Elastic wave propagation on the macroscale can proffer details about the defects and microstructure of composites and alloys. Experiments suggest that an effective link between different material properties can be established and that underpinning the underlying small-scale mechanics can lead to the effective prediction of correlations and microstructure identification. Microstructures described by a large number of parameters and with a high degree of variability present a significant challenge due to the absence of reliable analytical homogenization techniques. In this project, we present a multiscale approach based on representative volume elements and data-driven Machine Learning techniques to capture the dynamical response of highly heterogeneous media. The model is solved with the Finite Element Method and implemented in MOOSE (Multiphysics Object-Oriented Simulation Environment), developed primarily at Idaho National Lab. We present a detailed procedure for linking the macro to micro-scale properties, extensively test our model, and validate the simulation results with experimental data.

Data-driven Homogenization Models for Non-Destructive Testing

The Virtual Technical Meeting of the Society of Engineering Science 2020 is the first fully virtual meeting of the SES community to envision and connect during the hiatus created by Covid-19, providing a platform for the exchange of ideas, and to showcase the latest research in the field of engineering science.

[Visit the Programme](https://drive.google.com/file/d/1BoCPBzrr1IQFANuzQI8a-paXCJoOcCkB/view?usp=sharing) | [Visit the Prerecorded Video/Poster List](https://drive.google.com/file/d/1XHLeEXJxEFjCSsNScuLDfMwy1X7oy7oJ/view?usp=sharing) | [View the Session Grid](https://drive.google.com/file/d/1VeDOC-Ev5m_ymYjXnjaXSZHR3_DARiLt/view?usp=sharing)