Grain Microstructures

Grain microstructures, the arrangement and properties of individual grains within a material, significantly influence macroscopic material behavior. Current research focuses on computationally efficient methods for generating and analyzing these structures, employing machine learning techniques like generative adversarial networks (GANs), Markov Junior, and U-Net convolutional neural networks to create realistic virtual microstructures and predict material properties such as stress and phase evolution. These advancements accelerate simulations, enabling faster material design and optimization, particularly for applications in battery technology and metal manufacturing. The resulting speedups and improved accuracy offer significant advantages over traditional methods like finite element analysis.

Papers

July 7, 2024

Generating multi-scale NMC particles with radial grain architectures using spatial stochastics and GANs
Lukas Fuchs, Orkun Furat, Donal P. Finegan, Jeffery Allen, Francois L. E. Usseglio-Viretta, Bertan Ozdogru, Peter J. Weddle, Kandler Smith, Volker Schmidt
Generative Adversarial Network GAN Model Lithium Ion Battery Particle Cloud Grain Microstructures

November 21, 2023

Procedural Generation of Grain Orientations using the Wave Function Collapse Algorithm
G. Magny-Fokam, D. Madisetti, J. El-Awady
Procedural Content Generation Voronoi Tessellation Wave Function Electron Back Scattered Diffraction Grain Microstructures

March 15, 2023

Adapting U-Net for linear elastic stress estimation in polycrystal Zr microstructures
J. D. Langcaster, D. S. Balint, M. R. Wenman
U Net Critical Weld Key Performance Characteristic Grain Microstructures

October 31, 2022

Comparison of two artificial neural networks trained for the surrogate modeling of stress in materially heterogeneous elastoplastic solids
Sarthak Kapoor, Jaber Rezaei Mianroodi, Mohammad Khorrami, Nima S. Siboni, Bob Svendsen
Neural Network Consistent Comparison Surrogate Modeling Microstructure Model Continuous Chronic Stress Heterogeneous Material Grain Microstructures

May 4, 2022

Accelerating phase-field-based simulation via machine learning
Iman Peivaste, Nima H. Siboni, Ghasem Alahyarizadeh, Reza Ghaderi, Bob Svendsen, Dierk Raabe, Jaber R. Mianroodi
Machine Learning Simulation Study Microstructure Model 3D Microstructures Phase Field Grain Microstructures

Grain Microstructures

Papers

Generating multi-scale NMC particles with radial grain architectures using spatial stochastics and GANs

Procedural Generation of Grain Orientations using the Wave Function Collapse Algorithm

Adapting U-Net for linear elastic stress estimation in polycrystal Zr microstructures

Comparison of two artificial neural networks trained for the surrogate modeling of stress in materially heterogeneous elastoplastic solids

Accelerating phase-field-based simulation via machine learning