PolyCrysDiff: Controllable Generation of Three-Dimensional Computable Polycrystalline Material Structures

2026-03-12

AI Summary

This work addresses the challenge of controllably generating realistic and computationally viable three-dimensional polycrystalline microstructures to elucidate their structure–property relationships. To this end, it proposes the first end-to-end generative framework based on a conditional latent diffusion model, marking the inaugural application of conditional diffusion models to 3D polycrystalline microstructure modeling. The framework enables high-fidelity control over grain morphology, orientation distribution, and spatial correlations while ensuring physical validity and computational tractability of the generated structures. Validation via crystal plasticity finite element method (CPFEM) simulations and quantitative metrics (R² > 0.972) demonstrates that the generated microstructures significantly outperform existing approaches such as MRF and CNN in both morphological accuracy and statistical fidelity. Furthermore, the framework successfully enables systematic investigation of how grain characteristics influence mechanical properties.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Natural Language Processing: Generation

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

The three-dimensional (3D) microstructures of polycrystalline materials exert a critical influence on their mechanical and physical properties. Realistic, controllable construction of these microstructures is a key step toward elucidating structure-property relationships, yet remains a formidable challenge. Herein, we propose PolyCrysDiff, a framework based on conditional latent diffusion that enables the end-to-end generation of computable 3D polycrystalline microstructures. Comprehensive qualitative and quantitative evaluations demonstrate that PolyCrysDiff faithfully reproduces target grain morphologies, orientation distributions, and 3D spatial correlations, while achieving an $R^2$ over 0.972 on grain attributes (e.g., size and sphericity) control, thereby outperforming mainstream approaches such as Markov random field (MRF)- and convolutional neural network (CNN)-based methods. The computability and physical validity of the generated microstructures are verified through a series of crystal plasticity finite element method (CPFEM) simulations. Leveraging PolyCrysDiff's controllable generative capability, we systematically elucidate how grain-level microstructural characteristics affect the mechanical properties of polycrystalline materials. This development is expected to pave a key step toward accelerated, data-driven optimization and design of polycrystalline materials.

Problem

• polycrystalline materials
• 3D microstructure generation
• structure-property relationships
• controllable generation
• computable microstructures

Innovation

• conditional latent diffusion
• 3D polycrystalline microstructure
• controllable generation
• crystal plasticity FEM
• structure-property relationship