Computationally efficient finite difference method for metal additive manufacturing: A reduced-order DFAM tool applied to SLM

Highlights: • Metal Additive Manufacturing (MAM) simulation is computationally demanding. • Geometric simplifications allow MAM to be simulated in one-dimension (1D). • A tri-diagonal system of equations is developed to represent the MAM component. • Temperature profiles qualitatively match observations in MAM components. Metal Additive Manufacturing (MAM) allows the production of complex lattice structures that exceed the manufacturing capability of traditional processes. However, the MAM process is highly complex, including: transient thermo-mechanical loading, spatially and temporally transient boundary conditions and multi-scale affects. Furthermore, MAM is subject to significant experimental and statistical uncertainties. Consequently, the MAM process is poorly understood and build-process simulations are computationally demanding; this limits the availability of Design for Additive manufacturing (DFAM) tools, and necessitates experimental validation for commercial MAM applications. This research develops a novel finite difference method (FDM) simulation of the MAM temperature field that is based on a computationally efficient 1D transient tri-diagonal system of equations, and reduces data generation effort by the reuse of existing MAM production data. The 1D simulation is particularly suited to lattice geometries due to the slenderness of lattice strut elements. The reduced-order simulation method developed in this research provides a timely and useful DFAM tool that enables qualitative design insight early in the design phase and pre-production validation. The proposed method is validated by comparison to existing analytical results, numerical results and by application to a titanium and aluminium lattice structures manufactured by a commercial Selective Laser Melting (SLM) process.