A deep learning approach to the inverse problem of modulus identification in elasticity
The inverse elasticity problem of identifying elastic modulus distribution based on measured displacement/strain fields plays a key role in various non-destructive evaluation (NDE) techniques used in geological exploration, quality control, and medical diagnosis (e.g., elastography). Conventional me...
Saved in:
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Article |
Language: | English |
Published: |
2021
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/148674 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-148674 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1486742021-09-10T01:05:57Z A deep learning approach to the inverse problem of modulus identification in elasticity Ni, Bo Gao, Huajian School of Mechanical and Aerospace Engineering Institute of High Performance Computing, A*STAR Engineering::Mechanical engineering Diagnosis Elastic Moduli The inverse elasticity problem of identifying elastic modulus distribution based on measured displacement/strain fields plays a key role in various non-destructive evaluation (NDE) techniques used in geological exploration, quality control, and medical diagnosis (e.g., elastography). Conventional methods in this field are often computationally costly and cannot meet the increasing demand for real-time and high-throughput solutions for advanced manufacturing and clinical practices. Here, we propose a deep learning (DL) approach to address this challenge. By constructing representative sampling spaces of shear modulus distribution and adopting a conditional generative adversarial net, we demonstrate that the DL model can learn high-dimensional mapping between strain and modulus via training over a limited portion of the sampling space. The proposed DL approach bypasses the costly iterative solver in conventional methods and can be rapidly deployed with high accuracy, making it particularly suitable for applications such as real-time elastography and highthroughput NDE techniques. The authors acknowledge the support by the National Science Foundation (NSF) under the grant CMMI-1634492. The simulations were performed on resources provided by the Extreme Science and Engineering Discovery Environment (XSEDE) through grant MSS090046 and at the Center for Computation and Visualization at Brown University. 2021-09-10T01:05:12Z 2021-09-10T01:05:12Z 2021 Journal Article Ni, B. & Gao, H. (2021). A deep learning approach to the inverse problem of modulus identification in elasticity. MRS Bulletin, 46, 19-25. https://dx.doi.org/10.1557/s43577-020-00006-y 0883-7694 https://hdl.handle.net/10356/148674 10.1557/s43577-020-00006-y 46 19 25 en MRS Bulletin © 2021 Materials Research Society. All rights reserved. |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Mechanical engineering Diagnosis Elastic Moduli |
spellingShingle |
Engineering::Mechanical engineering Diagnosis Elastic Moduli Ni, Bo Gao, Huajian A deep learning approach to the inverse problem of modulus identification in elasticity |
description |
The inverse elasticity problem of identifying elastic modulus distribution based on measured displacement/strain fields plays a key role in various non-destructive evaluation (NDE) techniques used in geological exploration, quality control, and medical diagnosis (e.g., elastography). Conventional methods in this field are often computationally costly and cannot meet the increasing demand for real-time and high-throughput solutions for advanced manufacturing and clinical practices. Here, we propose a deep learning (DL) approach to address this challenge. By constructing representative sampling spaces of shear modulus distribution and adopting a conditional generative adversarial net, we demonstrate that the DL model can learn high-dimensional mapping between strain and modulus via training over a limited portion of the sampling space. The proposed DL approach bypasses the costly iterative solver in conventional methods and can be rapidly deployed with high accuracy, making it particularly suitable for applications such as real-time elastography and highthroughput NDE techniques. |
author2 |
School of Mechanical and Aerospace Engineering |
author_facet |
School of Mechanical and Aerospace Engineering Ni, Bo Gao, Huajian |
format |
Article |
author |
Ni, Bo Gao, Huajian |
author_sort |
Ni, Bo |
title |
A deep learning approach to the inverse problem of modulus identification in elasticity |
title_short |
A deep learning approach to the inverse problem of modulus identification in elasticity |
title_full |
A deep learning approach to the inverse problem of modulus identification in elasticity |
title_fullStr |
A deep learning approach to the inverse problem of modulus identification in elasticity |
title_full_unstemmed |
A deep learning approach to the inverse problem of modulus identification in elasticity |
title_sort |
deep learning approach to the inverse problem of modulus identification in elasticity |
publishDate |
2021 |
url |
https://hdl.handle.net/10356/148674 |
_version_ |
1710686940815163392 |