Machine Learning and Computer Vision for Processing-Property Analyses

Samantha Daly
Mechanical Engineering, University of California Santa Barbara

ABSTRACT: Microstructure influences deformation and failure mechanisms, such as twinning, slip, grain boundary sliding, and multi-crack systems. This influence includes complex stochastic and deterministic factors whose interactions are currently under active debate. Daly discusses the application of machine learning and computer vision to microscale displacement data for the high-throughput segmentation and identification of deformation mechanisms, and their evolution under load across mm-scale fields of view. Twinning in magnesium is an example. Daly presents a recently developed experimental approach to obtain high-resolution, large FOV microscale deformation maps. Also discussed is the analysis of deformation twinning in Mg WE43 over thousands of grains in each individual test, including the relative activity of specific variants automatically identified from microscale strain fields. The newly developed experimental and analytical approaches are length-scale independent and material agnostic. The approaches can be modified to identify a range of deformation and failure mechanisms.

(recording not available)