Fast optimisation procedure for the selection of L-PBF parameters based on utility function
Stefania Cacace, Quirico Semeraro
Abstract
L-PBF is an additive manufacturing process forming parts with complex geometries by adding material layer by layer. The selection of the process parameters in L-PBF has a significant impact on the mechanical properties of the printed parts. Scan speed, laser power, and hatch distance are among the most influential process parameters in L-PBF because, depending on their combination, different solidification mechanisms take place. However, the procedure for selecting these parameters can be expensive from an experimental point of view. Therefore, it is necessary to identify simplified models that allow fast and reliable optimization of the parameters in L-PBF. Furthermore, the choice of parameters cannot be based exclusively on qualitative aspects but must also consider the productivity of the process to obtain a satisfactory compromise. Increasing productivity leads to the formation of lack of fusion porosity which should be avoided. This paper proposes a procedure for selecting parameters based on a semi-analytical thermal model, which, together with a geometric-based defect model, allows identifying an optimality region where good solidification and productivity are considered. The optimization is carried using a properly defined utility function. The procedure is validated through the production of AISI 316L specimens using an industrial L-PBF system.