PSI - Issue 59

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 59 (2024) 17–23

© 2024 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 DMDP 2023 Organizers © 2024 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 DMDP 2023 Organizers Abstract The functional properties of the pseudoelastic alloy were predicted by machine learning methods, namely, the dependence of the dissipated energy and the strain range of NiTi alloy on the number of loading cycles. This study used a multilayered neural network perceptron (MLP) architecture and a random forest method. In particular, the correlation coefficient was 98% when modelling the dependence of the dissipated energy and the strain range of NiTi alloy on the number of loading cycles by the neural network method. The results obtained are in agreement with the experimental data. The random forests method was found to give the lowest prediction error of 3.9% and 7% in the test set of W d - N and Δ ε - N dependences, respectively. In comparison, the error of the neural network method was 5.5% and 7.5%, respectively. © 2024 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 DMDP 2023 Organizers VII International Conference “In -service Damage of Materials: Diagnostics and Prediction ” (DMDP 2023) Estimation of shape memory alloys functional properties by methods of artificial intelligence Oleh Yasniy, Volodymyr Iasnii, Oleh Pastukh, Iryna Didych*, Sergiy Fedak, Sofia Fedak, Lubov Tsymbaliuk Ternopil Ivan Puluj National Technical University, Ruska str. 56, Ternopil, 46001, Ukraine Abstract The functional properties of the pseudoelastic alloy were predicted by machine learning methods, namely, the dependence of the dissipated energy and the strain range of NiTi alloy on the number of loading cycles. This study used a multilayered neural network perceptron (MLP) architecture and a random forest method. In particular, the correlation coefficient was 98% when modelling the dependence of the dissipated energy and the strain range of NiTi alloy on the number of loading cycles by the neural network method. The results obtained are in agreement with the experimental data. The random forests method was found to give the lowest prediction error of 3.9% and 7% in the test set of W d - N and Δ ε - N dependences, respectively. In comparison, the error of the neural network method was 5.5% and 7.5%, respectively. VII International Conference “In -service Damage of Materials: Diagnostics and Prediction ” (DMDP 2023) Estimation of shape memory alloys functional properties by methods of artificial intelligence Oleh Yasniy, Volodymyr Iasnii, Oleh Pastukh, Iryna Didych*, Sergiy Fedak, Sofia Fedak, Lubov Tsymbaliuk Ternopil Ivan Puluj National Technical University, Ruska str. 56, Ternopil, 46001, Ukraine Keywords: pseudoelasticity; functional properties; artificial intelligence; machine learning; random forests; neural networks Keywords: pseudoelasticity; functional properties; artificial intelligence; machine learning; random forests; neural networks

* Corresponding author. Tel.: +380972272074. E-mail address: iryna.didych1101@gmail.com * Corresponding author. Tel.: +380972272074. E-mail address: iryna.didych1101@gmail.com

2452-3216 © 2024 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 DMDP 2023 Organizers 2452-3216 © 2024 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 DMDP 2023 Organizers

2452-3216 © 2024 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 DMDP 2023 Organizers 10.1016/j.prostr.2024.04.004