Search academic texts

Information × Registration Number 0225U001608, (0122U001858) , R & D reports Title Modelling of functional and structural properties of shape memory alloys by machine learning methods popup.stage_title Порівняння результатів прогнозування функціональних властивостей і втомної довговічності методами машинного навчання та детермінованими і ймовірнісно–статистичними методами. Розробка рекомендацій щодо застосування конкретних методів машинного навчання для прогнозування функціональних властивостей і втомної довговічності сплаву Head Yasnii Oleh P., Доктор технічних наук Registration Date 03-02-2025 Organization Ternopil National Technical University named after Ivan Puluj popup.description1 The development of lifetime assessmnet methods of shape memory alloys based on the revealed regularities of pseudoelastic behavior and lifetime of such elements, taking into account the influence of stress ratio and stress range. popup.description2  As a result of the work, a methodology for predicting fatigue life and functional properties of shape memory alloys was developed based on machine learning methods. The main patterns of the influence of the stress ratio on the functional properties of a pseudoelastic nickel-titanium shape memory alloy were obtained. The main patterns of the influence of the stress ratio on the fatigue life of a pseudoelastic nickel-titanium shape memory alloy were identified. Generalized data were obtained on the change in functional properties and fatigue life, considering the influence of the stress ratio. The errors in predicting the influence of the stress ratio on the functional properties and fatigue life of a nickel-titanium alloy by different methods were compared. The dependences of stress and strain range on the number of loading cycles for the specimens were predicted by machine learning methods. For each of the specimen, two models were built. The input of each model was the dependence of the corresponding physical quantity on the number of loading cycles. The number of loading cycles was considered an independent variable, and the corresponding physical quantity was treated as a dependent variable. Recommendations are given on using specific machine learning methods to predict nickel-titanium alloy functional properties and fatigue life, considering the influence of the stress ratio. Unlike existing methods for predicting the fatigue life of shape memory alloys, models have been built, and methods have been created using various machine learning methods for predicting the fatigue life of shape memory alloys based on the application of fracture criteria, considering the influence of the asymmetry of the loading cycle. The expected scientific and practical value of the results: prevention of possible accidents, destruction of structural elements and structures, and loss of working capacity by individuals. Product Description popup.authors Antonov Andrii M. Baran Denys Ya. Bukiv Nazarii Z. Boiko Andrii R. Brevus Vitalii M. Budz Volodymyr P. Havrysh Roman -. Homon Olha O. Didych Iryna S. Demchyk Vladyslav I. Mazepa Mykola R. Marushchak Olena V. Marchenko Liubov O. Myshkovych Olha V. Okipnyi Ihor B. Pasternak Yaroslav M. Sorochak Andrii P. Student Oleksandra Z. Tsymbaliuk Liubov I. Chornomaz Nataliia Yu. Yarema Ihor T. popup.nrat_date 2025-02-03 Close