Minimizing thermally induced residual stresses in metal additive manufacturing through peridynamics topology optimization

Abstract

Additive manufacturing (AM), characterized by layer-by-layer material addition using computer-aided design, presents an approach to manufacturing complex parts. However, AM processes naturally bring some difficulties such as thermally induced residual stresses, particularly in laser powder bed fusion (L-PBF) processes. This research paper proposes a novel topology optimization (TO) strategy that minimizes the residual stresses on the final product manufactured by metal AM process. In this context, peridynamics topology optimization (PD-TO) is utilized to perform TO within an integrated optimization framework, fed by thermal simulations of the L-PBF process. To calculate the residual stresses on classical TO results, we perform thermomechanical analyses based on the inherent strain method to model the manufacturing process. Afterward, those stress-concentrated regions are used to create virtual cracks for the subsequent PD-TO analysis. This approach allows us to embed potential cracks precisely in these high-stress regions, enhancing the efficacy of structural simulations. The methodology is interpreted through two comprehensive case studies: L-beam and Messerschmitt-Bolkow-Blohm (MBB) beam. The optimization process applied to both L-beam and MBB-Beam structures results in notable enhancements in their respective designs. For the L-beam, the optimization process, leads to a redistribution of material in the beam. This adjustment results in a decrease in residual stress by about 13%. For the MBB-Beam, optimization process induces geometric modifications that yield a stress distribution contributing to a reduction in residual stresses by about 15% for r(min) = 2, and 8% for r(min) = 3. Thus, in both structures, the optimization process effectively improves stress distribution.