隔振
材料科学
振动
非线性系统
结构工程
带隙
分离(微生物学)
声学
光电子学
工程类
物理
量子力学
微生物学
生物
作者
Seung Yoon On,Hyemi Moon,Seong Yeon Park,Tae Woong Ohm,Wonvin Kim,Hyunsoo Hong,Seong Su Kim
标识
DOI:10.1016/j.matdes.2022.111397
摘要
• Proposed periodic arched models to enhance vibration isolation performance based on integrating the structural nonlinearity and band gap effect. • Proposed machine learning optimization process to generate the optimum geometrical parameters of target band gap frequencies. • Determined the improvement in the vibration isolation performance of periodic arched structure by the numerical analysis and experimental results. In this study, a Roman-bridge-inspired metamaterial for vibration isolation is proposed that simultaneously realizes a geometrical nonlinear structure and band gap effect, through the arrangement of periodic materials in a semicircular arched structure. Finite element analysis and band diagram analysis are performed to verify the effect of the periodic arch-structured metamaterial by calculating the transmissibility and band gap frequency range of cuboid, arched and periodic arched models. Based on the calculation results, the relationship between the geometrical parameters of the periodic arch structure and band gap frequency are analyzed, and an empirical correlation is formulated for the generation of a training database for a machine learning optimization model. The geometrical parameters of the metamaterial were optimized using the machine learning model based on an autoencoder architecture with supervised learning. An optimized metamaterial with a band gap area at three target frequencies is generated by the trained neural network, and the effectiveness of the optimized structure is validated by both numerical analysis and a frequency response test.
科研通智能强力驱动
Strongly Powered by AbleSci AI