A physics-inspired neural network to solve partial differential equations - application in diffusion-induced stress

Analyzing and predicting diffusion-induced stress are of paramount importance in understanding the structural durability of lithium- and sodium-ion batteries, which generally require solving initial-boundary value problems, involving partial differential equations (PDEs) for mechanical equilibrium a...

Full description

Saved in:
Bibliographic Details
Published in:Physical chemistry chemical physics : PCCP Vol. 24; no. 13; p. 7937
Main Authors: Xue, Yuan, Li, Yong, Zhang, Kai, Yang, Fuqian
Format: Journal Article
Language:English
Published: England 30.03.2022
ISSN:1463-9084, 1463-9084
Online Access:Get more information
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Analyzing and predicting diffusion-induced stress are of paramount importance in understanding the structural durability of lithium- and sodium-ion batteries, which generally require solving initial-boundary value problems, involving partial differential equations (PDEs) for mechanical equilibrium and mass transport. Due to the complexity and nonlinear characteristics of the initial-boundary value problems, numerical methods, such as finite difference, finite element, spectral analysis, and so forth, have been used. In this work, we propose two whole loss functions as the sum of the residuals of the PDEs, initial conditions and boundary conditions for the problems with decoupling and coupling between diffusion and stress, respectively, and apply a physics-inspired neural network under the framework of DeepXDE to solve diffusion-induced stress in an elastic sphere in contrast to traditional numerical methods. Using time-space coordinates as inputs and displacement and the solute concentration as outputs of artificial neural networks, we solve the spatiotemporal evolution of the displacement and the solute concentration in the elastic sphere for both the decoupling and coupling problems. The numerical results from the physics-inspired neural network are validated by analytical solutions and a finite element simulation using the COMSOL package. The method developed in this work opens an approach to analyze the stress evolution in electrodes due to electrochemical cycling.
AbstractList Analyzing and predicting diffusion-induced stress are of paramount importance in understanding the structural durability of lithium- and sodium-ion batteries, which generally require solving initial-boundary value problems, involving partial differential equations (PDEs) for mechanical equilibrium and mass transport. Due to the complexity and nonlinear characteristics of the initial-boundary value problems, numerical methods, such as finite difference, finite element, spectral analysis, and so forth, have been used. In this work, we propose two whole loss functions as the sum of the residuals of the PDEs, initial conditions and boundary conditions for the problems with decoupling and coupling between diffusion and stress, respectively, and apply a physics-inspired neural network under the framework of DeepXDE to solve diffusion-induced stress in an elastic sphere in contrast to traditional numerical methods. Using time-space coordinates as inputs and displacement and the solute concentration as outputs of artificial neural networks, we solve the spatiotemporal evolution of the displacement and the solute concentration in the elastic sphere for both the decoupling and coupling problems. The numerical results from the physics-inspired neural network are validated by analytical solutions and a finite element simulation using the COMSOL package. The method developed in this work opens an approach to analyze the stress evolution in electrodes due to electrochemical cycling.
Analyzing and predicting diffusion-induced stress are of paramount importance in understanding the structural durability of lithium- and sodium-ion batteries, which generally require solving initial-boundary value problems, involving partial differential equations (PDEs) for mechanical equilibrium and mass transport. Due to the complexity and nonlinear characteristics of the initial-boundary value problems, numerical methods, such as finite difference, finite element, spectral analysis, and so forth, have been used. In this work, we propose two whole loss functions as the sum of the residuals of the PDEs, initial conditions and boundary conditions for the problems with decoupling and coupling between diffusion and stress, respectively, and apply a physics-inspired neural network under the framework of DeepXDE to solve diffusion-induced stress in an elastic sphere in contrast to traditional numerical methods. Using time-space coordinates as inputs and displacement and the solute concentration as outputs of artificial neural networks, we solve the spatiotemporal evolution of the displacement and the solute concentration in the elastic sphere for both the decoupling and coupling problems. The numerical results from the physics-inspired neural network are validated by analytical solutions and a finite element simulation using the COMSOL package. The method developed in this work opens an approach to analyze the stress evolution in electrodes due to electrochemical cycling.Analyzing and predicting diffusion-induced stress are of paramount importance in understanding the structural durability of lithium- and sodium-ion batteries, which generally require solving initial-boundary value problems, involving partial differential equations (PDEs) for mechanical equilibrium and mass transport. Due to the complexity and nonlinear characteristics of the initial-boundary value problems, numerical methods, such as finite difference, finite element, spectral analysis, and so forth, have been used. In this work, we propose two whole loss functions as the sum of the residuals of the PDEs, initial conditions and boundary conditions for the problems with decoupling and coupling between diffusion and stress, respectively, and apply a physics-inspired neural network under the framework of DeepXDE to solve diffusion-induced stress in an elastic sphere in contrast to traditional numerical methods. Using time-space coordinates as inputs and displacement and the solute concentration as outputs of artificial neural networks, we solve the spatiotemporal evolution of the displacement and the solute concentration in the elastic sphere for both the decoupling and coupling problems. The numerical results from the physics-inspired neural network are validated by analytical solutions and a finite element simulation using the COMSOL package. The method developed in this work opens an approach to analyze the stress evolution in electrodes due to electrochemical cycling.
Author Li, Yong
Xue, Yuan
Yang, Fuqian
Zhang, Kai
Author_xml – sequence: 1
  givenname: Yuan
  surname: Xue
  fullname: Xue, Yuan
  email: clyong1991@seu.edu.cn
  organization: Jiangsu Key Laboratory of Engineering Mechanics, School of Civil Engineering, Southeast University, Nanjing, Jiangsu 210096, China. clyong1991@seu.edu.cn
– sequence: 2
  givenname: Yong
  orcidid: 0000-0001-7607-0994
  surname: Li
  fullname: Li, Yong
  email: clyong1991@seu.edu.cn
  organization: Jiangsu Key Laboratory of Engineering Mechanics, School of Civil Engineering, Southeast University, Nanjing, Jiangsu 210096, China. clyong1991@seu.edu.cn
– sequence: 3
  givenname: Kai
  surname: Zhang
  fullname: Zhang, Kai
  email: zhangkai@tongji.edu.cn
  organization: School of Aerospace Engineering and Applied Mechanics, Tongji University, Shanghai 200092, China. zhangkai@tongji.edu.cn
– sequence: 4
  givenname: Fuqian
  orcidid: 0000-0001-6277-3082
  surname: Yang
  fullname: Yang, Fuqian
  email: fuqian.yang@uky.edu
  organization: Materials Program, Department of Chemical and Materials EngineeringUniversity of Kentucky, Lexington, KY 40506, USA. fuqian.yang@uky.edu
BackLink https://www.ncbi.nlm.nih.gov/pubmed/35311865$$D View this record in MEDLINE/PubMed
BookMark eNpNkMtOwzAQRS1URB-w4QOQl2wCdhyn9rKqeEmV2MA6cpwJGFLH9SSg_j1WKRKrO0dzdBd3Tia-90DIJWc3nAl923AbWKG0eDshM16UItNMFZN_95TMET8YY1xycUamQgrOVSlnBFc0vO_RWcycx-AiNNTDGE2XYvju4ycdeop99wU0mDi49Ghc20IEfwDYjWZwvUeaURNC5-wBqfMHb8QEqboZbWrGIQLiOTltTYdwccwFeb2_e1k_Zpvnh6f1apNZsRRD1ipZ59DmAEvQOctL0ZhWmxJ4zWprWyOtlIoVQtuGS61tqaxmApacG2DK5Aty_dsbYr8bAYdq69BC1xkP_YhVXhZpD5UmTOrVUR3rLTRViG5r4r76Gyr_Adgrbmk
CitedBy_id crossref_primary_10_1007_s00366_023_01914_8
crossref_primary_10_1016_j_apm_2025_115986
crossref_primary_10_1016_j_heliyon_2023_e15076
crossref_primary_10_1016_j_ssc_2023_115244
crossref_primary_10_1186_s13661_025_02021_x
crossref_primary_10_1002_advs_202306604
crossref_primary_10_1016_j_oceaneng_2023_114684
crossref_primary_10_1016_j_est_2023_107037
crossref_primary_10_1016_j_etran_2025_100420
crossref_primary_10_1016_j_ijengsci_2023_103841
crossref_primary_10_1515_zna_2024_0009
crossref_primary_10_1016_j_ijmecsci_2024_109296
ContentType Journal Article
DBID NPM
7X8
DOI 10.1039/d1cp04893g
DatabaseName PubMed
MEDLINE - Academic
DatabaseTitle PubMed
MEDLINE - Academic
DatabaseTitleList PubMed
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod no_fulltext_linktorsrc
Discipline Chemistry
EISSN 1463-9084
ExternalDocumentID 35311865
Genre Journal Article
GroupedDBID ---
-DZ
-JG
-~X
0-7
0R~
123
29O
4.4
53G
705
70~
7~J
87K
AAEMU
AAIWI
AAJAE
AAMEH
AANOJ
AAWGC
AAXHV
AAXPP
ABASK
ABDVN
ABEMK
ABJNI
ABPDG
ABRYZ
ABXOH
ACGFO
ACGFS
ACIWK
ACLDK
ACNCT
ADMRA
ADSRN
AEFDR
AENEX
AENGV
AESAV
AETIL
AFLYV
AFOGI
AFRDS
AFRZK
AFVBQ
AGEGJ
AGKEF
AGRSR
AGSTE
AHGCF
ALMA_UNASSIGNED_HOLDINGS
ANUXI
APEMP
ASKNT
AUDPV
AZFZN
BLAPV
BSQNT
C6K
CS3
D0L
DU5
EBS
ECGLT
EE0
EF-
F5P
GGIMP
GNO
H13
HZ~
H~N
IDZ
J3G
J3I
M4U
N9A
NHB
NPM
O9-
OK1
P2P
R7B
R7C
RAOCF
RCNCU
RNS
RPMJG
RRA
RRC
RSCEA
SKA
SKF
SLH
TN5
TWZ
UCJ
UHB
VH6
WH7
YNT
2WC
7X8
AAFBY
AGMRB
AKMSF
ALUYA
R56
ID FETCH-LOGICAL-c373t-f85b2ef2ee7e920263daf9a6e1b0bccfa5c5580439cd1599c68c903e711ae08a2
IEDL.DBID 7X8
ISICitedReferencesCount 17
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000771087800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1463-9084
IngestDate Sun Nov 09 12:06:08 EST 2025
Wed Feb 19 02:27:23 EST 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 13
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c373t-f85b2ef2ee7e920263daf9a6e1b0bccfa5c5580439cd1599c68c903e711ae08a2
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0001-6277-3082
0000-0001-7607-0994
OpenAccessLink https://pubs.rsc.org/en/content/articlepdf/2022/cp/d1cp04893g
PMID 35311865
PQID 2641518039
PQPubID 23479
ParticipantIDs proquest_miscellaneous_2641518039
pubmed_primary_35311865
PublicationCentury 2000
PublicationDate 20220330
PublicationDateYYYYMMDD 2022-03-30
PublicationDate_xml – month: 3
  year: 2022
  text: 20220330
  day: 30
PublicationDecade 2020
PublicationPlace England
PublicationPlace_xml – name: England
PublicationTitle Physical chemistry chemical physics : PCCP
PublicationTitleAlternate Phys Chem Chem Phys
PublicationYear 2022
SSID ssj0001513
Score 2.4677284
Snippet Analyzing and predicting diffusion-induced stress are of paramount importance in understanding the structural durability of lithium- and sodium-ion batteries,...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
StartPage 7937
Title A physics-inspired neural network to solve partial differential equations - application in diffusion-induced stress
URI https://www.ncbi.nlm.nih.gov/pubmed/35311865
https://www.proquest.com/docview/2641518039
Volume 24
WOSCitedRecordID wos000771087800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LS8QwEA7qCnrx_VhfRPAaNk2aNjmJLC4edNmDwt5Km0xlL92u3d3f76QP1osgeCmEdkqZTGcm8_oIeXChyZwCzpSLQoYLzrRKDbMilwIdFJvZulH4NR6P9XRqJm3ArWrLKjudWCtqN7c-Rj5Aw43GSXNpHssF86hRPrvaQmhsk55EV8aXdMXTzbRwJJBNd5FkhuuwG08qzcAFtuR-8Mrn765lbWJGh__9uCNy0DqX9KmRhmOyBcUJ2Rt2mG6npHqiTSijYrPC59jBUT_REomKph6cLucUxXENtPRChTc6CJV6AYtmNHhFGf2R-6azon5u5WNv-GqHAuNo04dyRj5Gz-_DF9bCLjArY7lkuVaZgFwAxGAEntGkS3OTRhBkPLM2T5VVSvuWWuvQGTI20tZwCXEQpMB1Ks7JTjEv4JJQIQEppEU3EkJlrHF4fnHcgcdcR13SJ_cdPxPkhM9VpAXMV1Wy4WifXDSbkpTN_I1Eot4IdKSu_kB9TfaFb1jwXYT8hvRy_Knhluza9XJWfd3V8oLX8eTtGxVtzh8
linkProvider ProQuest
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+physics-inspired+neural+network+to+solve+partial+differential+equations+-+application+in+diffusion-induced+stress&rft.jtitle=Physical+chemistry+chemical+physics+%3A+PCCP&rft.au=Xue%2C+Yuan&rft.au=Li%2C+Yong&rft.au=Zhang%2C+Kai&rft.au=Yang%2C+Fuqian&rft.date=2022-03-30&rft.issn=1463-9084&rft.eissn=1463-9084&rft.volume=24&rft.issue=13&rft.spage=7937&rft_id=info:doi/10.1039%2Fd1cp04893g&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1463-9084&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1463-9084&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1463-9084&client=summon