TY - JOUR
T1 - Pressure-adaptive ultra-thin hybrid metamaterials for broadband low-frequency underwater sound absorption
AU - Cai, Chuxiang
AU - Xin, Fengxian
N1 - Publisher Copyright:
© 2025 Elsevier Ltd
PY - 2025/11
Y1 - 2025/11
N2 - An ultra-thin hybrid metamaterial is proposed for broadband low-frequency underwater sound absorption. It combines rubber-coated Helmholtz resonators (RHRs) and rubber-plate composite structures (RPCs) of different sizes in parallel to absorb sound in both low and mid-to-high frequency bands. A finite element model is established to evaluate the acoustic performance of the hybrid metamaterial, and a predictive model based on a deep neural network (DNN) is developed to rapidly estimate the absorption coefficient. The analysis of the vibration velocity distribution shows that the a quasi-Helmholtz resonance occurs in the RHR under excitation of low-frequency sound waves. This resonance significantly affects the vibration distribution of the rubber in the adjacent RPC. The rubber between the parallel metal plates in the RPC undergoes wave mode conversion, effectively dissipating the sound energy and enabling a tunable quasi-perfect absorption peak in 300-1200 Hz. At mid-to-high frequencies, strong compressive vibration and wave mode conversion in the RPC primarily contribute to excellent absorption. By integrating the deep neural network (DNN) model with a genetic algorithm (GA), the key parameters of the RHR and RPC units are rapidly optimized. As a result, the hybrid metamaterial achieves an absorption coefficient greater than 0.8 in the frequency range of 364-10,000 Hz, with a structure thickness equal to 1/82 of the wavelength at 364 Hz. The optimized hybrid metamaterial exhibits robust acoustic performance under moderate changes in rubber material properties and maintains effective broadband absorption under oblique incidence angles below 60°, demonstrating good angular stability. In addition, the acoustic performance of the optimized hybrid metamaterial is investigated under various hydrostatic pressures. Even at 3 MPa, the absorption coefficient remains above 0.8 at frequencies above 410 Hz, demonstrating excellent low-frequency absorption under hydrostatic pressure. This work provides valuable guidance for the design of underwater hybrid metamaterials for broadband sound absorption.
AB - An ultra-thin hybrid metamaterial is proposed for broadband low-frequency underwater sound absorption. It combines rubber-coated Helmholtz resonators (RHRs) and rubber-plate composite structures (RPCs) of different sizes in parallel to absorb sound in both low and mid-to-high frequency bands. A finite element model is established to evaluate the acoustic performance of the hybrid metamaterial, and a predictive model based on a deep neural network (DNN) is developed to rapidly estimate the absorption coefficient. The analysis of the vibration velocity distribution shows that the a quasi-Helmholtz resonance occurs in the RHR under excitation of low-frequency sound waves. This resonance significantly affects the vibration distribution of the rubber in the adjacent RPC. The rubber between the parallel metal plates in the RPC undergoes wave mode conversion, effectively dissipating the sound energy and enabling a tunable quasi-perfect absorption peak in 300-1200 Hz. At mid-to-high frequencies, strong compressive vibration and wave mode conversion in the RPC primarily contribute to excellent absorption. By integrating the deep neural network (DNN) model with a genetic algorithm (GA), the key parameters of the RHR and RPC units are rapidly optimized. As a result, the hybrid metamaterial achieves an absorption coefficient greater than 0.8 in the frequency range of 364-10,000 Hz, with a structure thickness equal to 1/82 of the wavelength at 364 Hz. The optimized hybrid metamaterial exhibits robust acoustic performance under moderate changes in rubber material properties and maintains effective broadband absorption under oblique incidence angles below 60°, demonstrating good angular stability. In addition, the acoustic performance of the optimized hybrid metamaterial is investigated under various hydrostatic pressures. Even at 3 MPa, the absorption coefficient remains above 0.8 at frequencies above 410 Hz, demonstrating excellent low-frequency absorption under hydrostatic pressure. This work provides valuable guidance for the design of underwater hybrid metamaterials for broadband sound absorption.
KW - Broadband low-frequency sound absorption
KW - Deep neural network
KW - Hydrostatic pressure
KW - Underwater acoustic hybrid metamaterial
UR - https://www.scopus.com/pages/publications/105009349041
U2 - 10.1016/j.tws.2025.113603
DO - 10.1016/j.tws.2025.113603
M3 - 文章
AN - SCOPUS:105009349041
SN - 0263-8231
VL - 216
JO - Thin-Walled Structures
JF - Thin-Walled Structures
M1 - 113603
ER -