A Deep Convolutional Neural Network-Based Method to Predict Accurate Fracture Strength of Poly-Crystalline Graphene

A data-driven deep convolution neural network model is used to predict the fracture properties of polycrystalline graphene from the atomic resolution image. A large dataset is prepared using molecular dynamic simulations and atomic resolution image of polycrystalline graphene. The results show that our model after training can predict the fracture properties of polycrystalline graphene with high accuracy.