"MACHINE LEARNING" METHOD FOR THE SYNTHESIS OF A NiTiHf ALLOY

Abstract

Nitinol is a promising functional material used in aerospace, biotechnical, medical and other areas of modern high technology. The reason for this is its unique technical properties such as shape memory effect and superelasticity. These properties are manifested during phase transformations. In this paper, using machine learning (ML) methods, we present the results of studying the dependence of the transition temperature on processing conditions during the synthesis of the NiTiHf alloy, which is a modification of nitinol and has a wide range of applications. Machine methods such as GPR, SVM and Linear Regression have been applied. The online program Jupyter Notebook, which was created based on Python 3.9, was chosen as the software. In addition to discussing the results obtained, the paper makes suggestions for improving the results obtained.