PSI - Issue 68

Oleh Yasniy et al. / Procedia Structural Integrity 68 (2025) 132–138 O. Yasniy et al. / Structural Integrity Procedia 00 (2025) 000–000

137

6

The forecasting quality was assessed by the MAPE value, which was approximately 0.75% for the loading stage and 0.92% for the unloading stage. Fig. 7 shows the hysteresis loop for 127 loading-unloading cycles of the SMA, created by the ANN, and the corresponding experimental data.

Fig. 7. Predicted and experimental hysteresis loop for cycle 127.

The results obtained indicate the high accuracy of the ANN model in predicting the strain e under the influence of stress s , taking into account the load-unload cycle of the SMA material. 4. Conclusions In this paper, various machine learning methods, such as boosted trees, random forests, SVM, KNN, and ANN, were used to model the properties of SMA under the influence of load-unload cycles. Based on the experimental data, models were built to predict the strain of the material e depending on the applied stress s and the cycle number N . Among all the considered methods, the ANN-based model showed the best results, which had the lowest MAPE value. References Abitha, H., Kavitha, V., Gomathi, B., Ramachandran, B., 2020. A recent investigation on shape memory alloys and polymers based materials on bio artificial implants-hip and knee joint. Materials Today: Proceedings 33, 4458-4466. Didych, I. S., Pastukh, O., Pyndus, Y., Yasniy, O., 2018. The evaluation of durability of structural elements using neural networks. Acta Metallurgica Slovaca 24(1), 82-87. Iasnii V., Bykiv N., Yasniy O., Budz V., 2022. Methodology and some results of studying the influence of frequency on functional properties of pseudoelastic SMA. Scientific Journal of TNTU (Tern.), 107(3): 45-50. Klots, Y., Petliak, N., Titova, V., 2023. Evaluation of the efficiency of the system for detecting malicious outgoing traffic in public networks. In 2023 13th International Conference on Dependable Systems, Services and Technologies (DESSERT) (pp. 1-5). IEEE. Lyashuk, O., Stashkiv, M., Lytvynenko, I., Sakhno, V., Khoroshun, R., 2023. Information Technologies Use in the Study of Functional Properties of Wheeled Vehicles. In ITTAP (pp. 500-512). Melchane, S., Elmir, Y., Kacimi, F., 2024. Infectious diseases prediction based on machine learning: the impact of data reduction using feature extraction techniques. Procedia Computer Science 239, 675-683. Petliak, N., Klots, Y., Titova, V., Cheshun, V., Boyarchuk, A., 2023. Signature-based approach to detecting malicious outgoing traffic. 4th International Workshop on Intelligent Information Technologies and Systems of Information Security, IntellTSIS 2023. CEUR Workshop Proceedings, 3373, pp. 486–506. Quan, D., Hai, X., 2015. Shape memory alloy in various aviation field. Procedia Engineering 99, 1241-1246.