Application of artificial neural networks for the alloy features assessment based on the alloy compound

Nickel-based superalloys are compositions with complex doping. A large amount of data on the chemical composition and heat resistance of nickel alloys has been accumulated, however, there is also a lot of information on the properties that is absent and this is a problem. Modern information technologies have allowed substantiating a new approach to the analysis of service characteristics of metallic materials using artificial neural networks. The previously obtained real test data served as the basis for evaluating the missing values. Transformation of input data significantly increases the accuracy of calculations. As a result, a database for 210 marks of nickel-based superalloys and their heat resistance under different conditions was obtained.