Design of Artificial Neural Networks for Damage Estimation of Composite Laminates: Application to Delamination Failures in Ply Drops

This work presents a data-driven approach based on Artificial Neural Networks (ANN), that benefits from parametric non-linear finite element analyses, in order to provide a “cheaper” numerical alternative to the more expensive experimental testing of advanced composite laminates. The chosen subject...

Full description

Saved in:

Bibliographic Details
Published in:	Composite structures Vol. 304; no. 1; p. 116320
Main Authors:	Mendoza, Arturo, Friderikos, Orestis, Trullo, Roger, Baranger, Emmanuel
Format:	Journal Article
Language:	English
Published:	Elsevier Ltd 15.01.2023 Elsevier
Subjects:	Artificial Neural Networks Computer Aided Engineering Computer Science Damage mechanics Engineering Sciences Manifold learning Mechanics Mechanics of materials Neural and Evolutionary Computing Neural architecture learning Nonlinear finite element analysis Damage mechanics Nonlinear finite element analysis Manifold learning Artificial Neural Networks Neural architecture learning Artificial Neural Networks (ANN)
ISSN:	0263-8223, 1879-1085
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	This work presents a data-driven approach based on Artificial Neural Networks (ANN), that benefits from parametric non-linear finite element analyses, in order to provide a “cheaper” numerical alternative to the more expensive experimental testing of advanced composite laminates. The chosen subject of study is the damage evolution and associated delamination failures of ply drops. As such, physical-based modeling of these phenomena are used as training data for obtaining the most suitable ANN capable of estimating the damage evolution using only the prescribed initial conditions (i.e., laminate geometry, material orientation, applied loading). Additionally, this work aims at providing first-hand experience into the procedure that leads to obtaining such ANN. In order to do so, we detail each of the required steps, such as choosing a network architecture, defining a custom loss function, as well as deciding on the better learning parameters. Each stage on this process is guided by intuitive statistical analyses and simple criteria, such as selecting the “simpler” model over whenever a more complex one “improves” on the performance but only by a narrow margin. Indeed, for every step, multiple numerical experiences are provided so that the reader can also get a deeper understanding on the inner-workings of these ANN models. With this goal in mind, we employ dimensionality reduction techniques (PCA and t-SNE) to also propose a geometrical visualization of the nonlinear transformations performed by the ANN. Finally, additional tests regarding the network ability to generalize to unseen data showed that the optimal well-trained ANN is accurate and robust enough for near real-time predictions of the various damage evolution patterns, and outperforms other data-driven methods under comparison.
AbstractList	This work presents a data-driven approach based on Artificial Neural Networks (ANN), that benefits from parametric non-linear finite element analyses, in order to provide a “cheaper” numerical alternative to the more expensive experimental testing of advanced composite laminates. The chosen subject of study is the damage evolution and associated delamination failures of ply drops. As such, physical-based modeling of these phenomena are used as training data for obtaining the most suitable ANN capable of estimating the damage evolution using only the prescribed initial conditions (i.e., laminate geometry, material orientation, applied loading). Additionally, this work aims at providing first-hand experience into the procedure that leads to obtaining such ANN. In order to do so, we detail each of the required steps, such as choosing a network architecture, defining a custom loss function, as well as deciding on the better learning parameters. Each stage on this process is guided by intuitive statistical analyses and simple criteria, such as selecting the “simpler” model over whenever a more complex one “improves” on the performance but only by a narrow margin. Indeed, for every step, multiple numerical experiences are provided so that the reader can also get a deeper understanding on the inner-workings of these ANN models. With this goal in mind, we employ dimensionality reduction techniques (PCA and t-SNE) to also propose a geometrical visualization of the nonlinear transformations performed by the ANN. Finally, additional tests regarding the network ability to generalize to unseen data showed that the optimal well-trained ANN is accurate and robust enough for near real-time predictions of the various damage evolution patterns, and outperforms other data-driven methods under comparison. This work proposes a data driven approach which utilizes Artificial Neural Networks (ANN) in conjunction with parametric non-linear finite element analysis. The aim is to provide a low cost numerical counterpart to the expensive experimental testing of advanced composite laminates. The training data of ANN are obtained from physical based modeling of the damage evolution and associated delamination failures of ply drops. In contrast to a black-box ANN modeling approach, the core of this study concerns the development of a method for determining an optimal neural architecture. More specifically, we employed a random search handcrafted methodology for the neural net topology learning based on exploration and experimentation. This methodology is enhanced by a detailed statistical analysis used to make inferences about the procedural generation of architectures. In the same context, a series of experiments are performed to obtain an optimal set of hyperparameters to achieve a good performance in the training dataset. Furthermore, a visualization of the respective manifolds of the ANNs hidden layers is provided using two popular dimensionality reduction techniques, namely PCA and t-SNE, so as to transform the network layer output data into 2D representations. Additional tests, among others, regarding the network ability to generalize to unseen data, showed that the optimal well-trained neural network is accurate and robust enough for near real-time predictions of the various damage evolution patterns, and outperforms other data driven methods under comparison.
ArticleNumber	116320
Author	Mendoza, Arturo Trullo, Roger Baranger, Emmanuel Friderikos, Orestis
Author_xml	– sequence: 1 givenname: Arturo orcidid: 0000-0002-1746-9054 surname: Mendoza fullname: Mendoza, Arturo email: arturo.mendoza-quispe@safrangroup.com organization: Université Paris-Saclay, ENS Paris-Saclay, CentraleSupélec, CNRS, LMPS – Laboratoire de Mécanique Paris-Saclay, 4 avenue des sciences, 91190 Gif-sur-Yvette, France – sequence: 2 givenname: Orestis surname: Friderikos fullname: Friderikos, Orestis organization: Université Paris-Saclay, ENS Paris-Saclay, CentraleSupélec, CNRS, LMPS – Laboratoire de Mécanique Paris-Saclay, 4 avenue des sciences, 91190 Gif-sur-Yvette, France – sequence: 3 givenname: Roger surname: Trullo fullname: Trullo, Roger organization: Safran Tech, Rue des Jeunes Bois, 78772 Magny les Hameaux, France – sequence: 4 givenname: Emmanuel orcidid: 0000-0001-6295-9776 surname: Baranger fullname: Baranger, Emmanuel organization: Université Paris-Saclay, ENS Paris-Saclay, CentraleSupélec, CNRS, LMPS – Laboratoire de Mécanique Paris-Saclay, 4 avenue des sciences, 91190 Gif-sur-Yvette, France
BackLink	https://hal.science/hal-03826330$$DView record in HAL
BookMark	eNqNkEFPGzEQha2KSg2U_-Arh009dth4OSClCZRKUeEAZ8txxnTSzXplOyDu_PA62VaVeoHTyJ733uh9x-yoCx0yxkGMQUD9ZTN2YdunHHcuj6WQcgxQKyk-sBHoaVOB0OdHbCRkrSotpfrEjlPaCCH0BGDEXheY6LHjwfNZzOTJkW35D9zFw8jPIf5K3IfIF3ZrH5FfpUxbmykcPPNyOyTKyJd2S53NmC74rO9bcoMmB77Adtjt39eW2l3ExKnjd-0LX8TQp8_so7dtwtM_84Q9XF_dz2-q5e237_PZsnIKpqJaoQJcgT9vYOUnyk90XVvvpIapqnFtrXTYoG6kB71SZYXa1WvAxmk7dWqiTtjZkPvTtqaPpUd8McGSuZktzf5PKF04KfEERXs5aF0MKUX0xlE-dMixdDAgzJ6_2Zh__M2evxn4lwD9X8Dfi--wfh2sWGA8EUaTHGHncE0Ri3Yd6O2Q30sBq1M
CitedBy_id	crossref_primary_10_1016_j_heliyon_2024_e33681 crossref_primary_10_1016_j_tws_2024_112346 crossref_primary_10_1038_s41598_025_02390_4
Cites_doi	10.1016/j.compstruct.2005.01.020 10.1016/0266-3538(96)00005-X 10.1016/j.compositesa.2010.05.009 10.1016/j.paerosci.2005.02.001 10.1016/j.compstruct.2019.111653 10.1109/CVPR.2016.90 10.1557/mrc.2019.32 10.1007/s10443-021-09891-1 10.1016/j.compstruct.2006.04.047 10.1016/j.compstruct.2021.114287 10.1016/j.matdes.2004.04.008 10.1016/j.compscitech.2020.108573 10.1109/IJCNN.2001.939002 10.1177/073168449401300804 10.1016/j.matdes.2018.05.049 10.1109/COGANN.1992.273950 10.1016/S1359-835X(97)00102-4 10.3389/fpsyg.2018.01185 10.1080/00401706.2000.10485979 10.1016/j.compstruct.2020.113131 10.1037/h0071325 10.1016/S0266-3538(03)00106-4 10.1016/j.compstruct.2018.08.014
ContentType	Journal Article
Copyright	2022 Elsevier Ltd Distributed under a Creative Commons Attribution 4.0 International License
Copyright_xml	– notice: 2022 Elsevier Ltd – notice: Distributed under a Creative Commons Attribution 4.0 International License
DBID	AAYXX CITATION 1XC VOOES
DOI	10.1016/j.compstruct.2022.116320
DatabaseName	CrossRef Hyper Article en Ligne (HAL) Hyper Article en Ligne (HAL) (Open Access)
DatabaseTitle	CrossRef
DatabaseTitleList
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering Computer Science
EISSN	1879-1085
ExternalDocumentID	oai:HAL:hal-03826330v1 10_1016_j_compstruct_2022_116320 S0263822322010522
GroupedDBID	--K --M .~1 0R~ 1B1 1~. 1~5 4.4 457 4G. 5GY 5VS 6TJ 7-5 71M 8P~ 9JN AABNK AABXZ AACTN AAEDT AAEDW AAEPC AAIAV AAIKJ AAKOC AALRI AAOAW AAQFI AAXUO ABMAC ABXRA ABYKQ ACDAQ ACGFS ACRLP ADBBV ADEZE ADTZH AEBSH AECPX AEKER AENEX AEZYN AFKWA AFRZQ AFTJW AGHFR AGUBO AGYEJ AHHHB AHJVU AIEXJ AIKHN AITUG AJOXV ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AXJTR BJAXD BKOJK BLXMC CS3 DU5 EBS EFJIC EFLBG EO8 EO9 EP2 EP3 FDB FIRID FNPLU FYGXN G-Q GBLVA IHE J1W JJJVA KOM LY7 M24 M41 MAGPM MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 RNS ROL RPZ SDF SDG SES SPC SPCBC SSM SST SSZ T5K XPP ZMT ~02 ~G- 29F 9DU AAQXK AATTM AAXKI AAYWO AAYXX ABFNM ABJNI ABWVN ABXDB ACLOT ACNNM ACRPL ACVFH ADCNI ADIYS ADMUD ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKRWK AKYEP ANKPU APXCP ASPBG AVWKF AZFZN CITATION EFKBS EJD FEDTE FGOYB G-2 HVGLF HZ~ R2- SET SEW SMS WUQ ~HD 1XC VOOES
ID	FETCH-LOGICAL-c3170-be31eb1f591bf43f4866afc281736edaa2ce9e892f18b36afe8c6d1e9c8a7c343
ISSN	0263-8223
IngestDate	Tue Oct 14 20:30:56 EDT 2025 Sat Nov 29 07:19:50 EST 2025 Tue Nov 18 21:02:18 EST 2025 Fri Feb 23 02:38:32 EST 2024
IsDoiOpenAccess	true
IsOpenAccess	true
IsPeerReviewed	true
IsScholarly	true
Issue	1
Keywords	Damage mechanics Nonlinear finite element analysis Manifold learning Artificial Neural Networks Neural architecture learning Artificial Neural Networks (ANN)
Language	English
License	Distributed under a Creative Commons Attribution 4.0 International License: http://creativecommons.org/licenses/by/4.0
LinkModel	OpenURL
MergedId	FETCHMERGED-LOGICAL-c3170-be31eb1f591bf43f4866afc281736edaa2ce9e892f18b36afe8c6d1e9c8a7c343
ORCID	0000-0002-1746-9054 0000-0001-6295-9776
OpenAccessLink	https://hal.science/hal-03826330
ParticipantIDs	hal_primary_oai_HAL_hal_03826330v1 crossref_citationtrail_10_1016_j_compstruct_2022_116320 crossref_primary_10_1016_j_compstruct_2022_116320 elsevier_sciencedirect_doi_10_1016_j_compstruct_2022_116320
PublicationCentury	2000
PublicationDate	2023-01-15
PublicationDateYYYYMMDD	2023-01-15
PublicationDate_xml	– month: 01 year: 2023 text: 2023-01-15 day: 15
PublicationDecade	2020
PublicationTitle	Composite structures
PublicationYear	2023
Publisher	Elsevier Ltd Elsevier
Publisher_xml	– name: Elsevier Ltd – name: Elsevier
References	Chakraborty (b6) 2005; 26 Hotelling (b44) 1933; 24 Di Benedetto, Botelho, Janotti, Ancelotti Junior, Gomes (b9) 2021; 257 He, Zhang, Ren, Sun (b27) 2015 Khan, Kim (b8) 2018; 206 Khatir, Tiachacht, Le Thanh, Ghandourah, Mirjalili, Wahab (b15) 2021; 273 He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proceedings of the IEEE computer society conference on computer vision and pattern recognition, volume 2016-December, 2016. p. 770–778. Yang, Fu, Lee, Li (b10) 2021; 28 El Kadi (b17) 2006; 73 Allegri, Wisnom, Hallett (b18) 2010; 41 Zhang, Friedrich (b16) 2003; 63 Haykin (b25) 1999 Benzeggagh, Kenane (b22) 1996; 56 Schaffer JD, Whitley D, Eshelman LJ. Combinations of genetic algorithms and neural networks: A survey of the state of the art. In: International workshop on combinations of genetic algorithms and neural networks, 1992. p. 1–37. Cui, Wisnom, Jones (b19) 1994; 13 Kingma, Ba (b28) 2015 Snoek, Larochelle, Adams (b34) 2012; Vol. 4 Saxena, Verbeek (b40) 2016 Chen, Gu (b4) 2019; 9 Deenadayalu, Chattopadhyay, Chen (b5) 2004 Vapnik (b26) 1992; Vol. 4 Kesavan, Deivasigamani, John, Herszberg (b7) 2006; 75 Bishop (b24) 1995 Ronneberger, Fischer, Brox (b29) 2015; Vol. 9351 Zhang, Allegri, Hallett (b14) 2022; 236–237 Kitano (b38) 1990; Vol. 4 Petrossian, Wisnom (b20) 1998; 29 (b23) 2014 Han S, Mao H, Dally WJ. Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding. In: 4th international conference on learning representations, ICLR 2016 - conference track proceedings, 2016. Queipo, Haftka, Shyy, Goel, Vaidyanathan, Kevin Tucker (b2) 2005; 41 Krizhevsky (b43) 2014 McKay, Beckman, Conover (b21) 2000; 42 Hodas, Stinis (b41) 2018; 9 Guillon D, Lemasçon A, Callens C. QSP : An innovative process based on tailored preformsfor low cost and fast production of optimized thermoplastic composite parts. In: ECCM17-17th European conference on composite materials, 2016. p. 1. Allegri (b11) 2018; 155 Baker B, Gupta O, Naik N, Raskar R. Designing neural network architectures using reinforcement learning. In: 5th international conference on learning representations, ICLR 2017 - conference track proceedings, 2019. Zoph, Le (b36) 2017 Bergstra J, Yamins D, Cox DD. Making a science of model search: Hyperparameter optimization in hundreds of dimensions for vision architectures. In: 30th international conference on machine learning, ICML 2013, 2013. p. 115–23. Tao, Zhang, Ji, Qiu (b13) 2021; 203 Wilson DR, Martinez TR. The need for small learning rates on large problems. In: Proceedings of the international joint conference on neural networks, Vol. 1, 2001. p. 115–9. Altabey, Noori (b12) 2017; 3 Van Der Maaten, Hinton (b45) 2008; 9 Elsken, Metzen, Hutter (b31) 2019 Kyriakides, Margaritis (b32) 2020 Friderikos, Baranger, Guillon (b3) 2020; 234 Khan (10.1016/j.compstruct.2022.116320_b8) 2018; 206 Queipo (10.1016/j.compstruct.2022.116320_b2) 2005; 41 Zoph (10.1016/j.compstruct.2022.116320_b36) 2017 Chen (10.1016/j.compstruct.2022.116320_b4) 2019; 9 El Kadi (10.1016/j.compstruct.2022.116320_b17) 2006; 73 Zhang (10.1016/j.compstruct.2022.116320_b14) 2022; 236–237 Hodas (10.1016/j.compstruct.2022.116320_b41) 2018; 9 Ronneberger (10.1016/j.compstruct.2022.116320_b29) 2015; Vol. 9351 Allegri (10.1016/j.compstruct.2022.116320_b18) 2010; 41 Vapnik (10.1016/j.compstruct.2022.116320_b26) 1992; Vol. 4 Khatir (10.1016/j.compstruct.2022.116320_b15) 2021; 273 10.1016/j.compstruct.2022.116320_b42 Elsken (10.1016/j.compstruct.2022.116320_b31) 2019 He (10.1016/j.compstruct.2022.116320_b27) 2015 Deenadayalu (10.1016/j.compstruct.2022.116320_b5) 2004 Haykin (10.1016/j.compstruct.2022.116320_b25) 1999 Altabey (10.1016/j.compstruct.2022.116320_b12) 2017; 3 Van Der Maaten (10.1016/j.compstruct.2022.116320_b45) 2008; 9 Hotelling (10.1016/j.compstruct.2022.116320_b44) 1933; 24 Petrossian (10.1016/j.compstruct.2022.116320_b20) 1998; 29 Friderikos (10.1016/j.compstruct.2022.116320_b3) 2020; 234 10.1016/j.compstruct.2022.116320_b37 Kesavan (10.1016/j.compstruct.2022.116320_b7) 2006; 75 (10.1016/j.compstruct.2022.116320_b23) 2014 10.1016/j.compstruct.2022.116320_b39 Kitano (10.1016/j.compstruct.2022.116320_b38) 1990; Vol. 4 10.1016/j.compstruct.2022.116320_b33 Yang (10.1016/j.compstruct.2022.116320_b10) 2021; 28 McKay (10.1016/j.compstruct.2022.116320_b21) 2000; 42 Kyriakides (10.1016/j.compstruct.2022.116320_b32) 2020 10.1016/j.compstruct.2022.116320_b35 Saxena (10.1016/j.compstruct.2022.116320_b40) 2016 Benzeggagh (10.1016/j.compstruct.2022.116320_b22) 1996; 56 Kingma (10.1016/j.compstruct.2022.116320_b28) 2015 10.1016/j.compstruct.2022.116320_b30 Snoek (10.1016/j.compstruct.2022.116320_b34) 2012; Vol. 4 Allegri (10.1016/j.compstruct.2022.116320_b11) 2018; 155 Zhang (10.1016/j.compstruct.2022.116320_b16) 2003; 63 Bishop (10.1016/j.compstruct.2022.116320_b24) 1995 10.1016/j.compstruct.2022.116320_b1 Tao (10.1016/j.compstruct.2022.116320_b13) 2021; 203 Krizhevsky (10.1016/j.compstruct.2022.116320_b43) 2014 Di Benedetto (10.1016/j.compstruct.2022.116320_b9) 2021; 257 Cui (10.1016/j.compstruct.2022.116320_b19) 1994; 13 Chakraborty (10.1016/j.compstruct.2022.116320_b6) 2005; 26
References_xml	– volume: 9 start-page: 556 year: 2019 end-page: 566 ident: b4 article-title: Machine learning for composite materials publication-title: MRS Commun – start-page: 1 year: 2015 ident: b28 article-title: Adam: A method for stochastic optimization publication-title: 3rd international conference on learning representations, ICLR 2015 - conference track proceedings – year: 2019 ident: b31 article-title: Neural architecture search: A survey – start-page: 1026 year: 2015 end-page: 1034 ident: b27 article-title: Delving deep into rectifiers: Surpassing human-level performance on imagenet classification publication-title: Proceedings of the 2015 IEEE international conference on computer vision (ICCV), ICCV ’15 – volume: 26 start-page: 1 year: 2005 end-page: 7 ident: b6 article-title: Artificial neural network based delamination prediction in laminated composites publication-title: Mater Des – volume: 42 start-page: 55 year: 2000 end-page: 61 ident: b21 article-title: A comparison of three methods for selecting values of input variables in the analysis of output from a computer code publication-title: Technometrics – year: 2014 ident: b43 article-title: One weird trick for parallelizing convolutional neural networks – year: 2020 ident: b32 article-title: An introduction to neural architecture search for convolutional networks – volume: 13 start-page: 722 year: 1994 end-page: 739 ident: b19 article-title: An experimental and analytical study of delamination of unidirectional specimens with cut central plies publication-title: J Reinf Plast Compos – reference: Han S, Mao H, Dally WJ. Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding. In: 4th international conference on learning representations, ICLR 2016 - conference track proceedings, 2016. – volume: 41 start-page: 1 year: 2005 end-page: 28 ident: b2 article-title: Surrogate-based analysis and optimization publication-title: Prog Aerosp Sci – volume: Vol. 4 start-page: 831 year: 1992 end-page: 838 ident: b26 article-title: Principles of risk minimization for learning theory publication-title: Advances in Neural Information Processing Systems – volume: 273 year: 2021 ident: b15 article-title: An improved artificial neural network using arithmetic optimization algorithm for damage assessment in fgm composite plates publication-title: Compos Struct – volume: 234 year: 2020 ident: b3 article-title: Regularized least squares for the building of a sizing criterion based on damage mechanics publication-title: Compos Struct – reference: Baker B, Gupta O, Naik N, Raskar R. Designing neural network architectures using reinforcement learning. In: 5th international conference on learning representations, ICLR 2017 - conference track proceedings, 2019. – reference: He K, Zhang X, Ren S, Sun J. Deep residual learning for image recognition. In: Proceedings of the IEEE computer society conference on computer vision and pattern recognition, volume 2016-December, 2016. p. 770–778. – reference: Wilson DR, Martinez TR. The need for small learning rates on large problems. In: Proceedings of the international joint conference on neural networks, Vol. 1, 2001. p. 115–9. – volume: 3 start-page: 53 year: 2017 end-page: 78 ident: b12 article-title: Fatigue life prediction for carbon fibre/epoxy laminate composites under spectrum loading using two different neural network architectures publication-title: Int J Sustain Mater Struct Syst – start-page: 3874 year: 2004 end-page: 3903 ident: b5 article-title: Characterization and detection of delamination in composite laminates using artificial neural networks – volume: 73 start-page: 1 year: 2006 end-page: 23 ident: b17 article-title: Modeling the mechanical behavior of fiber-reinforced polymeric composite materials using artificial neural networks - a review publication-title: Compos Struct – volume: 206 start-page: 354 year: 2018 end-page: 362 ident: b8 article-title: Assessment of delaminated smart composite laminates via system identification and supervised learning publication-title: Compos Struct – volume: 29 start-page: 403 year: 1998 end-page: 414 ident: b20 article-title: Parametric study of delamination in composites with discontinuous plies using an analytical solution based on fracture mechanics publication-title: Composites A – volume: Vol. 4 start-page: 461 year: 1990 end-page: 476 ident: b38 article-title: Designing neural networks using genetic algorithms with graph generation system publication-title: Complex Systems – volume: 28 start-page: 809 year: 2021 end-page: 833 ident: b10 article-title: Artificial neural network (ANN)-based residual strength prediction of carbon fibre reinforced composites (CFRCs) after impact publication-title: Appl Compos Mater – year: 2014 ident: b23 article-title: Dassault systèmes, abaqus version 6.14.2, user’s manual – reference: Schaffer JD, Whitley D, Eshelman LJ. Combinations of genetic algorithms and neural networks: A survey of the state of the art. In: International workshop on combinations of genetic algorithms and neural networks, 1992. p. 1–37. – volume: 24 start-page: 417 year: 1933 end-page: 441 ident: b44 article-title: Analysis of a complex of statistical variables into principal components publication-title: J Educational Psychol – volume: 56 start-page: 439 year: 1996 end-page: 449 ident: b22 article-title: Measurement of mixed-mode delamination fracture toughness of unidirectional glass/epoxy composites with mixed-mode bending apparatus publication-title: Compos Sci Technol – reference: Bergstra J, Yamins D, Cox DD. Making a science of model search: Hyperparameter optimization in hundreds of dimensions for vision architectures. In: 30th international conference on machine learning, ICML 2013, 2013. p. 115–23. – volume: 257 year: 2021 ident: b9 article-title: Development of an artificial neural network for predicting energy absorption capability of thermoplastic commingled composites publication-title: Compos Struct – volume: Vol. 9351 start-page: 234 year: 2015 end-page: 241 ident: b29 article-title: U-net: Convolutional networks for biomedical image segmentation publication-title: Lecture notes in computer science (including subseries lecture notes in artificial intelligence and lecture notes in bioinformatics) – year: 1999 ident: b25 article-title: Neural networks: A comprehensive foundation – start-page: 4060 year: 2016 end-page: 4068 ident: b40 article-title: Convolutional neural fabrics publication-title: Advances in neural information processing systems – volume: Vol. 4 start-page: 2951 year: 2012 end-page: 2959 ident: b34 article-title: Practical Bayesian optimization of machine learning algorithms publication-title: Advances in neural information processing systems – volume: 75 start-page: 313 year: 2006 end-page: 320 ident: b7 article-title: Damage detection in T-joint composite structures publication-title: Compos Struct – volume: 155 start-page: 59 year: 2018 end-page: 70 ident: b11 article-title: Modelling fatigue delamination growth in fibre-reinforced composites: Power-law equations or artificial neural networks? publication-title: Mater Des – year: 1995 ident: b24 article-title: Neural networks for pattern recognition – volume: 9 year: 2018 ident: b41 article-title: Doing the impossible: Why neural networks can be trained at all publication-title: Front Psychol – reference: Guillon D, Lemasçon A, Callens C. QSP : An innovative process based on tailored preformsfor low cost and fast production of optimized thermoplastic composite parts. In: ECCM17-17th European conference on composite materials, 2016. p. 1. – year: 2017 ident: b36 article-title: Neural architecture search with reinforcement learning – volume: 63 start-page: 2029 year: 2003 end-page: 2044 ident: b16 article-title: Artificial neural networks applied to polymer composites: a review publication-title: Compos Sci Technol – volume: 203 year: 2021 ident: b13 article-title: Application of neural network to model stiffness degradation for composite laminates under cyclic loadings publication-title: Compos Sci Technol – volume: 236–237 year: 2022 ident: b14 article-title: Embedding artificial neural networks into twin cohesive zone models for composites fatigue delamination prediction under various stress ratios and mode mixities publication-title: Int J Solids Struct – volume: 9 start-page: 2579 year: 2008 end-page: 2625 ident: b45 article-title: Visualizing data using t-SNE publication-title: J Mach Learn Res – volume: 41 start-page: 1388 year: 2010 end-page: 1394 ident: b18 article-title: A simplified approach to the damage tolerance design of asymmetric tapered laminates, part I: Methodology development publication-title: Composites A – year: 2020 ident: 10.1016/j.compstruct.2022.116320_b32 – volume: 73 start-page: 1 year: 2006 ident: 10.1016/j.compstruct.2022.116320_b17 article-title: Modeling the mechanical behavior of fiber-reinforced polymeric composite materials using artificial neural networks - a review publication-title: Compos Struct doi: 10.1016/j.compstruct.2005.01.020 – volume: Vol. 9351 start-page: 234 year: 2015 ident: 10.1016/j.compstruct.2022.116320_b29 article-title: U-net: Convolutional networks for biomedical image segmentation – start-page: 3874 year: 2004 ident: 10.1016/j.compstruct.2022.116320_b5 – volume: 56 start-page: 439 year: 1996 ident: 10.1016/j.compstruct.2022.116320_b22 article-title: Measurement of mixed-mode delamination fracture toughness of unidirectional glass/epoxy composites with mixed-mode bending apparatus publication-title: Compos Sci Technol doi: 10.1016/0266-3538(96)00005-X – year: 2017 ident: 10.1016/j.compstruct.2022.116320_b36 – volume: 41 start-page: 1388 year: 2010 ident: 10.1016/j.compstruct.2022.116320_b18 article-title: A simplified approach to the damage tolerance design of asymmetric tapered laminates, part I: Methodology development publication-title: Composites A doi: 10.1016/j.compositesa.2010.05.009 – volume: 41 start-page: 1 year: 2005 ident: 10.1016/j.compstruct.2022.116320_b2 article-title: Surrogate-based analysis and optimization publication-title: Prog Aerosp Sci doi: 10.1016/j.paerosci.2005.02.001 – volume: 234 year: 2020 ident: 10.1016/j.compstruct.2022.116320_b3 article-title: Regularized least squares for the building of a sizing criterion based on damage mechanics publication-title: Compos Struct doi: 10.1016/j.compstruct.2019.111653 – volume: 3 start-page: 53 year: 2017 ident: 10.1016/j.compstruct.2022.116320_b12 article-title: Fatigue life prediction for carbon fibre/epoxy laminate composites under spectrum loading using two different neural network architectures publication-title: Int J Sustain Mater Struct Syst – year: 1995 ident: 10.1016/j.compstruct.2022.116320_b24 – ident: 10.1016/j.compstruct.2022.116320_b30 doi: 10.1109/CVPR.2016.90 – ident: 10.1016/j.compstruct.2022.116320_b39 – ident: 10.1016/j.compstruct.2022.116320_b1 – volume: 9 start-page: 556 year: 2019 ident: 10.1016/j.compstruct.2022.116320_b4 article-title: Machine learning for composite materials publication-title: MRS Commun doi: 10.1557/mrc.2019.32 – volume: 28 start-page: 809 year: 2021 ident: 10.1016/j.compstruct.2022.116320_b10 article-title: Artificial neural network (ANN)-based residual strength prediction of carbon fibre reinforced composites (CFRCs) after impact publication-title: Appl Compos Mater doi: 10.1007/s10443-021-09891-1 – volume: 75 start-page: 313 year: 2006 ident: 10.1016/j.compstruct.2022.116320_b7 article-title: Damage detection in T-joint composite structures publication-title: Compos Struct doi: 10.1016/j.compstruct.2006.04.047 – ident: 10.1016/j.compstruct.2022.116320_b35 – ident: 10.1016/j.compstruct.2022.116320_b33 – volume: 273 year: 2021 ident: 10.1016/j.compstruct.2022.116320_b15 article-title: An improved artificial neural network using arithmetic optimization algorithm for damage assessment in fgm composite plates publication-title: Compos Struct doi: 10.1016/j.compstruct.2021.114287 – year: 2014 ident: 10.1016/j.compstruct.2022.116320_b43 – volume: Vol. 4 start-page: 461 year: 1990 ident: 10.1016/j.compstruct.2022.116320_b38 article-title: Designing neural networks using genetic algorithms with graph generation system – year: 1999 ident: 10.1016/j.compstruct.2022.116320_b25 – volume: 26 start-page: 1 year: 2005 ident: 10.1016/j.compstruct.2022.116320_b6 article-title: Artificial neural network based delamination prediction in laminated composites publication-title: Mater Des doi: 10.1016/j.matdes.2004.04.008 – volume: 203 year: 2021 ident: 10.1016/j.compstruct.2022.116320_b13 article-title: Application of neural network to model stiffness degradation for composite laminates under cyclic loadings publication-title: Compos Sci Technol doi: 10.1016/j.compscitech.2020.108573 – ident: 10.1016/j.compstruct.2022.116320_b42 doi: 10.1109/IJCNN.2001.939002 – year: 2014 ident: 10.1016/j.compstruct.2022.116320_b23 – volume: 13 start-page: 722 year: 1994 ident: 10.1016/j.compstruct.2022.116320_b19 article-title: An experimental and analytical study of delamination of unidirectional specimens with cut central plies publication-title: J Reinf Plast Compos doi: 10.1177/073168449401300804 – start-page: 1026 year: 2015 ident: 10.1016/j.compstruct.2022.116320_b27 article-title: Delving deep into rectifiers: Surpassing human-level performance on imagenet classification – volume: Vol. 4 start-page: 831 year: 1992 ident: 10.1016/j.compstruct.2022.116320_b26 article-title: Principles of risk minimization for learning theory – year: 2019 ident: 10.1016/j.compstruct.2022.116320_b31 – volume: 155 start-page: 59 year: 2018 ident: 10.1016/j.compstruct.2022.116320_b11 article-title: Modelling fatigue delamination growth in fibre-reinforced composites: Power-law equations or artificial neural networks? publication-title: Mater Des doi: 10.1016/j.matdes.2018.05.049 – start-page: 1 year: 2015 ident: 10.1016/j.compstruct.2022.116320_b28 article-title: Adam: A method for stochastic optimization – volume: 9 start-page: 2579 year: 2008 ident: 10.1016/j.compstruct.2022.116320_b45 article-title: Visualizing data using t-SNE publication-title: J Mach Learn Res – ident: 10.1016/j.compstruct.2022.116320_b37 doi: 10.1109/COGANN.1992.273950 – volume: 29 start-page: 403 year: 1998 ident: 10.1016/j.compstruct.2022.116320_b20 article-title: Parametric study of delamination in composites with discontinuous plies using an analytical solution based on fracture mechanics publication-title: Composites A doi: 10.1016/S1359-835X(97)00102-4 – start-page: 4060 year: 2016 ident: 10.1016/j.compstruct.2022.116320_b40 article-title: Convolutional neural fabrics – volume: 9 year: 2018 ident: 10.1016/j.compstruct.2022.116320_b41 article-title: Doing the impossible: Why neural networks can be trained at all publication-title: Front Psychol doi: 10.3389/fpsyg.2018.01185 – volume: 42 start-page: 55 year: 2000 ident: 10.1016/j.compstruct.2022.116320_b21 article-title: A comparison of three methods for selecting values of input variables in the analysis of output from a computer code publication-title: Technometrics doi: 10.1080/00401706.2000.10485979 – volume: 257 year: 2021 ident: 10.1016/j.compstruct.2022.116320_b9 article-title: Development of an artificial neural network for predicting energy absorption capability of thermoplastic commingled composites publication-title: Compos Struct doi: 10.1016/j.compstruct.2020.113131 – volume: 236–237 year: 2022 ident: 10.1016/j.compstruct.2022.116320_b14 article-title: Embedding artificial neural networks into twin cohesive zone models for composites fatigue delamination prediction under various stress ratios and mode mixities publication-title: Int J Solids Struct – volume: 24 start-page: 417 year: 1933 ident: 10.1016/j.compstruct.2022.116320_b44 article-title: Analysis of a complex of statistical variables into principal components publication-title: J Educational Psychol doi: 10.1037/h0071325 – volume: 63 start-page: 2029 year: 2003 ident: 10.1016/j.compstruct.2022.116320_b16 article-title: Artificial neural networks applied to polymer composites: a review publication-title: Compos Sci Technol doi: 10.1016/S0266-3538(03)00106-4 – volume: Vol. 4 start-page: 2951 year: 2012 ident: 10.1016/j.compstruct.2022.116320_b34 article-title: Practical Bayesian optimization of machine learning algorithms – volume: 206 start-page: 354 year: 2018 ident: 10.1016/j.compstruct.2022.116320_b8 article-title: Assessment of delaminated smart composite laminates via system identification and supervised learning publication-title: Compos Struct doi: 10.1016/j.compstruct.2018.08.014
SSID	ssj0008411
Score	2.4038918
Snippet	This work presents a data-driven approach based on Artificial Neural Networks (ANN), that benefits from parametric non-linear finite element analyses, in order... This work proposes a data driven approach which utilizes Artificial Neural Networks (ANN) in conjunction with parametric non-linear finite element analysis....
SourceID	hal crossref elsevier
SourceType	Open Access Repository Enrichment Source Index Database Publisher
StartPage	116320
SubjectTerms	Artificial Neural Networks Computer Aided Engineering Computer Science Damage mechanics Engineering Sciences Manifold learning Mechanics Mechanics of materials Neural and Evolutionary Computing Neural architecture learning Nonlinear finite element analysis
Title	Design of Artificial Neural Networks for Damage Estimation of Composite Laminates: Application to Delamination Failures in Ply Drops
URI	https://dx.doi.org/10.1016/j.compstruct.2022.116320 https://hal.science/hal-03826330
Volume	304
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1879-1085 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0008411 issn: 0263-8223 databaseCode: AIEXJ dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1bb9MwFLa6jgd4QFzFGCAL8TZlSuxc7PFUaKeBqjFBkfoWuakDHW1SZVs1eObP8C85x06cSIBUhHhJm4vjNt9n-9j5zjmEvEg40zNfcWhpMDfBfEie0KHygkjPIxgAI6VMdP1xcnoqplN51uv9aHxhNsukKMT1tVz_V6jhGICNrrN_Abe7KRyA7wA6bAF22G4F_NBoMoyFWRkhEK6JYwwO82FE3yYGAwC-QsHOCBr5yhmO2D-gjksfjBWqZC6tZm7QvudGa3Wol_Ys7h-rBWrbjbD2bPn1YFiV64uu0dve00arxYsd0LqYl9_s4m4FZ0pHqAodBBdfrAzwHaYQWbhSk-qqfmf0vvzU6otfqcq4MZsOfrVSxVXtEFAvazAUdXnWsdOutTX-Nq246cKEiQVWMeuhfKhtly0S6aEPRbdP5zan8S_jg12qOEd41_YvH0LlDAaOmDO_HROdUvEDVok1MpQNgLG6Q3ZZEknRJ7uDN6PpWzfsi9Ake3Y_sZaNWTHh7-v7ky2087lZ1TdWzuQOuV1PT-jA0uou6eniHrnVCVp5n3y3BKNlTluCUUsw2hCMAsGoJRhtCYZlHBmoI9gR7dCLXpa0Sy_a0IsuCgr0ooZeD8jH49Hk9YlX5_LwMo65jWaaB2AW5JEMZnnI81DEscozJoKEx3quFMu01EKyPBAzDqe0yOJ5oGUmVJLxkD8k_aIs9CN4qnkoszkGkoNnDgaviAQUlzDVh6kMj_w9kjRPNc3qQPeYb2WZNorG87TFI0U8UovHHglcybUN9rJFmZcNcGlttFpjNAXObVH6OWDtKsNY7yeDcYrHfKBdzLm_CR7_UxX75Gbbvp6QPlygn5Ib2QZabfWsZvFPrvXSew
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=Design+of+Artificial+Neural+Networks+for+Damage+Estimation+of+Composite+Laminates%3A+Application+to+Delamination+Failures+in+Ply+Drops&rft.jtitle=Composite+structures&rft.au=Mendoza%2C+Arturo&rft.au=Friderikos%2C+Orestis&rft.au=Trullo%2C+Roger&rft.au=Baranger%2C+Emmanuel&rft.date=2023-01-15&rft.pub=Elsevier+Ltd&rft.issn=0263-8223&rft.eissn=1879-1085&rft.volume=304&rft_id=info:doi/10.1016%2Fj.compstruct.2022.116320&rft.externalDocID=S0263822322010522
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0263-8223&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0263-8223&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0263-8223&client=summon