Machine Learning Enhanced Development of Functionally Graded Materials Enabled by Directed Energy Deposition

Abstract

The ability to functionally grade materials provides designers with a new dimension of design flexibility that can be leveraged to improve functionality, reduce cost, or improve efficiency in a wide range of applications. This program was specifically focused on FGMs for hot and harsh gas path environments. These environments are common for the hot sections of jet engines and gas turbines, where parts undergo high temperature and mechanical loads in a corrosive environment. Expensive high γ' strengthened Ni superalloys such as René 41 (R41) and René 80 (R80) are generally used exclusively for a whole part, although only a section of the part demands such superalloys. To minimize cost, low/no γ' strengthened Ni superalloys such as Inconel 718 (INC718) could be used at less-demanding sections of the part could be welded to the high γ' strengthened Ni superalloys. However, the welding typically is a failure site due to the low durability at the welding interface. Functionally grading provides a welding alternative that can allow cost reduction without sacrificing mechanical performance.