Learning to predict 3D dislocation structures from X-ray patterns
Inferring Dislocation Microstructures from X-ray Diffraction via Cross-Modal Contrastive Learning
Machine Learning
Summary
X-ray diffraction patterns show how X-rays bounce off materials, but they only give indirect clues about tiny defects called dislocations inside. The authors developed a computer method that links simulated 3D dislocation structures with their X-ray patterns, so it can learn to guess the 3D defects just from the X-ray data. Their method works well on simulated examples, needing about 500 samples to accurately predict new structures. This could help scientists analyze materials more easily by using their X-ray data to find detailed internal defects.
What this means in practice
- •For materials engineers: Predict 3D dislocation structures directly from X-ray diffraction data to evaluate material properties without needing complex physical reconstruction methods.
- •For industrial nondestructive testing teams: Improve detection and analysis of internal defects in metals by rapidly inferring dislocation networks from standard diffraction scans in manufacturing quality control.
Tested on simulated data.
Authors
Benjamin Udofia, Nicolas Bertin, Markus Stricker
Abstract
Understanding and inferring dislocation microstructures from diffraction patterns remains an open challenge in materials characterization, as diffraction measurements provide only indirect information about the underlying dislocation structure. In this work, a cross-modal learning framework is developed to enable the prediction of 3D dislocation structures directly from diffraction data. Dislocation density fields generated from discrete dislocation dynamics simulations are paired with corresponding virtual X-ray diffraction patterns and embedded into a shared 2D latent space using contrastive learning. The alignment between structural and diffraction representations of dislocation structures is evaluated directly in the learned latent space using correlations between corresponding latent features. To estimate the role of dataset size for this approach, farthest point sampling is employed to construct representative and diverse training subsets of varying sizes. The results show strong cross-modal alignment and that model performance improves rapidly with increasing dataset size. Near-saturation is achieved with approximately 500 representative observations from a dataset of 10,000 observations, enabling accurate prediction of dislocation density fields from previously unseen diffraction data of the same distribution. Qualitative comparisons confirm that the predicted structures capture the dominant spatial features of the underlying dislocation microstructures. These findings demonstrate an efficient approach for learning structure-diffraction relationships and highlight the potential for inferring structural characteristics of dislocation networks directly from diffraction patterns, providing a pathway toward diffraction-based structural analysis and future extension to experimental data.