Deep Learning Approach to Mechanical Property Prediction of Single-Network Hydrogel

Hydrogel has a complex network structure with inhomogeneous and random distribution of polymer chains. Much effort been paid to fully understand the relationship between mesoscopic macroscopic mechanical properties hydrogels. In this paper, we develop deep learning approach predict hydrogels from structures. First, structural models are constructed scale using self-avoiding walk method. The model is similar real hydrogel network. Then, two proposed capture nonlinear mapping its macroscale property. A neural 3D convolutional containing physical information implemented nominal stress–stretch curves under uniaxial tension. Our results show that end-to-end framework can effectively within wide range structures, which demonstrates able internal structures properties. We hope provide guidance design material property different soft materials.