Intragranular strain estimation in far-field scanning X-ray diffraction using a Gaussian process
(2021) In Journal of Applied Crystallography 54(4). p.1057-1070- Abstract
- A new method for estimation of intragranular strain fields in polycrystalline materials based on scanning three-dimensional X-ray diffraction (scanning 3DXRD) data is presented and evaluated. Given an a priori known anisotropic compliance, the regression method enforces the balance of linear and angular momentum in the linear elastic strain field reconstruction. By using a Gaussian process (GP), the presented method can yield a spatial estimate of the uncertainty of the reconstructed strain field. Furthermore, constraints on spatial smoothness can be optimized with respect to measurements through hyperparameter estimation. These three features address weaknesses discussed for previously existing scanning 3DXRD reconstruction methods and,... (More)
- A new method for estimation of intragranular strain fields in polycrystalline materials based on scanning three-dimensional X-ray diffraction (scanning 3DXRD) data is presented and evaluated. Given an a priori known anisotropic compliance, the regression method enforces the balance of linear and angular momentum in the linear elastic strain field reconstruction. By using a Gaussian process (GP), the presented method can yield a spatial estimate of the uncertainty of the reconstructed strain field. Furthermore, constraints on spatial smoothness can be optimized with respect to measurements through hyperparameter estimation. These three features address weaknesses discussed for previously existing scanning 3DXRD reconstruction methods and, thus, offer a more robust strain field estimation. The method is twofold validated: firstly by reconstruction from synthetic diffraction data, and secondly by reconstruction of a previously studied tin (Sn) grain embedded in a polycrystalline specimen. Comparison against reconstructions achieved by a recently proposed algebraic inversion technique is also presented. It is found that the GP regression consistently produces reconstructions with lower root-mean-square errors, mean absolute errors and maximum absolute errors across all six components of strain. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/cc34fd87-f6cb-4466-8d6e-be9fcf693fde
- author
- Henningsson, Axel LU and Hendriks, Johannes
- organization
- publishing date
- 2021-05-12
- type
- Contribution to journal
- publication status
- published
- subject
- keywords
- three-dimensional X-ray diffraction, 3DXRD, intragranular strain, Gaussian processes, scanning X-ray diffraction
- in
- Journal of Applied Crystallography
- volume
- 54
- issue
- 4
- article number
- 1600-5767
- pages
- 14 pages
- publisher
- International Union of Crystallography
- external identifiers
-
- pmid:34429719
- scopus:85120959538
- ISSN
- 1600-5767
- DOI
- 10.1107/S1600576721005112
- language
- English
- LU publication?
- yes
- id
- cc34fd87-f6cb-4466-8d6e-be9fcf693fde
- date added to LUP
- 2021-08-31 14:21:00
- date last changed
- 2022-04-27 03:31:30
@article{cc34fd87-f6cb-4466-8d6e-be9fcf693fde,
abstract = {{A new method for estimation of intragranular strain fields in polycrystalline materials based on scanning three-dimensional X-ray diffraction (scanning 3DXRD) data is presented and evaluated. Given an a priori known anisotropic compliance, the regression method enforces the balance of linear and angular momentum in the linear elastic strain field reconstruction. By using a Gaussian process (GP), the presented method can yield a spatial estimate of the uncertainty of the reconstructed strain field. Furthermore, constraints on spatial smoothness can be optimized with respect to measurements through hyperparameter estimation. These three features address weaknesses discussed for previously existing scanning 3DXRD reconstruction methods and, thus, offer a more robust strain field estimation. The method is twofold validated: firstly by reconstruction from synthetic diffraction data, and secondly by reconstruction of a previously studied tin (Sn) grain embedded in a polycrystalline specimen. Comparison against reconstructions achieved by a recently proposed algebraic inversion technique is also presented. It is found that the GP regression consistently produces reconstructions with lower root-mean-square errors, mean absolute errors and maximum absolute errors across all six components of strain.}},
author = {{Henningsson, Axel and Hendriks, Johannes}},
issn = {{1600-5767}},
keywords = {{three-dimensional X-ray diffraction; 3DXRD; intragranular strain; Gaussian processes; scanning X-ray diffraction}},
language = {{eng}},
month = {{05}},
number = {{4}},
pages = {{1057--1070}},
publisher = {{International Union of Crystallography}},
series = {{Journal of Applied Crystallography}},
title = {{Intragranular strain estimation in far-field scanning X-ray diffraction using a Gaussian process}},
url = {{http://dx.doi.org/10.1107/S1600576721005112}},
doi = {{10.1107/S1600576721005112}},
volume = {{54}},
year = {{2021}},
}