PSI - Issue 72

Available online at www.sciencedirect.com

ScienceDirect

Procedia Structural Integrity 72 (2025) 188–194

12th Annual Conference of Society for Structural Integrity and Life (DIVK12) Modeling of shape memory alloys hysteresis behavior considering the loading cycle frequency Oleh Yasniy a , Dmytro Tymoshchuk a, *, Iryna Didych a, , Roman Zolotyi a , Vitaliy Tymoshchuk a a Ternopil Ivan Puluj National Technical University, Ruska str. 56, Ternopil, 46001, Ukraine Abstract Shape memory alloys (SMAs) can restore their original shape when exposed to temperature changes or mechanical stress. One of the main characteristics of SMA is superelasticity, which allows the material to return to its original shape after deformation. This property results from phase transitions between martensite and austenite under loading or temperature changes. The loading frequency plays an important role in the behavior of shape memory alloys, as it can significantly affect their functional properties. This study uses artificial neural networks (ANNs) to predict the hysteresis behavior of SMA nickel-titanium alloy at different cyclic loading frequencies. The experimentally obtained hysteresis loops for four loading frequencies were used to train the artificial neural network: 0.1, 0.5, 1, and 5 Hz. The input data contained such parameters as stress σ (MPa), cycle number N , and loading frequency f (Hz). The neural network learned to recognize patterns in the material behavior by analyzing the functional relationships between these parameters. This made it possible to predict the strain  (%) at different frequencies and loading conditions. The results show high accuracy of material behavior predictions, which is confirmed by the small values of MAE, MSE, and MAPE errors. © 2026 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 Aleksandar Sedmak, Branislav Djordjevic, Simon Sedmak Dr. Simon Sedmak, ssedmak@mas.bg.ac.rs, Innovation Center of Faculty of Mechanical Engineering, Belgrade, Serbia

Keywords: SMA; artificial intelligence; machine learning; frequency; hysteresis

* Corresponding author. Tel.: +380352519700. E-mail address: dmytro.tymoshchuk@gmail.com

2452-3216 © 2026 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 Aleksandar Sedmak, Branislav Djordjevic, Simon Sedmak Dr. Simon Sedmak, ssedmak@mas.bg.ac.rs, Innovation Center of Faculty of Mechanical Engineering, Belgrade, Serbia 10.1016/j.prostr.2025.08.091