Accelerated design of acoustic-mechanical multifunctional metamaterials via neural network

Despite the immense potential of metamaterials in acoustics and mechanics research, effectively integrating their advantages to meet specific application requirements remains a formidable challenge. In this study, we present a novel approach to designing a multifunctional metamaterial by combining Helmholtz resonant sound absorbers with lattice structures, successfully achieving integration of sound absorption and impact resistance capabilities. To accelerate the design process, we propose a neural network-driven impedance calculation model that can be flexibly applied to parallel structures with varying numbers of units. By integrating this with an optimization algorithm, our parallel array structure achieves an average sound absorption coefficient of approximately 0.87 within 300–600 Hz and on-demand sound absorption design based on the noise spectrum, while achieving a roughly 39 % increase in specific energy absorption performance compared to traditional Helmholtz resonators. Overall, our study provides an innovative paradigm for designing combined acoustic/mechanical metamaterials and accelerates optimal design of metamaterial parallel arrays for noise control.