Learning the dynamics of metamaterials from diffracted waves with convolutional neural networks

Scattered elastic waves provide non-invasive diagnostics and dynamic characterization of metamaterials, but extracting information from small-size samples is challenging. Here, convolutional neural networks are used to interpret diffracted waves, revealing how sample-edge scattering provides the most significant information on macroscopic metamaterial properties. Conventional methods used to identify the dynamical properties of unknown media from scattered mechanical waves rely on analytical or numerical manipulations of the wave equation. These methods show their limitations in scenarios where the analyzed medium is moderately sized and the diffraction from the material edges influences the scattered fields significantly, such as non-destructive diagnostics and metamaterial characterization. Here, we show that convolutional neural networks can interpret the diffracted fields and learn the mapping between the scattered fields and all the effective material parameters including mass density and stiffness tensors from a small set of numerical simulations. Furthermore, networks trained with synthetic data can process physical measurements and are very robust to measurement errors. More importantly, the trained network provides insight into the dynamic behavior of matter including quantitative measures of the scattered field sensitivity to each material property and how the sensitivity changes depending on the material under test.

Automated summary

Convolutional neural networks can interpret diffracted waves and learn the mapping between scattered fields and effective material parameters. It is a powerful tool for non-invasive diagnostics and dynamic characterization of metamaterials, providing valuable insights into the dynamic behavior of matter.