English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Microstructure is key to controlling and understanding the properties ofmetallic materials, but traditional approaches to describing microstructurecapture only a small number of features. To enable data-centric approaches tomaterials discovery, allow efficient storage of microstructural data and assistin quality control in metals processing, we require more complete descriptorsof microstructure. The concept of microstructural fingerprinting, using machinelearning (ML) to develop quantitative, low-dimensional descriptors ofmicrostructures, has recently attracted significant attention. However, it isdifficult to interpret conclusions drawn by ML algorithms, which are commonlyreferred to as "black boxes".  Here we explore variational autoencoders (VAEs), which can be trained toproduce microstructural fingerprints in a continuous latent space. VAEs enablethe reconstruction of images from fingerprints, allowing us to explore how keyfeatures of microstructure are encoded. We develop a VAE architecture based onResNet18 and train it on Ti-6Al-4V optical micrographs as an example of anindustrially important alloy where microstructural control is critical toperformance. The latent space is explored in several ways, including bysupplying interpolated and randomly perturbed fingerprints to the traineddecoder and via dimensionality reduction to explore the distribution ofmicrostructural features within the latent space of fingerprints. We show thatthe VAE fingerprints exhibit smooth, interpolable behaviour with stability tolocal perturbations, supporting their suitability as general purposedescriptors for microstructure. We also show that key properties of themicrostructures are strongly correlated with position in the latent space,supporting the use of VAE fingerprints for quantitative exploration ofprocess-structure-property relationships.

Exploring descriptors for titanium microstructure via digital fingerprints from variational autoencoders