Powder-bed-fusion additive manufacturing of molybdenum: Process simulation, optimization, and property prediction

Yuhang Wu,Meng Li,Ju Wang,Yang Wang,Xizhong An,Haitao Fu,Hao Zhang,Xiaohong Yang,Qingchuan Zou

DOI: https://doi.org/10.1016/j.addma.2022.103069

IF: 11

2022-10-01

Additive Manufacturing

Abstract:In this paper, the whole process of laser powder-bed-fusion (LPBF) additive manufacturing (AM) in the fabrication of molybdenum (Mo) material was numerically reproduced. Firstly, parameters of Mo powder with continuous size distribution utilized in actual AM production were calibrated and input into discrete element model (DEM) for parametric study on powder spreading. Then, the laser melting of the spread powder bed was simulated by computational fluid dynamics (CFD). On this basis, a back propagation neural network (BPNN) model was built for powder bed quality and molten track property prediction and evaluation. Results show that properly reducing the spreading velocity V or increasing the gap height H can contribute to the optimal structure of the powder bed. Corresponding mechanism analyses reveal that the residual velocity of particles and force chain block are the main reasons for the decrease of powder bed quality. And the molten track performance is not positively correlated with the packing density of the powder bed due to the defects such as balling and porosity caused by the over-thickness of powder bed. The selection rule of powder bed should satisfy as higher as possible the packing density within the bed thickness threshold. The BPNN model can accurately predict the powder bed quality and the molten track property and develop a reasonable map for the appropriate choice of operating parameters in real processes.

engineering, manufacturing,materials science, multidisciplinary