Stochastic reconstruction of heterogeneous microstructure combining sliced Wasserstein distance and gradient optimization
Zhenchuan Ma,Qizhi Teng,Pengcheng Yan,Xiaohong Wu,Xiaohai He
DOI: https://doi.org/10.1016/j.actamat.2024.120023
IF: 9.4
2024-05-20
Acta Materialia
Abstract:Computational reconstruction methods play an important role in integrated computational materials engineering, providing an efficient and inexpensive way for multi-modal and multi-length datasets construction, material design, and property prediction research. However, existing methods are either not computationally inefficient (traditional methods), or data-dependent and difficult to handle various reconstruction tasks (learning-based methods). Hybrid methods show advantages in addressing these problems, but they are still necessary to advance in low dataset dependence, generalization, and controllability. Herein, we propose a novel descriptor-based gradient optimization method, which utilizes multipoint statistical information and sliced Wasserstein metric for accurate microstructure characterization and reconstruction. Our method aims to respect and reproduce all the local patterns from training images, realizing similar higher-order statistics while better reproducing local features. To this end, sliced Wasserstein metric is introduced to measure the distance between higher-order distributions, and gradient optimization is adopted for fast and efficient reconstructions. Hierarchical and multitemplate strategies are developed to further strengthen characterization and reconstruction ability of heterogeneous structures. Additionally, our method can complete 2D-to-2D, 2D-to-3D, and 3D-to-3D reconstruction tasks even in the case of a single sample. Multiple sets of experiments are conducted under different reconstruction tasks for verification, and comparisons of visualization results, statistical parameters, and physical properties show the superiority of our method.
materials science, multidisciplinary,metallurgy & metallurgical engineering