Innovative Multi-scale/Multi-physics Model for Surface Finish Prediction and Optimization of Metal Additively Manufactured Parts

In this STTR effort, TDA and its team partner University of Louisville will focus on developing an innovative intelligent decision support tool using data-driven multi-scale multi-physics models (DDMM) to derive process-surface roughness relationships for selective laser melting (SLM). The proposed models account for both powder characteristics and AM processing/path planning, including powder size distribution, laser power, scanning speed, scanning strategy, geometry features, and build orientation. Critical experiments will be performed during the course of this research as part of verification and validation. The goal of this DDMM computational framework is to predict and optimize component-level surface roughness within a reasonable time. Proposed DDMM computational framework will address surface roughness caused by rippling marks, balling effect, staircase effect, and sintered powders. Solution for the DDMM framework will be obtained by discrete element method (DEM), computational fluid dynamics (CFD), finite element method (FEM) and data-driven modeling.