Accelerated search and design of stretchable graphene kirigami using machine learning

Abstract

Making kirigami-inspired cuts into a sheet has been shown to be an effective way of designing stretchable materials with metamorphic properties where the 2D shape can transform into complex 3D shapes. However, finding the optimal solutions is not straightforward as the number of possible cutting patterns grows exponentially with system size. Here, we report on how machine learning (ML) can be used to approximate the target properties, such as yield stress and yield strain, as a function of cutting pattern. Our approach enables the rapid discovery of kirigami designs that yield extreme stretchability as verified by molecular dynamics (MD) simulations. We find that convolutional neural networks (CNN), commonly used for classification in vision tasks, can be applied for regression to achieve an accuracy close to the precision of the MD simulations. This approach can then be used to search for optimal designs that maximize elastic stretchability with only 1000 training samples in a large design space of $\sim 4\times 10^6$ candidate designs. This example demonstrates the power and potential of ML in finding optimal kirigami designs at a fraction of iterations that would be required of a purely MD or experiment-based approach, where no prior knowledge of the governing physics is known or available.

References in corpus (12)

• Scikit-learn: Machine Learning in Python
• Machine learning phases of matter
• Discovery of low thermal conductivity compounds with first-principles anharmonic lattice dynamics calculations and Bayesian optimization
• Machine learning quantum phases of matter beyond the fermion sign problem
• Buckling-Induced Kirigami
• Initial rigid response and softening transition of highly stretchable kirigami sheet materials
• Polarization and valley switching in monolayer group-IV monochalcogenides
• Highly Deformable Graphene Kirigami
• The Relationship Between Local Structure and Relaxation in Out-of-Equilibrium Glassy Systems
• Kirigami Actuators
• Highly Stretchable MoS${}_2$ Kirigami
• Graphene and its elemental analogue: A molecular dynamics view of fracture phenomenon

Cited by in corpus (20)

• Automatic Design of Mechanical Metamaterial Actuators
• Supervised learning in a mechanical system
• Forward and Inverse Design of Kirigami via Supervised Autoencoder
• Drawing Phase Diagrams of Random Quantum Systems by Deep Learning the Wave Functions
• Design of pseudo-mechanisms and multistable units for mechanical metamaterials
• Theorem on the Compatibility of Spherical Kirigami Tessellations
• Predicting Mechanical Properties from Microstructure Images in Fiber-reinforced Polymers using Convolutional Neural Networks
• Valley notch filter in a graphene strain superlattice: Green's function and machine learning approach
• An Interpretable Machine Learning Model for Deformation of Multi-Walled Carbon Nanotubes
• Thermal buckling and symmetry breaking in thin ribbons under compression
• Elastocaloric Effect in Graphene Kirigami
• Anomalous thermal expansion in Ising-like puckered sheets
• Deep learning Local Reduced Density Matrices for Many-body Hamiltonian Estimation
• Exploration of mechanical, thermal conductivity and electromechanical properties of graphene nanoribbon springs
• Vibrations and tunneling of strained nanoribbons at finite temperature
• Predicting the mechanical response of oligocrystals with deep learning
• Exploring the potential of transfer learning for metamodels of heterogeneous material deformation
• Developing Artificial Mechanics Intuitions from Extremely Small Data
• Liquified protein vibrations, classification and cross-paradigm de novo image generation using deep neural networks
• Tunable Topological Phases in Quantum Kirigamis