PSI - Issue 48

Available online at www.sciencedirect.com Available online at www.sciencedirect.com ScienceDirect Structural Integrity Procedia 00 (2023) 000 – 000 Available online at www.sciencedirect.com ScienceDirect Structural Integrity Procedia 00 (2023) 000 – 000

www.elsevier.com/locate/procedia www.elsevier.com/locate/procedia

ScienceDirect

Procedia Structural Integrity 48 (2023) 183–189

© 2023 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers Abstract There was modeled the stress-strain diagram of 6061-T651 aluminum alloy by machine learning methods. In this study, methods of k -nearest neighbors and random forest were applied to obtain the best model for predicting the stress-strain diagram of 6061 T651 aluminum alloy at six temperatures (20, 100, 150, 200, 250, 300ºС). The obtained results agree well with the experimental data. It was determined that the errors of 9.6% and 5.9% for linear and non-linear regions, respectively, were obtained by the method of k -nearest neighbors in the test sample. The errors of the random forest method were 15% and 13.7%. © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers Keywords: stress-strain diagram, machine learning, method of k -nearest neighbors, random forest, 6061-T651 aliminum alloy 1. Introduction Stress-strain diagrams of materials are built to determine their mechanical characteristics. In particular, the curve will have a different shape depending on the type of material, its state, and the conditions under which the testing was carried out. That is, these are the dependences of strength parameters on deformation ones. In general, stress-strain diagrams are constructed by different methods. In particular, in the paper by Molkov (2013), the conditional and the actual deformation curves of 65G spring carbon steel is plotted using the standard approach, and the digital image correlation technique (DIC) is constructed and shows a good correlation of the obtained results. In the paper by Pylypenko et al. (2009) the advantages of the complete stress-strain softening diagrams for estimation of limiting material damage under complex Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) Application of machine learning for modeling of 6061-T651 aluminum alloy stress−strain diagram Oleh Yasniy a , Oleh Pastukh a , Iryna Didych a, *, Vasyl Yatsyshyn a , Ihor Chykhira a a Ternopil Ivan Puluj National Technical University, 56, Ruska Str., Ternopil, 46001, Ukraine Abstract There was modeled the stress-strain diagram of 6061-T651 aluminum alloy by machine learning methods. In this study, methods of k -nearest neighbors and random forest were applied to obtain the best model for predicting the stress-strain diagram of 6061 T651 aluminum alloy at six temperatures (20, 100, 150, 200, 250, 300ºС). The obtained results agree well with the experimental data. It was determined that the errors of 9.6% and 5.9% for linear and non-linear regions, respectively, were obtained by the method of k -nearest neighbors in the test sample. The errors of the random forest method were 15% and 13.7%. © 2023 The Authors. Published by ELSEVIER B.V. Keywords: stress-strain diagram, machine learning, method of k -nearest neighbors, random forest, 6061-T651 aliminum alloy 1. Introduction Stress-strain diagrams of materials are built to determine their mechanical characteristics. In particular, the curve will have a different shape depending on the type of material, its state, and the conditions under which the testing was carried out. That is, these are the dependences of strength parameters on deformation ones. In general, stress-strain diagrams are constructed by different methods. In particular, in the paper by Molkov (2013), the conditional and the actual deformation curves of 65G spring carbon steel is plotted using the standard approach, and the digital image correlation technique (DIC) is constructed and shows a good correlation of the obtained results. In the paper by Pylypenko et al. (2009) the advantages of the complete stress-strain softening diagrams for estimation of limiting material damage under complex Second International Symposium on Risk Analysis and Safety of Complex Structures and Components (IRAS 2023) Application of machine learning for modeling of 6061-T651 aluminum alloy stress−strain diagram Oleh Yasniy a , Oleh Pastukh a , Iryna Didych a, *, Vasyl Yatsyshyn a , Ihor Chykhira a a Ternopil Ivan Puluj National Technical University, 56, Ruska Str., Ternopil, 46001, Ukraine

* Corresponding author. Tel.: /; fax: /. E-mail address : iryna.didych1101@gmail.com * Corresponding author. Tel.: /; fax: /. E-mail address : iryna.didych1101@gmail.com

2452-3216 © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers 2452-3216 © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers

2452-3216 © 2023 The Authors. Published by ELSEVIER B.V. This is an open access article under the CC BY-NC-ND license (https://creativecommons.org/licenses/by-nc-nd/4.0) Peer-review under responsibility of the IRAS 2023 organizers 10.1016/j.prostr.2023.07.146