Machine learning approaches for elastic localization linkages in high-contrast composite materials


There has been a growing recognition of the opportunities afforded by advanced data science and informatics approaches in addressing the computational demands of modeling and simulation of multiscale materials science phenomena. More specifically, the mining of microstructure–property relationships by various methods in machine learning and data mining opens exciting new opportunities that can potentially result in a fast and efficient material design. This work explores and presents multiple viable approaches for computationally efficient predictions of the microscale elastic strain fields in a three-dimensional (3-D) voxel-based microstructure volume element (MVE). Advanced concepts in machine learning and data mining, including feature extraction, feature ranking and selection, and regression modeling, are explored as data experiments. Improvements are demonstrated in a gradually escalated fashion achieved by (1) feature descriptors introduced to represent voxel neighborhood characteristics, (2) a reduced set of descriptors with top importance, and (3) an ensemble-based regression technique.

In Integrating Materials and Manufacturing Innovation (2015), Volume 4, Issue 1, pp 192–208.
@article{liu2015machine, title={Machine learning approaches for elastic localization linkages in high-contrast composite materials}, author={Liu, Ruoqian and Yabansu, Yuksel C and Agrawal, Ankit and Kalidindi, Surya R and Choudhary, Alok N}, journal={Integrating Materials and Manufacturing Innovation}, volume={4}, number={1}, pages={13}, year={2015}, publisher={Springer}}