A discrete element solution method embedded within a Neural Network

This paper introduces a novel methodology, the Neural Network framework for the Discrete Element Method (NN4DEM), as part of a broader initiative to harness specialised AI hardware and software environments, marking a transition from traditional computational physics programming approaches. NN4DEM e...

Full description

Saved in:
Bibliographic Details
Published in:Powder technology Vol. 448; p. 120258
Main Authors: Naderi, Sadjad, Chen, Boyang, Yang, Tongan, Xiang, Jiansheng, Heaney, Claire E., Latham, John-Paul, Wang, Yanghua, Pain, Christopher C.
Format: Journal Article
Language:English
Published: Elsevier B.V 01.12.2024
Subjects:
ISSN:0032-5910
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract This paper introduces a novel methodology, the Neural Network framework for the Discrete Element Method (NN4DEM), as part of a broader initiative to harness specialised AI hardware and software environments, marking a transition from traditional computational physics programming approaches. NN4DEM enables GPU-parallelised computations by mapping particle data (coordinates and velocities) onto uniform grids as solution fields and computing contact forces by applying mathematical operations that can be found in convolutional neural networks (CNN). Essentially, this framework transforms a DEM problem into a series of layered “images” composed of pixels, using stencil operations to compute the DEM physics, which is inherently local. The method revolves around custom kernels, with operations prescribed by the laws of physics for contact detection and interaction. Therefore, unlike conventional AI methods, it eliminates the need for training data to determine network weights. NN4DEM utilises libraries such as PyTorch for relatively easier programmability and platform interoperability. This paper presents the theoretical foundations, implementation and validation of NN4DEM through hopper test benchmarks. An analysis of the results from random packing cases highlights the ability of NN4DEM to scale to 3D models with millions of particles. The paper concludes with potential research directions, including further integration with other physics-based models and applications across various multidisciplinary fields. [Display omitted] •A novel integration: neural networks with DEM principles.•Physical laws dictate network weights.•Neural Network solver benefits without training data.•GPU parallelisation with PyTorch.•Random packing demonstrates computational capacity.
AbstractList This paper introduces a novel methodology, the Neural Network framework for the Discrete Element Method (NN4DEM), as part of a broader initiative to harness specialised AI hardware and software environments, marking a transition from traditional computational physics programming approaches. NN4DEM enables GPU-parallelised computations by mapping particle data (coordinates and velocities) onto uniform grids as solution fields and computing contact forces by applying mathematical operations that can be found in convolutional neural networks (CNN). Essentially, this framework transforms a DEM problem into a series of layered “images” composed of pixels, using stencil operations to compute the DEM physics, which is inherently local. The method revolves around custom kernels, with operations prescribed by the laws of physics for contact detection and interaction. Therefore, unlike conventional AI methods, it eliminates the need for training data to determine network weights. NN4DEM utilises libraries such as PyTorch for relatively easier programmability and platform interoperability. This paper presents the theoretical foundations, implementation and validation of NN4DEM through hopper test benchmarks. An analysis of the results from random packing cases highlights the ability of NN4DEM to scale to 3D models with millions of particles. The paper concludes with potential research directions, including further integration with other physics-based models and applications across various multidisciplinary fields.
This paper introduces a novel methodology, the Neural Network framework for the Discrete Element Method (NN4DEM), as part of a broader initiative to harness specialised AI hardware and software environments, marking a transition from traditional computational physics programming approaches. NN4DEM enables GPU-parallelised computations by mapping particle data (coordinates and velocities) onto uniform grids as solution fields and computing contact forces by applying mathematical operations that can be found in convolutional neural networks (CNN). Essentially, this framework transforms a DEM problem into a series of layered “images” composed of pixels, using stencil operations to compute the DEM physics, which is inherently local. The method revolves around custom kernels, with operations prescribed by the laws of physics for contact detection and interaction. Therefore, unlike conventional AI methods, it eliminates the need for training data to determine network weights. NN4DEM utilises libraries such as PyTorch for relatively easier programmability and platform interoperability. This paper presents the theoretical foundations, implementation and validation of NN4DEM through hopper test benchmarks. An analysis of the results from random packing cases highlights the ability of NN4DEM to scale to 3D models with millions of particles. The paper concludes with potential research directions, including further integration with other physics-based models and applications across various multidisciplinary fields. [Display omitted] •A novel integration: neural networks with DEM principles.•Physical laws dictate network weights.•Neural Network solver benefits without training data.•GPU parallelisation with PyTorch.•Random packing demonstrates computational capacity.
ArticleNumber 120258
Author Pain, Christopher C.
Yang, Tongan
Naderi, Sadjad
Chen, Boyang
Latham, John-Paul
Xiang, Jiansheng
Wang, Yanghua
Heaney, Claire E.
Author_xml – sequence: 1
  givenname: Sadjad
  orcidid: 0000-0003-0721-3749
  surname: Naderi
  fullname: Naderi, Sadjad
  organization: Applied Modelling and Computation Group, Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
– sequence: 2
  givenname: Boyang
  orcidid: 0000-0002-9091-0352
  surname: Chen
  fullname: Chen, Boyang
  email: boyang.chen16@imperial.ac.uk
  organization: Applied Modelling and Computation Group, Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
– sequence: 3
  givenname: Tongan
  surname: Yang
  fullname: Yang, Tongan
  organization: Resource Geophysics Academy, Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
– sequence: 4
  givenname: Jiansheng
  surname: Xiang
  fullname: Xiang, Jiansheng
  organization: Applied Modelling and Computation Group, Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
– sequence: 5
  givenname: Claire E.
  surname: Heaney
  fullname: Heaney, Claire E.
  organization: Applied Modelling and Computation Group, Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
– sequence: 6
  givenname: John-Paul
  surname: Latham
  fullname: Latham, John-Paul
  organization: Applied Modelling and Computation Group, Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
– sequence: 7
  givenname: Yanghua
  surname: Wang
  fullname: Wang, Yanghua
  organization: Resource Geophysics Academy, Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
– sequence: 8
  givenname: Christopher C.
  surname: Pain
  fullname: Pain, Christopher C.
  email: c.pain@imperial.ac.uk
  organization: Applied Modelling and Computation Group, Department of Earth Science and Engineering, Imperial College London, London, SW7 2AZ, UK
BookMark eNqFkD9PwzAQxT0UiRb4BgwZWVLOTuwkDEhVxT-pggVmy7EvqksSF9uh4tuTKkwMsNwb7r2nu9-CzHrXIyGXFJYUqLjeLffuEFEvGbB8ScfJyxmZA2Qs5RWFU7IIYQcAIqMwJ-tVYmzQHiMm2GKHfUyCa4doXZ90GLfOJNjVaAya5GDj1vaJSp5x8KodJR6cfz8nJ41qA1786Bl5u797XT-mm5eHp_Vqk-qs5DHlUHDOoCkrVuvC1JxpLIvMCMNLlTNVUqiUyGoENII2WVVVBROlhlqJGsbNGbmaevfefQwYouzG07FtVY9uCDKjPKciE1CM1pvJqr0LwWMjtY3q-FT0yraSgjzSkjs50ZJHWnKiNYbzX-G9t53yX__FbqcYjgw-LXoZtMVeo7EedZTG2b8LvgGHOom8
CitedBy_id crossref_primary_10_1016_j_enggeo_2025_108033
crossref_primary_10_1007_s40571_025_00986_1
Cites_doi 10.14356/kona.2016023
10.1007/s11831-022-09851-3
10.1016/j.cpc.2022.108292
10.1016/j.cma.2023.116370
10.1088/0004-637X/727/2/120
10.1016/j.parco.2018.03.007
10.1007/s42241-020-0050-0
10.1016/j.ijpharm.2020.119427
10.3390/app11146278
10.1016/j.compstruc.2018.08.011
10.1002/nag.3293
10.1016/j.compchemeng.2017.04.019
10.1016/j.enggeo.2021.106444
10.1016/j.jcp.2017.01.014
10.1016/j.compgeo.2023.105677
10.3390/pr9101813
10.1016/j.powtec.2020.11.006
10.1002/nag.3701
10.1016/j.powtec.2020.09.009
10.1016/S0032-5910(99)00178-3
10.1016/j.powtec.2021.06.033
10.1038/188908a0
10.3390/app12063107
10.1038/nature06981
10.1016/j.compgeo.2024.106395
10.1016/j.ces.2021.116832
10.1016/j.powtec.2017.10.050
10.1016/j.apt.2023.104011
10.1016/j.powtec.2016.11.061
10.1016/0032-5910(93)85010-7
10.1016/j.softx.2016.04.004
10.1002/nme.7321
10.3390/chemengineering4040055
10.1080/00223131.2023.2231969
10.1007/s11044-020-09749-7
10.1016/j.cma.2024.116974
10.1016/j.cpc.2024.109196
10.1016/j.apt.2018.06.028
10.1016/j.powtec.2017.06.059
ContentType Journal Article
Copyright 2024 The Authors
Copyright_xml – notice: 2024 The Authors
DBID 6I.
AAFTH
AAYXX
CITATION
7S9
L.6
DOI 10.1016/j.powtec.2024.120258
DatabaseName ScienceDirect Open Access Titles
Elsevier:ScienceDirect:Open Access
CrossRef
AGRICOLA
AGRICOLA - Academic
DatabaseTitle CrossRef
AGRICOLA
AGRICOLA - Academic
DatabaseTitleList AGRICOLA

DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Physics
ExternalDocumentID 10_1016_j_powtec_2024_120258
S0032591024009021
GroupedDBID ---
--K
--M
-~X
.DC
.~1
0R~
123
1B1
1~.
1~5
4.4
457
4G.
5VS
6I.
7-5
71M
8P~
9JN
AABNK
AABXZ
AACTN
AAEDT
AAEDW
AAEPC
AAFTH
AAHBH
AAIKC
AAIKJ
AAKOC
AALRI
AAMNW
AAOAW
AAQFI
AARLI
AAXKI
AAXUO
ABJNI
ABMAC
ABNUV
ABXRA
ACDAQ
ACGFS
ACIWK
ACRLP
ADBBV
ADEWK
ADEZE
AEBSH
AEKER
AENEX
AEZYN
AFJKZ
AFKWA
AFRAH
AFRZQ
AFTJW
AGHFR
AGUBO
AGYEJ
AHHHB
AHPOS
AIEXJ
AIKHN
AITUG
AJOXV
AKRWK
AKURH
ALMA_UNASSIGNED_HOLDINGS
AMFUW
AMRAJ
AXJTR
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
ENUVR
EO8
EO9
EP2
EP3
FDB
FIRID
FLBIZ
FNPLU
FYGXN
G-Q
GBLVA
IHE
J1W
KOM
LX7
M41
MAGPM
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
ROL
RPZ
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SSG
SSM
SSZ
T5K
~02
~G-
29O
8WZ
9DU
A6W
AAQXK
AATTM
AAYWO
AAYXX
ABFNM
ABWVN
ABXDB
ACLOT
ACNNM
ACRPL
ACVFH
ADCNI
ADMUD
ADNMO
AEIPS
AEUPX
AFPUW
AGQPQ
AIGII
AIIUN
AKBMS
AKYEP
ANKPU
APXCP
ASPBG
AVWKF
AZFZN
BBWZM
CITATION
EFKBS
EFLBG
EJD
FEDTE
FGOYB
HLY
HVGLF
HZ~
NDZJH
R2-
SCB
SCE
T9H
WUQ
XPP
ZY4
~HD
7S9
L.6
ID FETCH-LOGICAL-c385t-5075520f892bc7db52ce873d6d58a42a8109a63be0ed61f39997268c0ba6b0a63
ISICitedReferencesCount 3
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001322191800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0032-5910
IngestDate Sat Sep 27 18:15:28 EDT 2025
Sat Nov 29 02:39:41 EST 2025
Tue Nov 18 20:53:07 EST 2025
Sat Dec 21 16:01:34 EST 2024
IsDoiOpenAccess true
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Keywords AI
Large computations
Neural network
Partial differential equations
Discrete Element Method
GPU-accelerated computing
Language English
License This is an open access article under the CC BY license.
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c385t-5075520f892bc7db52ce873d6d58a42a8109a63be0ed61f39997268c0ba6b0a63
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0002-9091-0352
0000-0003-0721-3749
OpenAccessLink https://dx.doi.org/10.1016/j.powtec.2024.120258
PQID 3154163607
PQPubID 24069
ParticipantIDs proquest_miscellaneous_3154163607
crossref_citationtrail_10_1016_j_powtec_2024_120258
crossref_primary_10_1016_j_powtec_2024_120258
elsevier_sciencedirect_doi_10_1016_j_powtec_2024_120258
PublicationCentury 2000
PublicationDate 2024-12-01
2024-12-00
20241201
PublicationDateYYYYMMDD 2024-12-01
PublicationDate_xml – month: 12
  year: 2024
  text: 2024-12-01
  day: 01
PublicationDecade 2020
PublicationTitle Powder technology
PublicationYear 2024
Publisher Elsevier B.V
Publisher_xml – name: Elsevier B.V
References Norouzi, Zarghami, Mostoufi (b11) 2017; 316
Wu, Wang (b27) 2022; 296
Zhang, Li, Yang, Zhang (b28) 2024
Zhang, Tagliafierro, Heuvel, Sabarwal, Bakke, Yue, Wei, Serban, Negruţ (b51) 2024; 300
Younes, Wautier, Wan, Millet, Nicot, Bouchard (b23) 2023; 162
PyTorch (b39) 2024
Zhao, Zhou, Bai, El-Emam, Agarwal (b16) 2023; 35
Lu, Gao, Dietiker, Shahnam, Rogers (b24) 2021; 245
Cheng, Guo, Lai (b43) 2000; 107
Li, Duan, Sakai (b2) 2024; 61
Alizadeh, Hassanpour, Pasha, Ghadiri, Bayly (b41) 2017; 319
Sakai (b12) 2016; 33
Govender, Wilke, Wu, Rajamani, Khinast, Glasser (b15) 2018; 29
NVIDIA Corporation (b48) 2021
Tsuji, Kawaguchi, Tanaka (b38) 1993; 77
Fang, Zhang, Vanden Heuvel, Serban, Negrut (b17) 2021; 9
Chen (b34) 2024
Phillips, Heaney, Chen, Buchan, Pain (b32) 2023; 124
He, Li, Chen (b13) 2024
Chen, Heaney, Pain (b31) 2024
Zhang, Vanden Heuvel, Schepelmann, Rogg, Apostolopoulos, Chandler, Serban, Negrut (b19) 2024
Wu, Wu, Zakhvatayeva, Wang, Liu, Yang, Zheng, Wu (b4) 2022; 78
Kelly, Olsen, Negrut (b49) 2020; 50
Li (b37) 2024
Dong, Yan, Cui (b1) 2022; 12
Bhalode, Ierapetritou (b3) 2020; 585
Zhao, Zhao (b10) 2023; 416
Liu, Xu, Govender, Wilke (b5) 2021; 377
Zhou, Xu, Lubbe (b20) 2021; 380
Wu, Wang (b26) 2021; 11
Tian, Zhang, Lin, Yang, Yang, Yang (b14) 2017; 104
Wang, Ihme, Chen, Anderson (b30) 2022; 274
Lai, Chen, Huang (b25) 2022; 46
Zheng, Zang, Chen, Zeng (b21) 2018; 209
Ahmadian, Zheng (b40) 2024; 171
Phillips, Heaney, Chen, Buchan, Pain (b35) 2023
Govender, Wilke, Kok (b18) 2016; 5
Gujjala, Kim, Ryu (b8) 2023; 30
Jaggannagari, Desu, Reimann, Gan, Moscardini, Annabattula (b44) 2021; 393
Song, Wang, Makse (b42) 2008; 453
Scott (b46) 1960; 188
Zhao, Xu, Ye, Liu (b29) 2020; 32
Jian, Gao (b50) 2023; 34
Spellings, Marson, Anderson, Glotzer (b6) 2017; 334
Sánchez, Scheeres (b7) 2011; 727
Liu, Wu (b22) 2020; 4
Yan, Regueiro (b9) 2018; 75
Li, Xiang, Chen, Heaney, Dargaville, Pain (b36) 2024
Chen, Heaney, Gomes, Matar, Pain (b33) 2024; 426
Qian, Wang, An, Wu, Wang, Zhao, Yang (b45) 2018; 325
NVIDIA Corporation (b47) 2020
Chen (10.1016/j.powtec.2024.120258_b31) 2024
Govender (10.1016/j.powtec.2024.120258_b18) 2016; 5
Li (10.1016/j.powtec.2024.120258_b2) 2024; 61
Govender (10.1016/j.powtec.2024.120258_b15) 2018; 29
Wu (10.1016/j.powtec.2024.120258_b4) 2022; 78
Zhao (10.1016/j.powtec.2024.120258_b29) 2020; 32
He (10.1016/j.powtec.2024.120258_b13) 2024
Kelly (10.1016/j.powtec.2024.120258_b49) 2020; 50
Zhao (10.1016/j.powtec.2024.120258_b10) 2023; 416
Chen (10.1016/j.powtec.2024.120258_b34) 2024
Phillips (10.1016/j.powtec.2024.120258_b35) 2023
Cheng (10.1016/j.powtec.2024.120258_b43) 2000; 107
Li (10.1016/j.powtec.2024.120258_b36) 2024
Zhou (10.1016/j.powtec.2024.120258_b20) 2021; 380
Spellings (10.1016/j.powtec.2024.120258_b6) 2017; 334
Jaggannagari (10.1016/j.powtec.2024.120258_b44) 2021; 393
Tsuji (10.1016/j.powtec.2024.120258_b38) 1993; 77
Gujjala (10.1016/j.powtec.2024.120258_b8) 2023; 30
Phillips (10.1016/j.powtec.2024.120258_b32) 2023; 124
Lai (10.1016/j.powtec.2024.120258_b25) 2022; 46
Wu (10.1016/j.powtec.2024.120258_b26) 2021; 11
Dong (10.1016/j.powtec.2024.120258_b1) 2022; 12
Zhao (10.1016/j.powtec.2024.120258_b16) 2023; 35
NVIDIA Corporation (10.1016/j.powtec.2024.120258_b47) 2020
Sánchez (10.1016/j.powtec.2024.120258_b7) 2011; 727
Tian (10.1016/j.powtec.2024.120258_b14) 2017; 104
Zhang (10.1016/j.powtec.2024.120258_b51) 2024; 300
Ahmadian (10.1016/j.powtec.2024.120258_b40) 2024; 171
Younes (10.1016/j.powtec.2024.120258_b23) 2023; 162
Zheng (10.1016/j.powtec.2024.120258_b21) 2018; 209
Wu (10.1016/j.powtec.2024.120258_b27) 2022; 296
Bhalode (10.1016/j.powtec.2024.120258_b3) 2020; 585
Wang (10.1016/j.powtec.2024.120258_b30) 2022; 274
Qian (10.1016/j.powtec.2024.120258_b45) 2018; 325
Liu (10.1016/j.powtec.2024.120258_b5) 2021; 377
Fang (10.1016/j.powtec.2024.120258_b17) 2021; 9
Alizadeh (10.1016/j.powtec.2024.120258_b41) 2017; 319
Zhang (10.1016/j.powtec.2024.120258_b28) 2024
Jian (10.1016/j.powtec.2024.120258_b50) 2023; 34
Norouzi (10.1016/j.powtec.2024.120258_b11) 2017; 316
Yan (10.1016/j.powtec.2024.120258_b9) 2018; 75
Zhang (10.1016/j.powtec.2024.120258_b19) 2024
NVIDIA Corporation (10.1016/j.powtec.2024.120258_b48) 2021
Lu (10.1016/j.powtec.2024.120258_b24) 2021; 245
PyTorch (10.1016/j.powtec.2024.120258_b39) 2024
Song (10.1016/j.powtec.2024.120258_b42) 2008; 453
Liu (10.1016/j.powtec.2024.120258_b22) 2020; 4
Scott (10.1016/j.powtec.2024.120258_b46) 1960; 188
Chen (10.1016/j.powtec.2024.120258_b33) 2024; 426
Li (10.1016/j.powtec.2024.120258_b37) 2024
Sakai (10.1016/j.powtec.2024.120258_b12) 2016; 33
References_xml – volume: 107
  start-page: 123
  year: 2000
  end-page: 130
  ident: b43
  article-title: Dynamic simulation of random packing of spherical particles
  publication-title: Powder Technol.
– volume: 104
  start-page: 231
  year: 2017
  end-page: 240
  ident: b14
  article-title: Implementing discrete element method for large-scale simulation of particles on multiple GPUs
  publication-title: Comput. Chem. Eng.
– volume: 453
  start-page: 629
  year: 2008
  end-page: 632
  ident: b42
  article-title: A phase diagram for jammed matter
  publication-title: Nature
– year: 2024
  ident: b31
  article-title: Using AI libraries for incompressible computational fluid dynamics
  publication-title: arXiv
– volume: 296
  year: 2022
  ident: b27
  article-title: Prediction of 3D contact force chains using artificial neural networks
  publication-title: Eng. Geol.
– volume: 30
  start-page: 1601
  year: 2023
  end-page: 1622
  ident: b8
  article-title: GPGPU-based parallel computation using discrete elements in geotechnics: A state-of-art review
  publication-title: Arch. Comput. Methods Eng.
– volume: 12
  start-page: 3107
  year: 2022
  ident: b1
  article-title: An efficient parallel framework for the discrete element method using GPU
  publication-title: Appl. Sci.
– year: 2020
  ident: b47
  article-title: NVIDIA Tensor Cores
– year: 2021
  ident: b48
  article-title: CUDA Programming Guide
– volume: 9
  start-page: 1813
  year: 2021
  ident: b17
  article-title: Chrono:: GPU: An open-source simulation package for granular dynamics using the discrete element method
  publication-title: Processes
– year: 2024
  ident: b13
  article-title: A novel unresolved/semi-resolved CFD-DEM coupling method with dynamic unstructured mesh
  publication-title: Int. J. Numer. Anal. Methods Geomech.
– volume: 5
  start-page: 62
  year: 2016
  end-page: 66
  ident: b18
  article-title: Blaze-DEMGPU: Modular high performance DEM framework for the GPU architecture
  publication-title: SoftwareX
– year: 2024
  ident: b28
  article-title: Prediction of constrained modulus for granular soil using 3D discrete element method and convolutional neural networks
  publication-title: J. Rock Mech. Geotech. Eng.
– volume: 32
  start-page: 735
  year: 2020
  end-page: 746
  ident: b29
  article-title: A TensorFlow-based new high-performance computational framework for CFD
  publication-title: J. Hydrodyn.
– volume: 188
  start-page: 908
  year: 1960
  end-page: 909
  ident: b46
  article-title: Packing of spheres: packing of equal spheres
  publication-title: Nature
– volume: 29
  start-page: 2476
  year: 2018
  end-page: 2490
  ident: b15
  article-title: Large-scale GPU based DEM modeling of mixing using irregularly shaped particles
  publication-title: Adv. Powder Technol.
– volume: 75
  start-page: 61
  year: 2018
  end-page: 87
  ident: b9
  article-title: Superlinear speedup phenomenon in parallel 3D Discrete Element Method (DEM) simulations of complex-shaped particles
  publication-title: Parallel Comput.
– volume: 416
  year: 2023
  ident: b10
  article-title: Revolutionizing granular matter simulations by high-performance ray tracing discrete element method for arbitrarily-shaped particles
  publication-title: Comput. Methods Appl. Mech. Engrg.
– volume: 319
  start-page: 313
  year: 2017
  end-page: 322
  ident: b41
  article-title: The effect of particle shape on predicted segregation in binary powder mixtures
  publication-title: Powder Technol.
– volume: 33
  start-page: 169
  year: 2016
  end-page: 178
  ident: b12
  article-title: How should the discrete element method be applied in industrial systems?: A review
  publication-title: KONA Powder Part. J.
– volume: 46
  start-page: 113
  year: 2022
  end-page: 140
  ident: b25
  article-title: Machine-learning-enabled discrete element method: Contact detection and resolution of irregular-shaped particles
  publication-title: Int. J. Numer. Anal. Methods Geomech.
– year: 2023
  ident: b35
  article-title: Solving the discretised Boltzmann transport equations using neural networks: Applications in neutron transport
– volume: 585
  year: 2020
  ident: b3
  article-title: Discrete element modeling for continuous powder feeding operation: Calibration and system analysis
  publication-title: Int. J. Pharm.
– volume: 245
  year: 2021
  ident: b24
  article-title: Machine learning accelerated discrete element modeling of granular flows
  publication-title: Chem. Eng. Sci.
– volume: 300
  year: 2024
  ident: b51
  article-title: Chrono DEM-Engine: A discrete element method dual-GPU simulator with customizable contact forces and element shape
  publication-title: Comput. Phys. Comm.
– volume: 380
  start-page: 394
  year: 2021
  end-page: 407
  ident: b20
  article-title: Multi-scale mechanics of sand based on FEM-DEM coupling method
  publication-title: Powder Technol.
– volume: 325
  start-page: 151
  year: 2018
  end-page: 160
  ident: b45
  article-title: DEM simulation on the vibrated packing densification of mono-sized equilateral cylindrical particles
  publication-title: Powder Technol.
– volume: 11
  start-page: 6278
  year: 2021
  ident: b26
  article-title: Estimating contact force chains using artificial neural network
  publication-title: Appl. Sci.
– volume: 124
  start-page: 4659
  year: 2023
  end-page: 4686
  ident: b32
  article-title: Solving the discretised neutron diffusion equations using neural networks
  publication-title: Internat. J. Numer. Methods Engrg.
– volume: 35
  year: 2023
  ident: b16
  article-title: Modeling and validation of coarse-grained computational fluid dynamics–discrete element method for dense gas—solid flow simulation in a bubbling fluidized bed
  publication-title: Phys. Fluids
– volume: 77
  start-page: 79
  year: 1993
  end-page: 87
  ident: b38
  article-title: Discrete particle simulation of two-dimensional fluidized bed
  publication-title: Powder Technol.
– year: 2024
  ident: b39
  article-title: Pairwisedistance
– volume: 50
  start-page: 355
  year: 2020
  end-page: 379
  ident: b49
  article-title: Billion degree of freedom granular dynamics simulation on commodity hardware via heterogeneous data-type representation
  publication-title: Multibody Syst. Dyn.
– volume: 209
  start-page: 74
  year: 2018
  end-page: 92
  ident: b21
  article-title: A GPU-based DEM-FEM computational framework for tire-sand interaction simulations
  publication-title: Comput. Struct.
– volume: 426
  year: 2024
  ident: b33
  article-title: Solving the discretised multiphase flow equations with interface capturing on structured grids using machine learning libraries
  publication-title: Comput. Methods Appl. Mech. Engrg.
– volume: 162
  year: 2023
  ident: b23
  article-title: DEM-LBM coupling for partially saturated granular assemblies
  publication-title: Comput. Geotech.
– volume: 34
  year: 2023
  ident: b50
  article-title: Investigation of spherical and non-spherical binary particles flow characteristics in a discharge hopper
  publication-title: Adv. Powder Technol.
– volume: 171
  year: 2024
  ident: b40
  article-title: Simulating the fluid–solid interaction of irregularly shaped particles using the LBM-DEM coupling method
  publication-title: Comput. Geotech.
– volume: 78
  year: 2022
  ident: b4
  article-title: Influence of moisture content on die filling of pharmaceutical powders
  publication-title: J. Drug Deliv. Sci. Technol.
– start-page: 1
  year: 2024
  end-page: 21
  ident: b19
  article-title: A GPU-accelerated simulator for the DEM analysis of granular systems composed of clump-shaped elements
  publication-title: Eng. Comput.
– volume: 4
  year: 2020
  ident: b22
  article-title: Modelling complex particle–fluid flow with a discrete element method coupled with Lattice Boltzmann methods (DEM-LBM)
  publication-title: ChemEngineering
– volume: 393
  start-page: 31
  year: 2021
  end-page: 59
  ident: b44
  article-title: DEM simulations of vibrated sphere packings in slender prismatic containers
  publication-title: Powder Technol.
– year: 2024
  ident: b36
  article-title: Implementing the discontinuous-Galerkin finite element method using graph neural networks
– volume: 727
  start-page: 120
  year: 2011
  ident: b7
  article-title: Simulating asteroid rubble piles with a self-gravitating soft-sphere distinct element method model
  publication-title: Astrophys. J.
– year: 2024
  ident: b37
  article-title: An AI-based integrated framework for anisotropic electrical resistivity imaging
– year: 2024
  ident: b34
  article-title: Solving the discretised shallow water equations using neural networks
– volume: 334
  start-page: 460
  year: 2017
  end-page: 467
  ident: b6
  article-title: GPU accelerated discrete element method (DEM) molecular dynamics for conservative, faceted particle simulations
  publication-title: J. Comput. Phys.
– volume: 61
  start-page: 285
  year: 2024
  end-page: 306
  ident: b2
  article-title: DEM simulations in nuclear engineering: a review of recent progress
  publication-title: J. Nucl. Sci. Technol.
– volume: 377
  start-page: 640
  year: 2021
  end-page: 656
  ident: b5
  article-title: Simulation of rock fracture process based on GPU-accelerated discrete element method
  publication-title: Powder Technol.
– volume: 274
  year: 2022
  ident: b30
  article-title: A TensorFlow simulation framework for scientific computing of fluid flows on tensor processing units
  publication-title: Comput. Phys. Comm.
– volume: 316
  start-page: 233
  year: 2017
  end-page: 244
  ident: b11
  article-title: New hybrid CPU-GPU solver for CFD-DEM simulation of fluidized beds
  publication-title: Powder Technol.
– year: 2024
  ident: 10.1016/j.powtec.2024.120258_b37
– volume: 33
  start-page: 169
  year: 2016
  ident: 10.1016/j.powtec.2024.120258_b12
  article-title: How should the discrete element method be applied in industrial systems?: A review
  publication-title: KONA Powder Part. J.
  doi: 10.14356/kona.2016023
– volume: 35
  issue: 4
  year: 2023
  ident: 10.1016/j.powtec.2024.120258_b16
  article-title: Modeling and validation of coarse-grained computational fluid dynamics–discrete element method for dense gas—solid flow simulation in a bubbling fluidized bed
  publication-title: Phys. Fluids
– volume: 30
  start-page: 1601
  issue: 3
  year: 2023
  ident: 10.1016/j.powtec.2024.120258_b8
  article-title: GPGPU-based parallel computation using discrete elements in geotechnics: A state-of-art review
  publication-title: Arch. Comput. Methods Eng.
  doi: 10.1007/s11831-022-09851-3
– volume: 274
  year: 2022
  ident: 10.1016/j.powtec.2024.120258_b30
  article-title: A TensorFlow simulation framework for scientific computing of fluid flows on tensor processing units
  publication-title: Comput. Phys. Comm.
  doi: 10.1016/j.cpc.2022.108292
– start-page: 1
  year: 2024
  ident: 10.1016/j.powtec.2024.120258_b19
  article-title: A GPU-accelerated simulator for the DEM analysis of granular systems composed of clump-shaped elements
  publication-title: Eng. Comput.
– year: 2021
  ident: 10.1016/j.powtec.2024.120258_b48
– volume: 416
  year: 2023
  ident: 10.1016/j.powtec.2024.120258_b10
  article-title: Revolutionizing granular matter simulations by high-performance ray tracing discrete element method for arbitrarily-shaped particles
  publication-title: Comput. Methods Appl. Mech. Engrg.
  doi: 10.1016/j.cma.2023.116370
– volume: 727
  start-page: 120
  issue: 2
  year: 2011
  ident: 10.1016/j.powtec.2024.120258_b7
  article-title: Simulating asteroid rubble piles with a self-gravitating soft-sphere distinct element method model
  publication-title: Astrophys. J.
  doi: 10.1088/0004-637X/727/2/120
– volume: 75
  start-page: 61
  year: 2018
  ident: 10.1016/j.powtec.2024.120258_b9
  article-title: Superlinear speedup phenomenon in parallel 3D Discrete Element Method (DEM) simulations of complex-shaped particles
  publication-title: Parallel Comput.
  doi: 10.1016/j.parco.2018.03.007
– volume: 32
  start-page: 735
  issue: 4
  year: 2020
  ident: 10.1016/j.powtec.2024.120258_b29
  article-title: A TensorFlow-based new high-performance computational framework for CFD
  publication-title: J. Hydrodyn.
  doi: 10.1007/s42241-020-0050-0
– volume: 585
  year: 2020
  ident: 10.1016/j.powtec.2024.120258_b3
  article-title: Discrete element modeling for continuous powder feeding operation: Calibration and system analysis
  publication-title: Int. J. Pharm.
  doi: 10.1016/j.ijpharm.2020.119427
– volume: 11
  start-page: 6278
  issue: 14
  year: 2021
  ident: 10.1016/j.powtec.2024.120258_b26
  article-title: Estimating contact force chains using artificial neural network
  publication-title: Appl. Sci.
  doi: 10.3390/app11146278
– volume: 209
  start-page: 74
  year: 2018
  ident: 10.1016/j.powtec.2024.120258_b21
  article-title: A GPU-based DEM-FEM computational framework for tire-sand interaction simulations
  publication-title: Comput. Struct.
  doi: 10.1016/j.compstruc.2018.08.011
– volume: 46
  start-page: 113
  issue: 1
  year: 2022
  ident: 10.1016/j.powtec.2024.120258_b25
  article-title: Machine-learning-enabled discrete element method: Contact detection and resolution of irregular-shaped particles
  publication-title: Int. J. Numer. Anal. Methods Geomech.
  doi: 10.1002/nag.3293
– volume: 104
  start-page: 231
  year: 2017
  ident: 10.1016/j.powtec.2024.120258_b14
  article-title: Implementing discrete element method for large-scale simulation of particles on multiple GPUs
  publication-title: Comput. Chem. Eng.
  doi: 10.1016/j.compchemeng.2017.04.019
– volume: 296
  year: 2022
  ident: 10.1016/j.powtec.2024.120258_b27
  article-title: Prediction of 3D contact force chains using artificial neural networks
  publication-title: Eng. Geol.
  doi: 10.1016/j.enggeo.2021.106444
– volume: 334
  start-page: 460
  year: 2017
  ident: 10.1016/j.powtec.2024.120258_b6
  article-title: GPU accelerated discrete element method (DEM) molecular dynamics for conservative, faceted particle simulations
  publication-title: J. Comput. Phys.
  doi: 10.1016/j.jcp.2017.01.014
– volume: 162
  year: 2023
  ident: 10.1016/j.powtec.2024.120258_b23
  article-title: DEM-LBM coupling for partially saturated granular assemblies
  publication-title: Comput. Geotech.
  doi: 10.1016/j.compgeo.2023.105677
– volume: 9
  start-page: 1813
  issue: 10
  year: 2021
  ident: 10.1016/j.powtec.2024.120258_b17
  article-title: Chrono:: GPU: An open-source simulation package for granular dynamics using the discrete element method
  publication-title: Processes
  doi: 10.3390/pr9101813
– volume: 380
  start-page: 394
  year: 2021
  ident: 10.1016/j.powtec.2024.120258_b20
  article-title: Multi-scale mechanics of sand based on FEM-DEM coupling method
  publication-title: Powder Technol.
  doi: 10.1016/j.powtec.2020.11.006
– year: 2024
  ident: 10.1016/j.powtec.2024.120258_b13
  article-title: A novel unresolved/semi-resolved CFD-DEM coupling method with dynamic unstructured mesh
  publication-title: Int. J. Numer. Anal. Methods Geomech.
  doi: 10.1002/nag.3701
– volume: 377
  start-page: 640
  year: 2021
  ident: 10.1016/j.powtec.2024.120258_b5
  article-title: Simulation of rock fracture process based on GPU-accelerated discrete element method
  publication-title: Powder Technol.
  doi: 10.1016/j.powtec.2020.09.009
– volume: 107
  start-page: 123
  issue: 1–2
  year: 2000
  ident: 10.1016/j.powtec.2024.120258_b43
  article-title: Dynamic simulation of random packing of spherical particles
  publication-title: Powder Technol.
  doi: 10.1016/S0032-5910(99)00178-3
– volume: 393
  start-page: 31
  year: 2021
  ident: 10.1016/j.powtec.2024.120258_b44
  article-title: DEM simulations of vibrated sphere packings in slender prismatic containers
  publication-title: Powder Technol.
  doi: 10.1016/j.powtec.2021.06.033
– volume: 188
  start-page: 908
  issue: 4754
  year: 1960
  ident: 10.1016/j.powtec.2024.120258_b46
  article-title: Packing of spheres: packing of equal spheres
  publication-title: Nature
  doi: 10.1038/188908a0
– volume: 12
  start-page: 3107
  issue: 6
  year: 2022
  ident: 10.1016/j.powtec.2024.120258_b1
  article-title: An efficient parallel framework for the discrete element method using GPU
  publication-title: Appl. Sci.
  doi: 10.3390/app12063107
– volume: 453
  start-page: 629
  issue: 7195
  year: 2008
  ident: 10.1016/j.powtec.2024.120258_b42
  article-title: A phase diagram for jammed matter
  publication-title: Nature
  doi: 10.1038/nature06981
– volume: 171
  year: 2024
  ident: 10.1016/j.powtec.2024.120258_b40
  article-title: Simulating the fluid–solid interaction of irregularly shaped particles using the LBM-DEM coupling method
  publication-title: Comput. Geotech.
  doi: 10.1016/j.compgeo.2024.106395
– volume: 245
  year: 2021
  ident: 10.1016/j.powtec.2024.120258_b24
  article-title: Machine learning accelerated discrete element modeling of granular flows
  publication-title: Chem. Eng. Sci.
  doi: 10.1016/j.ces.2021.116832
– volume: 325
  start-page: 151
  year: 2018
  ident: 10.1016/j.powtec.2024.120258_b45
  article-title: DEM simulation on the vibrated packing densification of mono-sized equilateral cylindrical particles
  publication-title: Powder Technol.
  doi: 10.1016/j.powtec.2017.10.050
– volume: 34
  issue: 5
  year: 2023
  ident: 10.1016/j.powtec.2024.120258_b50
  article-title: Investigation of spherical and non-spherical binary particles flow characteristics in a discharge hopper
  publication-title: Adv. Powder Technol.
  doi: 10.1016/j.apt.2023.104011
– volume: 316
  start-page: 233
  year: 2017
  ident: 10.1016/j.powtec.2024.120258_b11
  article-title: New hybrid CPU-GPU solver for CFD-DEM simulation of fluidized beds
  publication-title: Powder Technol.
  doi: 10.1016/j.powtec.2016.11.061
– year: 2024
  ident: 10.1016/j.powtec.2024.120258_b34
– volume: 77
  start-page: 79
  issue: 1
  year: 1993
  ident: 10.1016/j.powtec.2024.120258_b38
  article-title: Discrete particle simulation of two-dimensional fluidized bed
  publication-title: Powder Technol.
  doi: 10.1016/0032-5910(93)85010-7
– year: 2020
  ident: 10.1016/j.powtec.2024.120258_b47
– volume: 5
  start-page: 62
  year: 2016
  ident: 10.1016/j.powtec.2024.120258_b18
  article-title: Blaze-DEMGPU: Modular high performance DEM framework for the GPU architecture
  publication-title: SoftwareX
  doi: 10.1016/j.softx.2016.04.004
– year: 2024
  ident: 10.1016/j.powtec.2024.120258_b39
– year: 2024
  ident: 10.1016/j.powtec.2024.120258_b36
– volume: 124
  start-page: 4659
  issue: 21
  year: 2023
  ident: 10.1016/j.powtec.2024.120258_b32
  article-title: Solving the discretised neutron diffusion equations using neural networks
  publication-title: Internat. J. Numer. Methods Engrg.
  doi: 10.1002/nme.7321
– volume: 4
  issue: 4
  year: 2020
  ident: 10.1016/j.powtec.2024.120258_b22
  article-title: Modelling complex particle–fluid flow with a discrete element method coupled with Lattice Boltzmann methods (DEM-LBM)
  publication-title: ChemEngineering
  doi: 10.3390/chemengineering4040055
– year: 2024
  ident: 10.1016/j.powtec.2024.120258_b31
  article-title: Using AI libraries for incompressible computational fluid dynamics
  publication-title: arXiv
– year: 2023
  ident: 10.1016/j.powtec.2024.120258_b35
– volume: 61
  start-page: 285
  issue: 3
  year: 2024
  ident: 10.1016/j.powtec.2024.120258_b2
  article-title: DEM simulations in nuclear engineering: a review of recent progress
  publication-title: J. Nucl. Sci. Technol.
  doi: 10.1080/00223131.2023.2231969
– volume: 50
  start-page: 355
  year: 2020
  ident: 10.1016/j.powtec.2024.120258_b49
  article-title: Billion degree of freedom granular dynamics simulation on commodity hardware via heterogeneous data-type representation
  publication-title: Multibody Syst. Dyn.
  doi: 10.1007/s11044-020-09749-7
– volume: 426
  year: 2024
  ident: 10.1016/j.powtec.2024.120258_b33
  article-title: Solving the discretised multiphase flow equations with interface capturing on structured grids using machine learning libraries
  publication-title: Comput. Methods Appl. Mech. Engrg.
  doi: 10.1016/j.cma.2024.116974
– volume: 300
  year: 2024
  ident: 10.1016/j.powtec.2024.120258_b51
  article-title: Chrono DEM-Engine: A discrete element method dual-GPU simulator with customizable contact forces and element shape
  publication-title: Comput. Phys. Comm.
  doi: 10.1016/j.cpc.2024.109196
– volume: 29
  start-page: 2476
  issue: 10
  year: 2018
  ident: 10.1016/j.powtec.2024.120258_b15
  article-title: Large-scale GPU based DEM modeling of mixing using irregularly shaped particles
  publication-title: Adv. Powder Technol.
  doi: 10.1016/j.apt.2018.06.028
– volume: 319
  start-page: 313
  year: 2017
  ident: 10.1016/j.powtec.2024.120258_b41
  article-title: The effect of particle shape on predicted segregation in binary powder mixtures
  publication-title: Powder Technol.
  doi: 10.1016/j.powtec.2017.06.059
– year: 2024
  ident: 10.1016/j.powtec.2024.120258_b28
  article-title: Prediction of constrained modulus for granular soil using 3D discrete element method and convolutional neural networks
  publication-title: J. Rock Mech. Geotech. Eng.
– volume: 78
  year: 2022
  ident: 10.1016/j.powtec.2024.120258_b4
  article-title: Influence of moisture content on die filling of pharmaceutical powders
  publication-title: J. Drug Deliv. Sci. Technol.
SSID ssj0006310
Score 2.4530609
Snippet This paper introduces a novel methodology, the Neural Network framework for the Discrete Element Method (NN4DEM), as part of a broader initiative to harness...
SourceID proquest
crossref
elsevier
SourceType Aggregation Database
Enrichment Source
Index Database
Publisher
StartPage 120258
SubjectTerms computer software
Discrete Element Method
environment
GPU-accelerated computing
Large computations
Neural network
neural networks
Partial differential equations
particles
physics
powders
seeds
solutions
testing
weight
Title A discrete element solution method embedded within a Neural Network
URI https://dx.doi.org/10.1016/j.powtec.2024.120258
https://www.proquest.com/docview/3154163607
Volume 448
WOSCitedRecordID wos001322191800001&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: 0032-5910
  databaseCode: AIEXJ
  dateStart: 19950101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0006310
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3Pb9sgFEZZusN2mPZT69pNTNrNcmQbg-GYVa3aHqIeMimnWWBwlai1o8b9sf--jwBJ1mjretjFsrDBxO_L4_H88YHQNyoLo2pC41RoFue1lrFguo6pYjm3MULF1HKziWI04pOJOOv1foa1MDcXRdPwuzsx_6-mhjIwtl06-wRzrxqFAjgHo8MRzA7HfzL80H50gViwM5Fx3PAoPNDvFx2ZS2XA4Tji-bSJZGQ1OsBYI0cK34xYz9pbKzfRbaXgR5YDPXVZZT2Tek0UcJ7se_tL-mHRehWflx63zfkakJOpLz6FkwVUPN_MQmT5BqPDe1YCk1rhKares-ZORNP7xhTqOZn2LbftMgizwby9hZ8zsA8YrG__XSX7wei14hQGutqsdK2UtpXStfIM7WQF9K6PdoYnh5PT1VjNSOqFO13vw-LKJQNwuzd_Cl4eDOPL2GT8Gr3ykwo8dGB4g3qmeYtebkhNvkMHQxxggT0scIAFdrDAARbYwQJL7GCBPSzeox9Hh-OD49hvoBFXhNMuhlif0iypuchUVWhFs8rwgmimKZd5JnmaCMmIMonRLK0hVhVFxniVKMlUAlc-oH7TNuYjwlLXXIncVAQceG64pNKuYTa14qqmQu0iEt5MWXl1ebvJyUX5N7vsonhVa-7UVR65vwgvvfQRoov8SkDSIzW_BhuV4EDtVzHZmPZ6URKYRKRWNa_49MTe7KEX67_CPup3V9fmM3pe3XTTxdUXD7V700eVLg
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=A+discrete+element+solution+method+embedded+within+a+Neural+Network&rft.jtitle=Powder+technology&rft.au=Naderi%2C+Sadjad&rft.au=Chen%2C+Boyang&rft.au=Yang%2C+Tongan&rft.au=Xiang%2C+Jiansheng&rft.date=2024-12-01&rft.issn=0032-5910&rft.volume=448&rft.spage=120258&rft_id=info:doi/10.1016%2Fj.powtec.2024.120258&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_powtec_2024_120258
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0032-5910&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0032-5910&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0032-5910&client=summon