Deep Learning-Accelerated Designs of Tunable Magneto-Mechanical Metamaterials

Abstract

Metamaterials are artificially structured materials with unusual properties, such as negative Poisson’s ratio, acoustic band gap, and energy absorption. However, metamaterials made of conventional materials lack tunability after fabrication. Thus, active metamaterials using magneto-mechanical actuation for untethered, fast, and reversible shape configurations are developed to tune the mechanical response and property of metamaterials. Although the magneto-mechanical metamaterials have shown promising capabilities in tunable mechanical stiffness, acoustic band gaps, and electromagnetic behaviors, the existing demonstrations rely on the forward design methods based on experience or simulations, by which the metamaterial properties are revealed only after the design. Considering the massive design space due to the material and structural programmability, a robust inverse design strategy is desired to create the magneto-mechanical metamaterials with preferred tunable properties. In this work, we develop an inverse design framework where a deep residual network replaces the conventional finite-element analysis for acceleration, realizing metamaterials with predetermined global strains under magnetic actuations. For validation, a direct-ink-writing printing method of the magnetic soft materials is adopted to fabricate the designed complex metamaterials. The deep learning-accelerated design framework opens avenues for the designs of magneto-mechanical metamaterials and other active metamaterials with target mechanical, acoustic, thermal, and electromagnetic properties.

Origami Tribo-Metamaterials with Mechanoelectrical Multistability

“…Mechanical metamaterials are the structural materials periodically assembled by microstructures to exhibit extraordinary mechanical characteristics, , which can be characterized between natural materials that are based on the intrinsic properties of materials and manmade structures that are mainly affected by structural properties. , As a consequence, the localized behavior of mechanical metamaterials at the microstructure level is structure-like, and the overall performance at the metamaterial level is similar to homogenized materials. , Mechanical metamaterials can overcome the trade-off challenge that natural materials typically have to face between physical properties and mechanical performance, such as origami metamaterials reported with bistable force–displacement relations . Since origami metamaterials are significantly dependent on their origami cells, well tunability over mechanical properties can be induced by rationally designing those cells. , Studies have been conducted on designing origami cells to obtain the origami metamaterials with desirable configurations, − unprecedented mechanical properties, such as negative Poisson’s ratio, − negative stiffness, , and advanced functions. , Recent research interests have been shifted to exploring the functionality of mechanical metamaterials using functional materials, such as energy materials for energy generation, power absorption, , energy storage, thermal materials for thermophotovoltaic response, thermomechanical response and thermoelastic response, , magnetic materials for electromagnetic energy harvesting and absorption, − and so forth. Recent development of advanced functional materials (e.g., self-healable materials, , nanomaterials, , etc.)…”

Multimodal Deep-Learning Framework for Accurate Prediction of Wettability Evolution of Laser-Textured Surfaces

“…A model that can rapidly determine the target performance of complex micro/nanostructures and easily generalize them to similar systems is required to identify laser-textured surfaces with superhydrophobic properties among large design spaces within reasonable time frames. Deep learning, which has been widely used in materials science, − can independently extract and learn features and make intelligent decisions to rapidly establish the regression relationship between process parameters or microstructure (inputs) and target performance (outputs). Using a convolutional neural network (CNN), Kim et al effectively developed a prediction model of the stress–strain curves of unidirectional composites with complex microstructures, presenting an interesting example .…”

Micro-scale Mechanical Metamaterial with a Controllable Transition in the Poisson's Ratio and Band Gap Formation

“…In the case of macroscopic systems, one of the most promising approaches corresponds to the use of appropriately distributed magnet-like inclusions that respond to the application of an external magnetic field [53,54]. However, this task becomes much more challenging at small scales where one of the more feasible approaches seems to be the use of magnetic nanoparticles dispersed within a non-magnetic host material [10,[55][56][57]. Another approach making it possible to construct an active mechanical metamaterial is associated with the use of several materials with different thermal expansion coefficients.…”