions of the Limp frame model agreed well with the experimental data within the 500–6400 Hz frequency range. Through use of the multiscale model developed in this study, a deterministic relationship between microstructure and acoustic properties was established, revealing that the inertial interactions between sound waves and the nanofiber skeleton, as well as the slip boundary effect at the nanofiber surfaces, are among the primary mechanisms contributing to the flow resistance and superior sound absorption performance of nanofibers.