[DSRP Evidence](https://dsrpevidence.org/)

# Prediction of grain boundary energy and understanding of grain boundary structure-energy relationship by machine learning

Lu et al., 2026, Acta Mechanica Sinica — Chemistry

Patterns: [Distinctions](https://dsrpevidence.org/pattern/distinctions)

## In short

The paper treats the grain boundary itself as the fundamental distinction-line separating two crystal orientation domains, and shows that the geometry of that distinction alone determines a measurable physical property.

## What they found (results)

Machine learning models trained on atomistic simulations of about 100,000 aluminum grain boundaries predicted grain-boundary energy 10.6-12.9% more accurately than established theoretical models based on the boundary's structural parameters.

These researchers were not testing DSRP. The finding is theirs; the correspondence to DSRP is drawn by this site.

[Source](https://doi.org/10.1007/s10409-026-51015-x)