Gradient-enhanced PINN with residual unit for studying forward-inverse problems of variable coefficient equations

Physics-informed neural network (PINN) is a powerful emerging method for studying forward-inverse problems of partial differential equations (PDEs), even from limited sample data. Variable coefficient PDEs, which model real-world phenomena, are of considerable physical significance and research valu...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Physica. D Ročník 481; s. 134764
Hlavní autoři: Zhou, Hui-Juan, Chen, Yong
Médium: Journal Article
Jazyk:angličtina
Vydáno: Elsevier B.V 01.11.2025
Témata:
ISSN:0167-2789
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Abstract Physics-informed neural network (PINN) is a powerful emerging method for studying forward-inverse problems of partial differential equations (PDEs), even from limited sample data. Variable coefficient PDEs, which model real-world phenomena, are of considerable physical significance and research value. This study proposes a gradient-enhanced PINN with residual unit (R-gPINN) method to solve the data-driven solution and function discovery for variable coefficient PDEs. On the one hand, the proposed method incorporates residual units into the neural networks to mitigate gradient vanishing and network degradation, unify linear and nonlinear coefficient problem. We present two types of residual unit structures in this work to offer more flexible solutions in problem-solving. On the other hand, by including gradient terms of variable coefficients, the method penalizes collocation points that fail to satisfy physical properties. This enhancement improves the network’s adherence to physical constraints and aligns the prediction function more closely with the objective function. Numerical experiments including solve the forward-inverse problems of variable coefficient Burgers equation, variable coefficient KdV equation, variable coefficient Sine–Gordon equation, and high-dimensional variable coefficient Kadomtsev–Petviashvili equation. The results show that using R-gPINN method can greatly improve the accuracy of predict solution and predict variable coefficient in solving variable coefficient equations.
AbstractList Physics-informed neural network (PINN) is a powerful emerging method for studying forward-inverse problems of partial differential equations (PDEs), even from limited sample data. Variable coefficient PDEs, which model real-world phenomena, are of considerable physical significance and research value. This study proposes a gradient-enhanced PINN with residual unit (R-gPINN) method to solve the data-driven solution and function discovery for variable coefficient PDEs. On the one hand, the proposed method incorporates residual units into the neural networks to mitigate gradient vanishing and network degradation, unify linear and nonlinear coefficient problem. We present two types of residual unit structures in this work to offer more flexible solutions in problem-solving. On the other hand, by including gradient terms of variable coefficients, the method penalizes collocation points that fail to satisfy physical properties. This enhancement improves the network’s adherence to physical constraints and aligns the prediction function more closely with the objective function. Numerical experiments including solve the forward-inverse problems of variable coefficient Burgers equation, variable coefficient KdV equation, variable coefficient Sine–Gordon equation, and high-dimensional variable coefficient Kadomtsev–Petviashvili equation. The results show that using R-gPINN method can greatly improve the accuracy of predict solution and predict variable coefficient in solving variable coefficient equations.
ArticleNumber 134764
Author Zhou, Hui-Juan
Chen, Yong
Author_xml – sequence: 1
  givenname: Hui-Juan
  surname: Zhou
  fullname: Zhou, Hui-Juan
  organization: School of Science, Shanghai Maritime University, Shanghai, 201306, PR China
– sequence: 2
  givenname: Yong
  orcidid: 0000-0002-6008-6542
  surname: Chen
  fullname: Chen, Yong
  email: ychen@sei.ecnu.edu.cn
  organization: School of Mathematical Sciences, Key Laboratory of MEA (Ministry of Education) and Shanghai Key Laboratory of PMMP, East China Normal University, Shanghai, 200241, China
BookMark eNp9kEFPAjEUhHvAREB_gZf-gV3b7m539-DBEEUSgh703JT2VUqghbYL4d-7K549vUzezGTyTdDIeQcIPVCSU0L54zY_bC5R54ywKqdFWfNyhMb9p85Y3bS3aBLjlhBC66Ieo-M8SG3BpQzcRjoFGn8sVit8tmmDA0SrO7nDnbMJGx9wTJ2-WPc9iLMMOrPuBCECPgS_3sE-Ym_wSQYre4WVB2OsGuoxHDuZrHfxDt0YuYtw_3en6Ov15XP2li3f54vZ8zJTrKpSxriRirJCl6wFxrkma6pJq5lhQBtWKWLKVhrerGmraFnWtJWEN5z1Xq6Kupii4tqrgo8xgBGHYPcyXAQlYiAltuKXlBhIiSupPvV0TUE_7WQhiDjs77nYACoJ7e2_-R-6iHio
Cites_doi 10.1016/j.physleta.2021.127739
10.1016/j.physleta.2022.128536
10.1016/j.physd.2023.133729
10.1143/JPSJ.30.272
10.1103/PhysRevE.108.045303
10.1002/sapm1987762133
10.1016/0021-9991(83)90085-2
10.1016/j.chaos.2021.111393
10.1016/j.physd.2022.133629
10.1016/0893-6080(89)90020-8
10.1016/j.cma.2022.114823
10.1016/j.ijleo.2019.01.018
10.1007/s11071-023-08641-1
10.1002/mma.5982
10.1017/S0022112078001214
10.1016/j.physd.2023.133851
10.1016/j.jcp.2024.113090
10.1016/j.jcp.2018.10.045
10.1002/sapm19898011
10.1016/j.physd.2023.133945
10.1140/epjd/e2011-20518-0
10.1016/S0375-9601(02)00033-6
10.1103/PhysRevLett.40.233
10.1016/j.physleta.2021.127408
10.1007/s11071-023-08712-3
10.1007/s11424-024-3467-7
10.1016/S0022-247X(02)00445-6
10.1016/0045-7825(92)90088-2
10.1063/1.1693097
10.1038/s42254-021-00314-5
10.1090/qam/42889
10.1016/0375-9601(75)90124-3
ContentType Journal Article
Copyright 2025
Copyright_xml – notice: 2025
DBID AAYXX
CITATION
DOI 10.1016/j.physd.2025.134764
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Physics
ExternalDocumentID 10_1016_j_physd_2025_134764
S0167278925002416
GroupedDBID --K
--M
-~X
.~1
0R~
1B1
1RT
1~.
1~5
29O
4.4
457
4G.
5VS
7-5
71M
8P~
9JN
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AAQXK
AATTM
AAXKI
AAXUO
AAYWO
ABAOU
ABFNM
ABMAC
ABNEU
ABWVN
ABXDB
ACDAQ
ACFVG
ACGFS
ACLOT
ACNCT
ACNNM
ACRLP
ACRPL
ADBBV
ADEZE
ADGUI
ADIYS
ADMUD
ADNMO
ADVLN
AEBSH
AEIPS
AEKER
AFFNX
AFJKZ
AFTJW
AGHFR
AGQPQ
AGUBO
AGYEJ
AHHHB
AIEXJ
AIGVJ
AIIUN
AIKHN
AITUG
AIVDX
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
APXCP
ARUGR
ASPBG
AVWKF
AXJTR
AZFZN
BBWZM
BKOJK
BLXMC
EBS
EFJIC
EFKBS
EFLBG
EJD
EO8
EO9
EP2
EP3
F5P
FDB
FEDTE
FGOYB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
HMV
HVGLF
HZ~
H~9
IHE
J1W
K-O
KOM
M38
M41
MHUIS
MO0
MVM
N9A
NDZJH
O-L
O9-
OAUVE
OGIMB
OZT
P-8
P-9
P2P
PC.
Q38
R2-
RNS
ROL
RPZ
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SPD
SPG
SSQ
SSW
SSZ
T5K
TN5
TWZ
WUQ
XJT
XPP
YNT
YYP
~02
~G-
~HD
9DU
AAYXX
CITATION
ID FETCH-LOGICAL-c255t-26fac123d429e266d0b1d09d2f2e1825c0f49af68b19c144719a068622666c373
ISICitedReferencesCount 1
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001516751600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0167-2789
IngestDate Sat Nov 29 07:37:03 EST 2025
Sat Sep 27 17:12:47 EDT 2025
IsPeerReviewed true
IsScholarly true
Keywords Physics-informed neural network
Function discovery
Variable coefficient equation
Residual network
Data-driven solution
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c255t-26fac123d429e266d0b1d09d2f2e1825c0f49af68b19c144719a068622666c373
ORCID 0000-0002-6008-6542
ParticipantIDs crossref_primary_10_1016_j_physd_2025_134764
elsevier_sciencedirect_doi_10_1016_j_physd_2025_134764
PublicationCentury 2000
PublicationDate November 2025
2025-11-00
PublicationDateYYYYMMDD 2025-11-01
PublicationDate_xml – month: 11
  year: 2025
  text: November 2025
PublicationDecade 2020
PublicationTitle Physica. D
PublicationYear 2025
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Ott, Sudan (b9) 1970; 13
Zhou (b26) 2024; 37
Güngör, Winternitz (b46) 2002; 276
Song, Yan (b29) 2023; 448
Moghimi, Hejazi (b35) 2007; 33
Hua, Shuklab, Karniadakis, Kawaguchia (b12) 2024; 176
Lin, Chen (b28) 2023; 459
Batchelor (b2) 1967
David, Levi, Winternitz (b44) 1987; 76
Grimshaw (b39) 1979; 368
Wang, Cui (b22) 2022; 456
Wazwaz (b42) 2019; 182
Parker (b33) 1992; 438
Ko, Kuehl (b8) 1978; 40
Wu, Fang, Wang, Wu, Dai (b21) 2021; 152
Zhou, Pu, Chen (b25) 2023; 111
Tian, Niu, Li (b20) 2023; 111
Mo, Ling, Zeng (b23) 2022; 421
Shuning, Yong (b18) 2022; 457
Liu, Wang, Zhang, Lu, Liu (b24) 2023; 108
Fletcher (b36) 1983; 51
Pu, Chen (b16) 2024; 510
Grimshaw (b6) 1978; 86
Cohen-Tannoudji, Diu, Laloë (b3) 2019
Raissi, Perdikaris, Karniadakis (b10) 2019; 378
Dia, Chau, Lu, Zheng (b34) 2020; 43
David, Levi, Winternitz (b45) 1989; 80
Karniadakis, Kevrekidis, Lu (b13) 2021; 3
Yu, Lu, Meng, Karniadakis (b31) 2022; 393
Kadomtsev, Petviashvili (b43) 1970; Vol. 192
Keener, Sneyd (b4) 2009
Braun, Kivshar (b41) 2004
Lin, Chen (b17) 2023; 445
Fan (b40) 2002; 294
Awawdeh, Jaradat, Al-Shara (b38) 2012; 66
Li, Chen (b14) 2020; 72
Kakutani (b5) 1971; 30
Carslaw (b1) 1906
Wang, Yan (b19) 2021; 404
Hornik, Stinchcombe, White (b11) 1989; 2
Miao, Chen (b27) 2023; 456
He, Zhang, Ren, Sun (b30) 2016
Nishikawa, Kaw (b7) 1975; 50
Pu, Chen (b15) 2023; 454
Cole (b32) 1951; 9
Ali, Gardner, Gardner (b37) 1992; 100
Zhou (10.1016/j.physd.2025.134764_b26) 2024; 37
Raissi (10.1016/j.physd.2025.134764_b10) 2019; 378
Moghimi (10.1016/j.physd.2025.134764_b35) 2007; 33
Kakutani (10.1016/j.physd.2025.134764_b5) 1971; 30
Mo (10.1016/j.physd.2025.134764_b23) 2022; 421
Karniadakis (10.1016/j.physd.2025.134764_b13) 2021; 3
Hornik (10.1016/j.physd.2025.134764_b11) 1989; 2
Liu (10.1016/j.physd.2025.134764_b24) 2023; 108
Wang (10.1016/j.physd.2025.134764_b19) 2021; 404
Wang (10.1016/j.physd.2025.134764_b22) 2022; 456
Ott (10.1016/j.physd.2025.134764_b9) 1970; 13
Grimshaw (10.1016/j.physd.2025.134764_b6) 1978; 86
Parker (10.1016/j.physd.2025.134764_b33) 1992; 438
David (10.1016/j.physd.2025.134764_b44) 1987; 76
Ko (10.1016/j.physd.2025.134764_b8) 1978; 40
Tian (10.1016/j.physd.2025.134764_b20) 2023; 111
Wazwaz (10.1016/j.physd.2025.134764_b42) 2019; 182
Cohen-Tannoudji (10.1016/j.physd.2025.134764_b3) 2019
He (10.1016/j.physd.2025.134764_b30) 2016
Cole (10.1016/j.physd.2025.134764_b32) 1951; 9
Pu (10.1016/j.physd.2025.134764_b16) 2024; 510
Miao (10.1016/j.physd.2025.134764_b27) 2023; 456
Awawdeh (10.1016/j.physd.2025.134764_b38) 2012; 66
Zhou (10.1016/j.physd.2025.134764_b25) 2023; 111
Ali (10.1016/j.physd.2025.134764_b37) 1992; 100
Batchelor (10.1016/j.physd.2025.134764_b2) 1967
Wu (10.1016/j.physd.2025.134764_b21) 2021; 152
Shuning (10.1016/j.physd.2025.134764_b18) 2022; 457
Keener (10.1016/j.physd.2025.134764_b4) 2009
Kadomtsev (10.1016/j.physd.2025.134764_b43) 1970; Vol. 192
Pu (10.1016/j.physd.2025.134764_b15) 2023; 454
Hua (10.1016/j.physd.2025.134764_b12) 2024; 176
Yu (10.1016/j.physd.2025.134764_b31) 2022; 393
Fletcher (10.1016/j.physd.2025.134764_b36) 1983; 51
Nishikawa (10.1016/j.physd.2025.134764_b7) 1975; 50
Song (10.1016/j.physd.2025.134764_b29) 2023; 448
Lin (10.1016/j.physd.2025.134764_b17) 2023; 445
David (10.1016/j.physd.2025.134764_b45) 1989; 80
Güngör (10.1016/j.physd.2025.134764_b46) 2002; 276
Li (10.1016/j.physd.2025.134764_b14) 2020; 72
Grimshaw (10.1016/j.physd.2025.134764_b39) 1979; 368
Braun (10.1016/j.physd.2025.134764_b41) 2004
Fan (10.1016/j.physd.2025.134764_b40) 2002; 294
Lin (10.1016/j.physd.2025.134764_b28) 2023; 459
Carslaw (10.1016/j.physd.2025.134764_b1) 1906
Dia (10.1016/j.physd.2025.134764_b34) 2020; 43
References_xml – volume: 111
  start-page: 16467
  year: 2023
  end-page: 16482
  ident: b20
  article-title: Mix-training physics-informed neural networks for high-order rogue waves of cmKdV equation
  publication-title: Nonlinear Dynam.
– volume: 510
  year: 2024
  ident: b16
  article-title: Lax pairs informed neural networks solving integrable systems
  publication-title: J. Comput. Phys.
– volume: 2
  start-page: 359
  year: 1989
  end-page: 366
  ident: b11
  article-title: Multilayer feedforward networks are universal approximators
  publication-title: Neural Netw.
– volume: 456
  year: 2023
  ident: b27
  article-title: VC-PINN: Variable coefficient physics-informed neural network for forward and inverse problems of PDEs with variable coefficient
  publication-title: Phys. D
– volume: 182
  start-page: 605
  year: 2019
  end-page: 610
  ident: b42
  article-title: The integrable time-dependent sine-Gordon equation with multiple optical kink solutions
  publication-title: Optik
– volume: 9
  start-page: 225
  year: 1951
  end-page: 236
  ident: b32
  article-title: On a quasi-linear parabolic equation occurring in aerodynamics
  publication-title: Quart. Appl. Math.
– volume: 40
  start-page: 233
  year: 1978
  end-page: 236
  ident: b8
  article-title: Korteweg–de Vries soliton in a slowly varying medium
  publication-title: Phys. Rev. Lett.
– year: 1906
  ident: b1
  article-title: Introduction to the Mathematical Theory of the Conduction of Heat in Solids
– year: 1967
  ident: b2
  article-title: An Introduction to Fluid Dynamics
– volume: 176
  year: 2024
  ident: b12
  article-title: Tackling the curse of dimensionality with physics-informed neural networks
  publication-title: Neural Netw.
– volume: 448
  year: 2023
  ident: b29
  article-title: Deep learning soliton dynamics and complex potentials recognition for 1D and 2D PT-symmetric saturable nonlinear Schrödinger equations
  publication-title: Phys. D
– volume: 66
  start-page: 40
  year: 2012
  ident: b38
  article-title: Applications of a simplified bilinear method to ion-acoustic solitary waves in plasma
  publication-title: Eur. Phys. J. D
– volume: 445
  year: 2023
  ident: b17
  article-title: Physics-informed neural network methods based on Miura transformations and discovery of new localized wave solutions
  publication-title: Phys. D
– volume: 76
  start-page: 133
  year: 1987
  end-page: 168
  ident: b44
  article-title: Integrable nonlinear equations for water waves in straits of varying depth and width
  publication-title: Stud. Appl. Math.
– volume: 51
  start-page: 159
  year: 1983
  end-page: 188
  ident: b36
  article-title: A comparison of finite element and finite difference solutions of the one-and two-dimensional Burgers equations
  publication-title: J. Comput. Phys.
– volume: 454
  year: 2023
  ident: b15
  article-title: Complex dynamics on the one-dimensional quantum droplets via time piecewise PINNs
  publication-title: Phys. D
– volume: 100
  start-page: 325
  year: 1992
  end-page: 337
  ident: b37
  article-title: A collocation solution for Burgers’ equation using cubic B-spline finite elements
  publication-title: Comput. Methods Appl. Mech. Engrg.
– volume: 276
  start-page: 314
  year: 2002
  end-page: 328
  ident: b46
  article-title: Generalized Kadomtsev–Petviashvili equation with an infinite-dimensional symmetry algebra
  publication-title: J. Math. Anal. Appl.
– volume: 33
  start-page: 1756
  year: 2007
  end-page: 1761
  ident: b35
  article-title: Variational iteration method for solving generalized Burger–Fisher and Burger equations
  publication-title: Chaos
– year: 2004
  ident: b41
  article-title: The Frenkel-Kontorova Model: Concepts, Methods, and Applications
– volume: 457
  year: 2022
  ident: b18
  article-title: A two-stage physics-informed neural network method based on conserved quantities and applications in localized wave solutions
  publication-title: J. Comput. Phys.
– volume: 86
  start-page: 415
  year: 1978
  end-page: 431
  ident: b6
  article-title: Long nonlinear internal waves in channels of arbitrary cross-section
  publication-title: J. Fluid Mech.
– start-page: 630
  year: 2016
  end-page: 645
  ident: b30
  article-title: Identity Mappings in Deep Residual Networks
– volume: 3
  start-page: 422
  year: 2021
  end-page: 440
  ident: b13
  article-title: Physics-informed machine learning
  publication-title: Nat. Rev. Phys.
– volume: 50
  start-page: 455
  year: 1975
  end-page: 456
  ident: b7
  article-title: Propagation of solitary ion acoustic waves in inhomogeneous plasmas
  publication-title: Phys. Lett. A
– volume: 459
  year: 2023
  ident: b28
  article-title: Gradient-enhanced physics-informed neural networks based on transfer learning for inverse problems of the variable coefficient differential equations
  publication-title: Phys. D
– volume: 72
  year: 2020
  ident: b14
  article-title: Solving second-order nonlinear evolution partial differential equations using deep learning
  publication-title: Commun. Theor. Phys. (Beijing)
– volume: 368
  start-page: 359
  year: 1979
  end-page: 375
  ident: b39
  article-title: Slowly varying solitary waves. I. Korteweg–de Vries equation
  publication-title: Proc. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci.
– volume: Vol. 192
  start-page: 753
  year: 1970
  end-page: 756
  ident: b43
  article-title: On the stability of solitary waves in weakly dispersing media
  publication-title: Doklady Akademii Nauk
– volume: 80
  start-page: 1
  year: 1989
  end-page: 23
  ident: b45
  article-title: Solitons in shallow seas of variable depth and in marine straits
  publication-title: Stud. Appl. Math.
– volume: 421
  year: 2022
  ident: b23
  article-title: Data-driven vector soliton solutions of coupled nonlinear Schrödinger equation using a deep learning algorithm
  publication-title: Phys. Lett. A
– volume: 43
  start-page: 2171
  year: 2020
  end-page: 2188
  ident: b34
  article-title: Mathematical studies of the solution of Burgers’ equations by adomian decomposition method
  publication-title: Math. Methods Appl. Sci.
– volume: 30
  start-page: 272
  year: 1971
  end-page: 276
  ident: b5
  article-title: Effect of an uneven bottom on gravity waves
  publication-title: J. Phys. Soc. Japan
– volume: 438
  start-page: 93
  year: 1992
  end-page: 108
  ident: b33
  article-title: On periodic solutions of the burgers equation: a unified approach
  publication-title: Proc. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci.
– year: 2019
  ident: b3
  article-title: Quantum Mechanics: Concepts and Applications
– volume: 152
  year: 2021
  ident: b21
  article-title: Predicting the dynamic process and model parameters of the vector optical solitons in birefringent fibers via the modified PINN
  publication-title: Chaos Solitons Fractals
– year: 2009
  ident: b4
  article-title: Mathematical Physiology 1: Cellular Physiology
– volume: 378
  start-page: 686
  year: 2019
  end-page: 707
  ident: b10
  article-title: Physics-informed neural net- works: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
  publication-title: J. Comput. Phys.
– volume: 456
  year: 2022
  ident: b22
  article-title: Prediction of the number of solitons for initial value of nonlinear Schrödinger equation based on the deep learning method
  publication-title: Phys. Lett. A
– volume: 13
  start-page: 1432
  year: 1970
  end-page: 1434
  ident: b9
  article-title: Damping of solitary waves
  publication-title: Phys. Fluids
– volume: 404
  year: 2021
  ident: b19
  article-title: Data-driven rogue waves and parameter discovery in the defocusing nonlinear Schrödinger equation with a potential using the PINN deeplearning
  publication-title: Phys. Lett. A
– volume: 393
  year: 2022
  ident: b31
  article-title: Gradient-enhanced physics-informed neural networks for forward and inverse PDE problems
  publication-title: Comput. Methods Appl. Mech. Engrg.
– volume: 294
  start-page: 26
  year: 2002
  end-page: 30
  ident: b40
  article-title: Auto-Bäcklund transformation and similarity reductions for general variable coefficient KdV equations
  publication-title: Phys. Lett. A
– volume: 111
  start-page: 14667
  year: 2023
  end-page: 14693
  ident: b25
  article-title: Data-driven forward–inverse problems for the variable coefficients Hirota equation using deep learning method
  publication-title: Nonlinear Dynam.
– volume: 37
  start-page: 511
  year: 2024
  end-page: 544
  ident: b26
  article-title: Parallel physics-informed neural networks method with regularization strategies for the forward-inverse problems of the variable coefficient modified KdV equation
  publication-title: J. Syst. Sci. Complex.
– volume: 108
  year: 2023
  ident: b24
  article-title: Prediction of phase transition and time-varying dynamics of the (2+1)-dimensional Boussinesq equation by the parameter-integrated physics-informed neural networks with phase domain decomposition
  publication-title: Phys. Rev. E
– volume: 421
  year: 2022
  ident: 10.1016/j.physd.2025.134764_b23
  article-title: Data-driven vector soliton solutions of coupled nonlinear Schrödinger equation using a deep learning algorithm
  publication-title: Phys. Lett. A
  doi: 10.1016/j.physleta.2021.127739
– volume: 456
  year: 2022
  ident: 10.1016/j.physd.2025.134764_b22
  article-title: Prediction of the number of solitons for initial value of nonlinear Schrödinger equation based on the deep learning method
  publication-title: Phys. Lett. A
  doi: 10.1016/j.physleta.2022.128536
– volume: 448
  year: 2023
  ident: 10.1016/j.physd.2025.134764_b29
  article-title: Deep learning soliton dynamics and complex potentials recognition for 1D and 2D PT-symmetric saturable nonlinear Schrödinger equations
  publication-title: Phys. D
  doi: 10.1016/j.physd.2023.133729
– volume: 30
  start-page: 272
  year: 1971
  ident: 10.1016/j.physd.2025.134764_b5
  article-title: Effect of an uneven bottom on gravity waves
  publication-title: J. Phys. Soc. Japan
  doi: 10.1143/JPSJ.30.272
– volume: 108
  year: 2023
  ident: 10.1016/j.physd.2025.134764_b24
  article-title: Prediction of phase transition and time-varying dynamics of the (2+1)-dimensional Boussinesq equation by the parameter-integrated physics-informed neural networks with phase domain decomposition
  publication-title: Phys. Rev. E
  doi: 10.1103/PhysRevE.108.045303
– year: 2004
  ident: 10.1016/j.physd.2025.134764_b41
– volume: 76
  start-page: 133
  issue: 2
  year: 1987
  ident: 10.1016/j.physd.2025.134764_b44
  article-title: Integrable nonlinear equations for water waves in straits of varying depth and width
  publication-title: Stud. Appl. Math.
  doi: 10.1002/sapm1987762133
– volume: 459
  year: 2023
  ident: 10.1016/j.physd.2025.134764_b28
  article-title: Gradient-enhanced physics-informed neural networks based on transfer learning for inverse problems of the variable coefficient differential equations
  publication-title: Phys. D
– volume: 176
  year: 2024
  ident: 10.1016/j.physd.2025.134764_b12
  article-title: Tackling the curse of dimensionality with physics-informed neural networks
  publication-title: Neural Netw.
– volume: 51
  start-page: 159
  issue: 1
  year: 1983
  ident: 10.1016/j.physd.2025.134764_b36
  article-title: A comparison of finite element and finite difference solutions of the one-and two-dimensional Burgers equations
  publication-title: J. Comput. Phys.
  doi: 10.1016/0021-9991(83)90085-2
– volume: 152
  year: 2021
  ident: 10.1016/j.physd.2025.134764_b21
  article-title: Predicting the dynamic process and model parameters of the vector optical solitons in birefringent fibers via the modified PINN
  publication-title: Chaos Solitons Fractals
  doi: 10.1016/j.chaos.2021.111393
– volume: 445
  year: 2023
  ident: 10.1016/j.physd.2025.134764_b17
  article-title: Physics-informed neural network methods based on Miura transformations and discovery of new localized wave solutions
  publication-title: Phys. D
  doi: 10.1016/j.physd.2022.133629
– volume: 2
  start-page: 359
  year: 1989
  ident: 10.1016/j.physd.2025.134764_b11
  article-title: Multilayer feedforward networks are universal approximators
  publication-title: Neural Netw.
  doi: 10.1016/0893-6080(89)90020-8
– volume: 457
  year: 2022
  ident: 10.1016/j.physd.2025.134764_b18
  article-title: A two-stage physics-informed neural network method based on conserved quantities and applications in localized wave solutions
  publication-title: J. Comput. Phys.
– volume: 393
  year: 2022
  ident: 10.1016/j.physd.2025.134764_b31
  article-title: Gradient-enhanced physics-informed neural networks for forward and inverse PDE problems
  publication-title: Comput. Methods Appl. Mech. Engrg.
  doi: 10.1016/j.cma.2022.114823
– volume: 182
  start-page: 605
  year: 2019
  ident: 10.1016/j.physd.2025.134764_b42
  article-title: The integrable time-dependent sine-Gordon equation with multiple optical kink solutions
  publication-title: Optik
  doi: 10.1016/j.ijleo.2019.01.018
– volume: 33
  start-page: 1756
  issue: 5
  year: 2007
  ident: 10.1016/j.physd.2025.134764_b35
  article-title: Variational iteration method for solving generalized Burger–Fisher and Burger equations
  publication-title: Chaos
– volume: 111
  start-page: 14667
  year: 2023
  ident: 10.1016/j.physd.2025.134764_b25
  article-title: Data-driven forward–inverse problems for the variable coefficients Hirota equation using deep learning method
  publication-title: Nonlinear Dynam.
  doi: 10.1007/s11071-023-08641-1
– volume: 43
  start-page: 2171
  issue: 5
  year: 2020
  ident: 10.1016/j.physd.2025.134764_b34
  article-title: Mathematical studies of the solution of Burgers’ equations by adomian decomposition method
  publication-title: Math. Methods Appl. Sci.
  doi: 10.1002/mma.5982
– volume: 86
  start-page: 415
  year: 1978
  ident: 10.1016/j.physd.2025.134764_b6
  article-title: Long nonlinear internal waves in channels of arbitrary cross-section
  publication-title: J. Fluid Mech.
  doi: 10.1017/S0022112078001214
– volume: 454
  year: 2023
  ident: 10.1016/j.physd.2025.134764_b15
  article-title: Complex dynamics on the one-dimensional quantum droplets via time piecewise PINNs
  publication-title: Phys. D
  doi: 10.1016/j.physd.2023.133851
– volume: 510
  year: 2024
  ident: 10.1016/j.physd.2025.134764_b16
  article-title: Lax pairs informed neural networks solving integrable systems
  publication-title: J. Comput. Phys.
  doi: 10.1016/j.jcp.2024.113090
– volume: 438
  start-page: 93
  year: 1992
  ident: 10.1016/j.physd.2025.134764_b33
  article-title: On periodic solutions of the burgers equation: a unified approach
  publication-title: Proc. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci.
– volume: 378
  start-page: 686
  year: 2019
  ident: 10.1016/j.physd.2025.134764_b10
  article-title: Physics-informed neural net- works: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
  publication-title: J. Comput. Phys.
  doi: 10.1016/j.jcp.2018.10.045
– volume: 80
  start-page: 1
  issue: 1
  year: 1989
  ident: 10.1016/j.physd.2025.134764_b45
  article-title: Solitons in shallow seas of variable depth and in marine straits
  publication-title: Stud. Appl. Math.
  doi: 10.1002/sapm19898011
– volume: 456
  year: 2023
  ident: 10.1016/j.physd.2025.134764_b27
  article-title: VC-PINN: Variable coefficient physics-informed neural network for forward and inverse problems of PDEs with variable coefficient
  publication-title: Phys. D
  doi: 10.1016/j.physd.2023.133945
– start-page: 630
  year: 2016
  ident: 10.1016/j.physd.2025.134764_b30
– volume: 72
  year: 2020
  ident: 10.1016/j.physd.2025.134764_b14
  article-title: Solving second-order nonlinear evolution partial differential equations using deep learning
  publication-title: Commun. Theor. Phys. (Beijing)
– volume: 66
  start-page: 40
  year: 2012
  ident: 10.1016/j.physd.2025.134764_b38
  article-title: Applications of a simplified bilinear method to ion-acoustic solitary waves in plasma
  publication-title: Eur. Phys. J. D
  doi: 10.1140/epjd/e2011-20518-0
– volume: 294
  start-page: 26
  issue: 1
  year: 2002
  ident: 10.1016/j.physd.2025.134764_b40
  article-title: Auto-Bäcklund transformation and similarity reductions for general variable coefficient KdV equations
  publication-title: Phys. Lett. A
  doi: 10.1016/S0375-9601(02)00033-6
– volume: Vol. 192
  start-page: 753
  year: 1970
  ident: 10.1016/j.physd.2025.134764_b43
  article-title: On the stability of solitary waves in weakly dispersing media
– year: 1967
  ident: 10.1016/j.physd.2025.134764_b2
– volume: 40
  start-page: 233
  year: 1978
  ident: 10.1016/j.physd.2025.134764_b8
  article-title: Korteweg–de Vries soliton in a slowly varying medium
  publication-title: Phys. Rev. Lett.
  doi: 10.1103/PhysRevLett.40.233
– volume: 404
  year: 2021
  ident: 10.1016/j.physd.2025.134764_b19
  article-title: Data-driven rogue waves and parameter discovery in the defocusing nonlinear Schrödinger equation with a potential using the PINN deeplearning
  publication-title: Phys. Lett. A
  doi: 10.1016/j.physleta.2021.127408
– volume: 111
  start-page: 16467
  year: 2023
  ident: 10.1016/j.physd.2025.134764_b20
  article-title: Mix-training physics-informed neural networks for high-order rogue waves of cmKdV equation
  publication-title: Nonlinear Dynam.
  doi: 10.1007/s11071-023-08712-3
– volume: 37
  start-page: 511
  year: 2024
  ident: 10.1016/j.physd.2025.134764_b26
  article-title: Parallel physics-informed neural networks method with regularization strategies for the forward-inverse problems of the variable coefficient modified KdV equation
  publication-title: J. Syst. Sci. Complex.
  doi: 10.1007/s11424-024-3467-7
– volume: 368
  start-page: 359
  year: 1979
  ident: 10.1016/j.physd.2025.134764_b39
  article-title: Slowly varying solitary waves. I. Korteweg–de Vries equation
  publication-title: Proc. R. Soc. Lond. Ser. A Math. Phys. Eng. Sci.
– volume: 276
  start-page: 314
  issue: 1
  year: 2002
  ident: 10.1016/j.physd.2025.134764_b46
  article-title: Generalized Kadomtsev–Petviashvili equation with an infinite-dimensional symmetry algebra
  publication-title: J. Math. Anal. Appl.
  doi: 10.1016/S0022-247X(02)00445-6
– year: 2019
  ident: 10.1016/j.physd.2025.134764_b3
– volume: 100
  start-page: 325
  issue: 3
  year: 1992
  ident: 10.1016/j.physd.2025.134764_b37
  article-title: A collocation solution for Burgers’ equation using cubic B-spline finite elements
  publication-title: Comput. Methods Appl. Mech. Engrg.
  doi: 10.1016/0045-7825(92)90088-2
– volume: 13
  start-page: 1432
  year: 1970
  ident: 10.1016/j.physd.2025.134764_b9
  article-title: Damping of solitary waves
  publication-title: Phys. Fluids
  doi: 10.1063/1.1693097
– volume: 3
  start-page: 422
  year: 2021
  ident: 10.1016/j.physd.2025.134764_b13
  article-title: Physics-informed machine learning
  publication-title: Nat. Rev. Phys.
  doi: 10.1038/s42254-021-00314-5
– volume: 9
  start-page: 225
  issue: 3
  year: 1951
  ident: 10.1016/j.physd.2025.134764_b32
  article-title: On a quasi-linear parabolic equation occurring in aerodynamics
  publication-title: Quart. Appl. Math.
  doi: 10.1090/qam/42889
– year: 1906
  ident: 10.1016/j.physd.2025.134764_b1
– year: 2009
  ident: 10.1016/j.physd.2025.134764_b4
– volume: 50
  start-page: 455
  year: 1975
  ident: 10.1016/j.physd.2025.134764_b7
  article-title: Propagation of solitary ion acoustic waves in inhomogeneous plasmas
  publication-title: Phys. Lett. A
  doi: 10.1016/0375-9601(75)90124-3
SSID ssj0001737
Score 2.4729505
Snippet Physics-informed neural network (PINN) is a powerful emerging method for studying forward-inverse problems of partial differential equations (PDEs), even from...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 134764
SubjectTerms Data-driven solution
Function discovery
Physics-informed neural network
Residual network
Variable coefficient equation
Title Gradient-enhanced PINN with residual unit for studying forward-inverse problems of variable coefficient equations
URI https://dx.doi.org/10.1016/j.physd.2025.134764
Volume 481
WOSCitedRecordID wos001516751600001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: Elsevier SD Freedom Collection Journals 2021
  issn: 0167-2789
  databaseCode: AIEXJ
  dateStart: 19950101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0001737
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1LT9tAEF7xKBI9IKBF0ALaA8c68mu93iNClIdQxAFQ2otl70MkBxuSGPHzmX0lhlSoHLhYlmWvLX-fZ2fG38widFSSMleUAQJCJkHKkyhgSVgGLK1EVoJ1pOZH-90V7ffzwYBdu4YKE7OcAK3r_PmZPXwq1HAMwNalsx-AezYoHIB9AB22ADts_wv4s7FRcU0DWd_b3_sQufdtwhVia1t81cKXbBSGpr2sU1MaAe2w1kINXT5lVpoxSo8nCKhNiRVvpGk5oQUE8rHtZPucf3ttYe_NhcR_75vWzG_tMLhs51w8cWUhfxo3d7rUQ0xcDV4nGwlWVpfSds1papdgcQZRV6raNuULttqmDUY9ncLRPVtj0puf_boz9psZa6Yj9BK1UWEGKfQghR1kGa3GlDCw1avHF6eDy9n0HFHbSNU_u29FZUR_C8_yb3el44LcbKINFzvgY4v5FlqS9Tb62ukouY3WLACTb-hxgQdY8wBrHmDPA6x5gAF67HmA3_AAex7gRmHPA9zhAZ7x4Du6_X16c3IeuPU1Ag6B5DSIM1Vy8FwE-CQSHDURVpEImYhVLCHsJDxUKStVllcR4xB404iVpqIIzs14QpMdtFI3tdxFWMRSVJXgQoksJTmv4lLRSIVESZLxnOyhX_41Fg-2jUrxDnh7KPOvunCeoPXwCiDPexf--Nh9fqL1Oa_30cp03MoD9IU_TYeT8aFjzgsWA4MM
linkProvider Elsevier
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=Gradient-enhanced+PINN+with+residual+unit+for+studying+forward-inverse+problems+of+variable+coefficient+equations&rft.jtitle=Physica.+D&rft.au=Zhou%2C+Hui-Juan&rft.au=Chen%2C+Yong&rft.date=2025-11-01&rft.issn=0167-2789&rft.volume=481&rft.spage=134764&rft_id=info:doi/10.1016%2Fj.physd.2025.134764&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_physd_2025_134764
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0167-2789&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0167-2789&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0167-2789&client=summon