TY  - JOUR
A1  - Evsevleev, Sergei
A1  - Paciornik, Sidnei
A1  - Bruno, Giovanni
T1  - Advanced deep learning-based 3D microstructural characterization of multiphase metal matrix composites
JF  - Advanced engineering materials
N2  - The quantitative analysis of microstructural features is a key to understanding the micromechanical behavior of metal matrix composites (MMCs), which is a premise for their use in practice. Herein, a 3D microstructural characterization of a five-phase MMC is performed by synchrotron X-ray computed tomography (SXCT). A workflow for advanced deep learning-based segmentation of all individual phases in SXCT data is shown using a fully convolutional neural network with U-net architecture. High segmentation accuracy is achieved with a small amount of training data. This enables extracting unprecedently precise microstructural parameters (e.g., volume fractions and particle shapes) to be input, e.g., in micromechanical models.
KW  - computed tomography
KW  - convolutional neural networks
KW  - deep learning
KW  - metal
KW  - matrix composites
KW  - segmentations
Y1  - 2020
U6  - https://doi.org/10.1002/adem.201901197
SN  - 1438-1656
SN  - 1527-2648
VL  - 22
IS  - 4
PB  - Wiley-VCH
CY  - Weinheim
ER  - 
TY  - JOUR
A1  - Mishurova, Tatiana
A1  - Bruno, Giovanni
A1  - Evsevleev, Sergei
A1  - Sevostianov, Igor
T1  - Determination of macroscopic stress from diffraction experiments
BT  - a critical discussion
JF  - Journal of applied physics
N2  - The paper is motivated by some inconsistencies and contradictions present in the literature on the calculation of the so-called diffraction elastic constants. In an attempt at unifying the views that the two communities of Materials Science and Mechanics of Materials have on the subject, we revisit and define the terminology used in the field. We also clarify the limitations of the commonly used approaches and show that a unified methodology is also applicable to textured materials with a nearly arbitrary grain shape. We finally compare the predictions based on this methodology with experimental data obtained by in situ synchrotron radiation diffraction on additively manufactured Ti-6Al-4V alloy. We show that (a) the transverse isotropy of the material yields good agreement between the best-fit isotropy approximation (equivalent to the classic Kroner's model) and the experimental data and (b) the use of a general framework allows the calculation of all components of the tensor of diffraction elastic constants, which are not easily measurable by diffraction methods. This allows us to extend the current state-of-the-art with a predictive tool.
Y1  - 2020
U6  - https://doi.org/10.1063/5.0009101
SN  - 0021-8979
SN  - 1089-7550
VL  - 128
IS  - 2
PB  - American Institute of Physics
CY  - Melville
ER  -