Optimizing Metal 3D Printing using Artificial Neural Network based Seagull Optimization Algorithm

The infill parameters are important according to overall saving and cost material when printing metal 3D model. The Fused Deposition Modeling (FDM) is applied in different industries to formed functional regions through polymers and metal reinforced polymer mixtures to obtain chosen properties. The Laser Powder Bed Fusion (LPBF) is complex to complete as surface roughness which is based on orientation. Hence, this research explored the finishing effects SS 316L LPBF components in different parts. The Artificial Neural Network (ANN) based Seagull Optimization Algorithm (SOA) is proposed for optimizing metal 3D printing. Moreover, the impact of critical speed at milling procedure on surface roughness SS 316L LPBF is analyzed with layer thickness and orientations of LPBF components. The printing temperature, layer height, compressive, flexural and tensile strength are considered as input and output parameters. The result exposed that machined surface suffers important orientation and layer thickness variations. The ANN-SOA achieved
$0.21\mu \mathrm{m}, 0.47\mu \mathrm{m},0.19\mu \mathrm{m}, 0.38\mu \mathrm{m}, 0.15\mu \mathrm{m}$
and
$0.56\mu \mathrm{m}$
for 1–6 experiments respectively.