Studying Complex Evolution of Hyperelastic Materials under External Field Stimuli using Artificial Neural Networks with Spatiotemporal Features in a Small‐Scale Dataset

Abstract

Deep‐learning (DL) methods, in consideration of their excellence in dealing with highly complex structure–performance relationships for materials, are expected to become a new design paradigm for breakthroughs in material performance. However, in most cases, it is impractical to collect massive‐scale experimental data or open‐source theoretical databases to support training DL models with sufficient prediction accuracy. In a dataset consisting of 483 porous silicone rubber observations generated via ink‐writing additive manufacturing, this work demonstrates that constructing low‐dimensional, accurate descriptors is the prerequisite for obtaining high‐precision DL models based on small experimental datasets. On this basis, a unique convolutional bidirectional long short‐term memory model with spatiotemporal features extraction capability is designed, whose hierarchical learning mechanism further reduces the requirement for the amount of data by taking full advantage of data information. The proposed approach can be expected as a powerful tool for innovative material design on small experimental datasets, which can also be used to explore the evolutionary mechanisms of the structures and properties of materials under complex working conditions.

Detecting Microstructural Criticality/Degeneracy through Hybrid Learning Strategies Trained by Molecular Dynamics Simulations

“…The latent features can be divided into two groups, one for n = 0 and the other for n = 1−5. Therefore, criticality in the interfacial behavior occurs when the interphase thickness n changes from 0 to 1, and degeneracy in the interfacial behavior occurs when n takes arbitrary values within the range of [1,5]. For convenience, the former will be referred to as microstructural criticality in interphase thickness and the latter as microstructural degeneracy in interphase thickness.…”

“…Modern advancements in experimental or computational techniques have enabled the acquisition of large volumes of data in a reasonably short time, which provides a great opportunity for DL-based methodologies to assist the development of novel materials. The DL-based approach has already been adopted to understand the evolutionary mechanisms and properties of the 3D-printed porous silicone rubber structures under complex external stimuli, 1 to predict the strain and stress field for hierarchical structures, 2 to improve the structural design efficiency of MnNiSi-MTX alloys, 3 and to interpret the physical mechanism of the yield strength of polycrystalline metals, 4 to name just a few. Among these applications, the DL-aided microstructure design for composite materials is gaining more and more attention.…”

3D Printed Silicones with Shape Morphing and Low-Temperature Ultraelasticity

“…The rising three-dimensional (3D) printing technique, with its unparalleled freedom to create complex, customized geometries with low cost, shows great promise in controlling the internal morphologies and architectures of cellular materials. − Especially, 3D printing of silicones could be realized using direct ink writing (DIW), ,,,− inkjet printing, , embedded 3D printing, , vat polymerization, , and expanded techniques for higher resolution. , Mechanical responses of the printed foams could be well predicted, designed, and/or optimized by digital techniques such as simulation and machine learning , and further tailored by controlling the inner structure (such as the polymer network , and filler orientation , ) of the printed filaments. In addition, by introducing micro- or nanoscale pores in the 3D printed filaments using a sacrificial templating concept, − a hierarchical porous structure could be achieved, endowing the foam with ultraelasticity (i.e., extreme compressibility and cyclic endurance) and much enhanced active surface area compared to its nonhierarchical counterparts, , which is favorable for high-tech fields such as aerospace, energy, and bioengineering.…”

Synthesis and Characterization of a Liquid Crystal-Modified Polydimethylsiloxane Rubber with Mechanical Adaptability Based on Chain Extension in the Process of Crosslinking

“…The mechanical adaptive material is a key material in connection devices, artificial muscles, and other areas, which can effectively dissipate energy and suppress the increase of stress under continuous strain in a large deformation area. − Several attempts have been made to broaden the platform of the material’s mechanical adaptability and improve its mechanical strength. , We expect to introduce a novel energy dissipation mechanism from the molecular chain scale to further improve mechanical adaptability. Among them, liquid crystal elastomers have become the forefront of recent research due to their characteristic energy dissipation mechanism, which can undergo internal phase transition under external stimulation. − Recently, Yu et al developed 3D-printing liquid crystal elastomer foams to enhance energy dissipation under mechanical insult .…”

“… 1 − 6 Several attempts have been made to broaden the platform of the material’s mechanical adaptability and improve its mechanical strength. 7 , 8 We expect to introduce a novel energy dissipation mechanism from the molecular chain scale to further improve mechanical adaptability. Among them, liquid crystal elastomers have become the forefront of recent research due to their characteristic energy dissipation mechanism, which can undergo internal phase transition under external stimulation.…”