Cellular scanning strategy for selective laser melting: evolution of optimal grid-based scanning path and parametric approach to thermal homogeneity

Selective laser melting, as a rapid manufacturing technology, is uniquely poised to enforce a paradigm shift in the manufacturing industry by eliminating the gap between job- and batch-production techniques. Products from this process, however, tend to show an increased amount of defects such as distortions, residual stresses and cracks; primarily attributed to the high temperatures and temperature gradients occurring during the process. A unit cell approach towards the building of a standard sample, based on literature, has been investigated in the present work. A pseudo-analytical model has been developed and validated using thermal distributions obtained using different existing scanning strategies. Several existing standard and non-standard scanning methods have been evaluated and compared using the empirical model as well as a 3D-thermal finite element model. Finally, a new grid-based scan strategy has been developed for processing the standard sample, one unit cell at a time, using genetic algorithms, with an objective of reducing thermal asymmetries.