Deep-Learning-Assisted Resonant Ultrasound Spectroscopy for Cubic Solids

Fukuda, Hiroki; Nagakubo, Akira; Wright, Oliver B.; Kyotani, Kazuhiro; Ogi, Hirotsugu

doi:https://doi.org/10.1103/physrevapplied.20.034048

Deep-Learning-Assisted Resonant Ultrasound Spectroscopy for Cubic Solids

Fukuda, Hiroki; Nagakubo, Akira; Wright, Oliver B. et al.

Physical Review Applied, 2023, 20, 034048

Number of Access:529（2025-10-09 11:46 Counts）

Identifier to cite or link to this item: https://hdl.handle.net/11094/92584

Link to primary information

File	Format	Terms of use	Size	Views	Date.Available	Description	information
PhysRevApplied_20_034048	pdf	None	5.61 MB	344	2023.09.28

Item Information

Output File Export EndNote Basic Export Mendeley

Title	Deep-Learning-Assisted Resonant Ultrasound Spectroscopy for Cubic Solids
Creator	CreatorName Fukuda, Hiroki
	NRID 1000070751113 1000070751113 ORCID 0000-0001-7494-5099 0000-0001-7494-5099 researchmap permalink nagakubo nagakubo Scopus Author ID 55329400300 55329400300 CreatorName Nagakubo, Akira
	CreatorName Wright, Oliver B.
	CreatorName Kyotani, Kazuhiro
	NRID 1000090252626 1000090252626 researchmap permalink read0042771 read0042771 Scopus Author ID 35563683500 35563683500 CreatorName Ogi, Hirotsugu
Description	Fukuda Hiroki, Nagakubo Akira, Wright Oliver B., et al. Deep-Learning-Assisted Resonant Ultrasound Spectroscopy for Cubic Solids. Physical Review Applied 20, 228 (2023); https://doi.org/10.1103/physrevapplied.20.034048.
Abstract	This paper solves a long-standing problem in resonant ultrasound spectroscopy (RUS) for cubic solids through the use of two-dimensional deep learning (DL). By means of inverse methods, conventional RUS can determine all independent elastic constants of a crystalline solid specimen by comparing measured and calculated free-vibration resonance frequencies. However, to avoid invalid local minima in the inverse process, good initial guesses of the elastic constants must be available. Here, we propose a DL scheme to remedy this problem, which utilizes an input elasticity image composed of three layers obtained from resonance frequency data. After network training, this scheme is executed in two steps: DL processing by a neural network to output elastic constants in a Blackman diagram classification, followed by the use of a regression network around the classified point in this diagram for more accurate determination. By means of simulations, we demonstrate that this DL scheme yields the principal elastic constants within an error of approximately 5% without any inverse processing even in the unfavorable case of five missing modes for 111 existing cubic crystals.
Publisher	American Physical Society (APS)
Source Title	Physical Review Applied
Volume (Issue)	20
Page	034048
Date of Issued	2023-09-21
Language	English
Handle URL	https://hdl.handle.net/11094/92584
eISSN	23317019
Relation.isIdenticalTo	DOI (Publisher Version) https://doi.org/10.1103/physrevapplied.20.034048
Access Rights	open access
CopyRight	Copyright 2023 by the American Physical Society
oaire:version	Version of Record
Category	学術雑誌論文 Journal Article

Resource Type	journal article
Local Type	Journal Article
DCMI Type	Text
DCTERMS.bibliographicCitation	Physical Review Applied. 2023, 20, p. 034048
DC.title	Deep-Learning-Assisted Resonant Ultrasound Spectroscopy for Cubic Solids
DC.creator	Fukuda, Hiroki
	Nagakubo, Akira
	Wright, Oliver B.
	Kyotani, Kazuhiro
	Ogi, Hirotsugu
DC.publisher	American Physical Society (APS)
DC.language' scheme='DCTERMS.RFC1766	English
DC.type' scheme='DCTERMS.DCMIType	Text
DCTERMS.issued' scheme='DCTERMS.W3CDTF	2023-09-21
DC.identifier	https://doi.org/10.1103/physrevapplied.20.034048
DC.description	Fukuda Hiroki, Nagakubo Akira, Wright Oliver B., et al. Deep-Learning-Assisted Resonant Ultrasound Spectroscopy for Cubic Solids. Physical Review Applied 20, 228 (2023); https://doi.org/10.1103/physrevapplied.20.034048.
DCTERMS.abstract	This paper solves a long-standing problem in resonant ultrasound spectroscopy (RUS) for cubic solids through the use of two-dimensional deep learning (DL). By means of inverse methods, conventional RUS can determine all independent elastic constants of a crystalline solid specimen by comparing measured and calculated free-vibration resonance frequencies. However, to avoid invalid local minima in the inverse process, good initial guesses of the elastic constants must be available. Here, we propose a DL scheme to remedy this problem, which utilizes an input elasticity image composed of three layers obtained from resonance frequency data. After network training, this scheme is executed in two steps: DL processing by a neural network to output elastic constants in a Blackman diagram classification, followed by the use of a regression network around the classified point in this diagram for more accurate determination. By means of simulations, we demonstrate that this DL scheme yields the principal elastic constants within an error of approximately 5% without any inverse processing even in the unfavorable case of five missing modes for 111 existing cubic crystals.
DC.rights	Copyright 2023 by the American Physical Society
DC.format	92584
citation_title	Deep-Learning-Assisted Resonant Ultrasound Spectroscopy for Cubic Solids
citation_author	Fukuda, Hiroki
	Nagakubo, Akira
	Wright, Oliver B.
	Kyotani, Kazuhiro
	Ogi, Hirotsugu
citation_publisher	American Physical Society (APS)
citation_language	English
citation_date	2023-09-21
citation_journal_title	Physical Review Applied
citation_volume	20
citation_firstpage	034048
citation_doi	https://doi.org/10.1103/physrevapplied.20.034048
citation_public_url	https://hdl.handle.net/11094/92584

Deep-Learning-Assisted Resonant Ultrasound Spectroscopy for Cubic Solids

Link to primary information

Item Information

Show details