Accelerated multiscale mechanics modeling in a deep learning framework

Microstructural heterogeneity affects the macro-scale behavior of materials. Conversely, load distribution at changes microstructural response. These up-scaling and down-scaling relations are often modeled using multiscale finite element (FE) approaches such as FE-squared (FE2). However, FE2 requires numerous calculations micro-scale, which renders this approach intractable. This paper reports an enormously faster machine learning (ML) based for mechanics modeling. The proposed ML-driven analysis uses ML-model that predicts local stress tensor fields in a linear elastic fiber-reinforced composite microstructure. ML-model, specifically U-Net deep convolutional neural network (CNN), is trained separately to perform mapping between spatial arrangement fibers corresponding 2D fields. provides effective material properties subsequent framework. Several numerical examples demonstrate substantial reduction computational cost when compared with traditional modeling full-scale FE analysis, homogenization analysis. has tremendous potential efficient complex heterogeneous materials, applications uncertainty quantification, design, optimization.