A Graph Algorithm for the Design of Functionally Graded Alloy Components

Abstract One of the promising capabilities of metal additive manufacturing (AM) is the potential to manufacture metal parts with spatially varying alloy compositions. It is hypothesized that these compositionally graded alloy (CGA) parts can satisfy competing cost and performance requirements that cannot be met with a monolithic alloy. However, there are significant challenges in their design and manufacture, such as the formation of deleterious material phases and solidification cracking during AM. Prior work addresses the materials design of CGAs between two alloy targets in trivial geometries. However, there are no formal approaches to designing complex CGA parts, limiting the application of these promising materials for further research and implementation. In current practice, most manufactured CGAs are unidirectional, with composition changing in planar, cylindrical, or spherical layers. In this paper, we introduce a novel graph-based algorithm for designing manufacturable general CGA components. Given a component geometry with two or more alloy composition targets at select locations, the algorithm uses a labeled property graph representation of material compatibilities and properties to generate a valid alloy gradient throughout the entire part. We show that the algorithm will always find an alloy gradient solution if one exists, demonstrate it on subsea valve body design problem using the Fe-Ni-Cr ternary system, and study the algorithm’s scalability concerning several key problem discretization parameters.