Check out our recent peer-reviewed publications

Abstract

Improvements to synchrotron-based micro-computed tomography scanning capabilities have gifted researchers the ability to characterize 4D material thermomechanical responses more thoroughly than ever before. These advancements, however, have brought about new challenges in analyzing the resulting deluge of data. We report on a nickel-based superalloy specimen imaged 26 times in-situ during cyclic loading at Argonne National Laboratory Advanced Photon Source beamline 1ID, in order to monitor crack growth within the microstructure. Several deep learning approaches which utilize convolutional neural networks are implemented to segment crack features from reconstructed tomography scans. U-Net architecture implementations are found to be especially effective, achieving IoU = 0.995 ± 0.004 and Matthews correlation coefficient scores of ϕ = 0.826 ± 0.085. These advancements broaden possibilities for scientists seeking to automate segmentation analyses of similar large datasets.

Graphical abstract

first_publication_image.jpg

Citation:

D. Menasche, P. Shade, S. Safriet, P. Kenesei, J. Park, W. Musinski. Deep learning approaches to semantic segmentation of fatigue cracking within cyclically loaded nickel superalloy, Computational Materials Science, 198: 110683, 2021. DOI: 10.1016/j.commatsci.2021.110683.

Abstract

We describe 3D characterization of an additively manufactured Inconel 625 nickel-base superalloy specimen conducted during a uniaxial tension test using a suite of nondestructive x-ray techniques. High-energy diffraction microscopy in both near- and far-field modalities are employed in situ to track evolution of the material orientation and stress–strain fields at six points during the mechanical test, and these data streams are registered with micro-computed tomography reconstructions which probe the material density. This data volume was matched to a multi-modal serial sectioning characterization of the specimen taken after loading, described in this article’s companion. Twenty-eight grains which were monitored throughout the experiment were selected to form the basis for AFRL AM Modeling Series Challenge 4, Microscale Structure-to-Properties.

Graphical abstract

second_publication_image.png

Citation:

D. Menasche, W. Musinski, M. Obstalecki, M. Shah, S. Donegan, J. Bernier, P. Kenesei, J. Park, P. Shade. AFRL Additive Manufacturing Modeling Series: Challenge 4, In Situ Mechanical Test of an IN625 Sample with Concurrent High-Energy Diffraction Microscopy Characterization, Integrating Materials and Manufacturing Innovation, 2193-9772, 2021. DOI: 10.1007/s40192-021-00218-3

Additional Publications

Citation:

M. Chapman, M. Shah, S. Donegan, J. Scott, P. Shade, D. Menasche, M. Uchic. AFRL Additive Manufacturing Modeling Series: Challenge 4, 3D Reconstruction of an IN625 High-Energy Diffraction Microscopy Sample Using Multi-modal Serial Sectioning, Computational Materials Science, 10: 129-141, 2021. DOI: 10.1007/s40192-021-00212-9.

Citation:

D. Menasche, P. Shade, S. Safriet, R. Suter. Accuracy and precision of near-field high-energy diffraction microscopy forward-model-based microstructure reconstructions, Journal of Applied Crystallography, 53: 107-116, 2021. DOI: 10.1107/S1600576719016005.