Discovery of Cellular Unit Cells With High Natural Frequency and Energy Absorption Capabilities by an Inverse Machine Learning Framework

Abstract

Cellular materials have been widely used in load carrying lightweight structures. Although lightweight increases natural frequency, low stiffness of cellular structures reduces natural frequency. Designing structures with higher natural frequency can usually avoid resonance. In addition, because of the less amount of materials used in cellular structures, the energy absorption capability usually decreases such as under impact loading. Therefore, designing cellular structures with higher natural frequency and higher energy absorption capability is highly desired. In this study, machine learning and novel inverse design techniques enable to search a huge space of unexplored structural designs. In this study, machine learning regression and Generative Neural Networks (GANs) were used to form an inverse design framework. Optimal cellular unit cells that surpass the performance of biomimetic structures inspired from honeycomb, plant stems and trabecular bone in terms of natural frequency and impact resistance were discovered using machine learning. The discovered optimal cellular unit cells exhibited 30–100% higher natural frequency and 300% higher energy absorption than those of the biomimetic counterparts. The discovered optimal unit cells were validated through experimental and simulation comparisons. The machine learning framework in this study would help in designing load carrying engineering structures with increased natural frequency and enhanced energy absorption capability.

Insight in thermomechanical constitutive modeling of shape memory polymers

“…Shafe et al (2022) studied the effect of atomistic fingerprints on thermomechanical properties of epoxy-diamine thermoset shape memory polymers, which facilitates machine learning discovery of TSMPs. Recently, we also established both forward and inverse machine learning frameworks and discovered quite a few columns, lattice unit cells, and thin-walled structures with much improved structural capacities, including structures three-dimensional-printed by SMPs (Challapalli and Li, 2020;Challapalli et al, 2021a;Challapalli et al, 2021b;Challapalli and Li, 2021). Following the same line, it is expected that machine learning may also be used to establish the constitutive laws for SMPs.…”

A laminated vitrimer composite with strain sensing, delamination self-healing, deicing, and room-temperature shape restoration properties

“…From the literature pertaining to impact tolerance of composite laminates, initiation energy and propagation energy are the predominant and most significant parameters. From Li et al 15,32,33 and Agarwal et al, 34 the initiation energy is a measurement to evaluate the elastic energy transfer capability, and the propagation energy is the energy absorbed by the target through plastic deformation and damage. For brittle materials, the propagation energy is primarily absorbed by damage.…”

Overcoming the barrier: designing novel thermally robust shape memory vitrimers by establishing a new machine learning framework

“…37 In the field of metamaterials, Adithya et al designed new cellular structures and lattice structures with improved properties (load-carrying capacity, natural frequency, energy absorption capabilities) using a generative adversarial network (GAN). 38,39 Despite the previous successes, no ML-based methods have been applied to the field of vitrimer discovery, hence we decided to employ a design framework based on ML to address the gap in this field.…”