Designing aperiodic metamaterials using mechanical neural networks
- 1 April 2026
- journal article
- research article
- Published by Royal Society of Chemistry (RSC) in Materials Horizons
- Vol. 13 (10) , 4973-4981
Abstract
Mechanic neural network design aperiodic metamaterials by first learning the desired behaviors via tuning the stiffness of their active beams. Passive beam design with those same stiffness are accordingly organized within the final design.Keywords
Funding Information
- Air Force Office of Scientific Research (FA9550-22-1-0008)
This publication has 42 references indexed in Scilit:
- Electro‐Thermally Controlled Active Mechanical Metamaterials with Programmable Stiffness and NonreciprocityAdvanced Science, 2025
- 3D Tiled Auxetic Metamaterial: A New Family of Mechanical Metamaterial with High Resilience and Mechanical HysteresisAdvanced Materials, 2024
- Kirigami Design and Modeling for Strong, Lightweight MetamaterialsAdvanced Functional Materials, 2022
- An Active Microarchitectured Material that Utilizes Piezo Actuators to Achieve Programmable PropertiesAdvanced Engineering Materials, 2016
- Programmable Elastic MetamaterialsAdvanced Engineering Materials, 2015
- Material interpolation schemes in topology optimizationArchive of Applied Mechanics, 1999
- Cellular solid structures with unbounded thermal expansionJournal of Materials Science Letters, 1996
- Optimal shape design as a material distribution problemStructural and Multidisciplinary Optimization, 1989
- Multilayer feedforward networks are universal approximatorsNeural Networks, 1989
- Generating optimal topologies in structural design using a homogenization methodComputer Methods in Applied Mechanics and Engineering, 1988