Deep-subwavelength lightweight metastructures for low-frequency vibration isolation. (March 2022)

• Record Type: Journal Article
• Authors: Jin, Yabin; Zeng, Shixuan; Wen, Zhihui; He, Liangshu; Li, Yong; Li, Yan
• Is Part Of: Materials & design. Volume 215(2022)
• Publication Date: 2022-03
• Subjects: Lightweight metastructure -- Deep-subwavelength bandgap -- Vibration isolation -- Machine learning -- Inverse design; Materials -- Periodicals; Engineering design -- Periodicals; Matériaux -- Périodiques; Conception technique -- Périodiques; Electronic journals; 620.11
• DOI: 10.1016/j.matdes.2022.110499
• Languages: English
• ISSNs: 0264-1275

Abstract

Graphical abstract: Highlights: Archimedean spiral and a double-layer corrugated sandwich core are combined for low-frequency vibration isolation, light weight and high static load-bearing properties. The metastructures realize vibration isolation below 150 Hz with the corresponding flexural wavelengths 40 times the lattice constant. A machine learning method is proposed to realize on-demand inverse design of the metastructures. The proposed metastructures provide a promising way for potential functional integration devices for vibration and noise control in industries. Abstract: How to endow diverse functions for lightweight structures without increasing mass has always been a major challenge (e.g., to have simultaneous high static stiffness and extreme-low dynamic stiffness in one structure) from both fundamental science and engineering application. In this work, Archimedean spiral metastructures are proposed for low-frequency vibration isolation by combining a double-layer corrugated sandwich core and two spiral metamaterial plates, which the former provides lightweight and high static load-bearing properties, and the latter possesses deep-subwavelength bandgaps. Both of the two proposed metastructures made of resin and Aluminum show efficient vibration isolation below 150 Hz where the corresponding flexural wavelengths are about 40 times the lattice constant, being a highly compact design for deep-subwavelength vibration control. We further develop a machine learning method to realize on-demand inverse design of the metastructures with targeted bandgap properties. The experimental and numerical results of the bandgaps are highly consistent, and further validate the accuracy of the machine-learning prediction. The slits between the Archimedean spiral arms make the metastructures be possible to further integrate sound absorption and isolation functions in future. The proposed metastructures provide a promising way for potential functional integration devices for vibration and noise control in industries such as aerospace engineering.