Nowadays additive manufacturing is a production process in strong expansion for its adaptability in handling many materials as well as the possibility to create geometries unfeasible with conventional techniques. A distinctive aspect of additive manufacturing is the variability of mechanical properties of the printed pieces, which is mainly due to the physical dependence on process parameters. More in general, both the shape of the piece to manufacture and features of the specific printing technology used strongly influence the distribution of thermal gradients and, consequently, the mechanical characteristics of the final product. In this work, a numerical model was set using commercial software, to predict how the mechanical properties of printed parts are influenced by process parameters. The built-up model was also validated against experimental data. The use of such a model, generally applicable to additive manufacturing technologies, was showcased for the selective laser melting technology. Adopting this approach and performing a numerical campaign foreseeing a sensitivity analysis, a database of the properties of the manufactured components can thus be generated and the estimate of the final properties can be done by inspecting it. Considering the main findings gained, since the design phase such a computational means can effectively support designers and computer-aided engineering analysts in the identification of the combination of process parameters that guarantee the achievement of the sought properties and performance of the final product.
Additive manufacturing (AM) is considered a production technique and no longer just a rapid prototyping tool. This is due to the possibility of using different materials and creating shapes otherwise impossible when adopting more conventional techniques. One of the most popular techniques of AM is the selective laser melting (SLM): although the process starts from the powder of a material with certain characteristics, the properties found in the final product are likely to be extremely variable. Due to the intrinsic slowness and high cost, it is not always thinkable to proceed to manufacture with a trial-and-error approach. For this reason, numerical tools can be adopted as an alternative way to predict the variability induced by the process parameters (e.g. building direction, layer thickness, laser power) preventively and, thus, to avoid the possible problems in 3D-printing operations (e.g. detachment from the building platform). An important consequence of the variability of the mechanical properties is the residual stresses state: this condition is very relevant due to the high thermal gradients developed during the printing process, which are mainly caused by the different thermal conductivity of powder and solidified material.
