A Lagrangian meshfree mesoscale simulation of powder bed fusion additive manufacturing of metals. (13th October 2020)

• Record Type: Journal Article
• Authors: Fan, Zongyue; Wang, Hao; Huang, Zhida; Liao, Huming; Fan, Jiang; Lu, Jian; Liu, Chong; Li, Bo
• Is Part Of: International journal for numerical methods in engineering. Volume 122:Number 2(2021)
• Page Start: 483
• Page End: 514
• Publication Date: 2020-10-13
• Subjects: defects formation -- HOTM method -- meshfree methods -- powder bed fusion additive manufacturing -- powder‐scale simulation; Numerical analysis -- Periodicals; Engineering mathematics -- Periodicals; 620.001518
• DOI: 10.1002/nme.6546
• Languages: English
• ISSNs: 0029-5981

Abstract

Abstract: We present a powder‐scale computational framework to predict the microstructure evolution of metals in powder bed fusion additive manufacturing (PBF AM) processes based on the hot optimal transportation meshfree (HOTM) method. The powder bed is modeled as discrete and deformable three‐dimensional bodies by integrating statistic information from experiments, including particle size and shape, and powder packing density. Tractions in Lagrangian framework are developed to model the recoil pressure and surface tension. The laser beam is applied to surfaces of particles and substrate dynamically as a heat flux with user‐specified beam size, power, scanning speed, and path. The linear momentum and energy conservation equations are formulated in the Lagrangian configuration and solved simultaneously in a monolithic way by the HOTM method to predict the deformation, temperature, contact mechanisms, and fluid‐structure interactions in the powder bed. The numerical results are validated against single track experiments. Various powder bed configurations, laser powers, and speed are investigated to understand the influence of dynamic contact and inelastic material behavior on the deformation, heat transfer, and phase transition of the powder bed. The formation of defects in the microstructure of 3D printed metals, including pores, partially, and unmelted particles, is predicted by the proposed computational scheme.