Multiscale modeling of heterogeneous materials using mechanics, imaging and machine learning

Many natural and synthetic materials, e.g. composites, cementitious materials and geological materials are heterogeneous and multi-scale in nature. They typically demonstrate a nonlinear and path-dependent behavior, which is determined by the properties, shape, and distribution of the constituents at the microscopic scale. While analysis of the mechanical behavior of these materials is often performed using phenomenological continuum-scale constitutive models, multiscale models are needed to fully capture the complex nonlinear material behavior. Some of the major challenges associated with multi-scale modeling of systems made of heterogeneous materials include computational costs of high-fidelity simulations, transferring the information across different length scales, and characterization of the material microstructure. In this talk, we present the recent advances in multi-scale modeling of heterogenous materials aided by advanced imaging techniques and machine learning, which have emerged as a promising alternative to the traditional constitutive models to enhance simulations of complex systems.