Issue 53

P. Ferro et alii, Frattura ed Integrità Strutturale, 53 (2020) 252-284; DOI: 10.3221/IGF-ESIS.53.21 255 Figure 3: Porosity percentage (%) as a function of process parameters (laser power, exposure time) and iso-volumetric energy density (Ed) curves. Material = AlSi10Mg alloy; h = 80 μ m; d = 30 μ m [14]. A second experimental strategy for process parameters optimization is the in-situ monitoring and in-line quality control of the process itself [3,18-20]. However, they are still challenging to implement in actual machines for industrial production. In the last years, the simulation of AM processes emerged as a powerful tool that allows performing virtual experiments without the need of raw materials and machines [21-24], sample preparation and extensive characterization methods. For this reason, it might be used to overcome the drawbacks coming from experimental techniques but also to support them by reducing the number of tests necessary for process parameters optimization [25,26]. Finally, it provides a quick feedback loop to designers. It sounds good, if only it were so simple. In fact, modeling all phenomena involved in PBFPs is not an easy task. Analytical models, due to their high time efficiency, may be useful for parametric analyses but are strongly limited to the prediction of thermal fields in very simple geometries. Numerical modeling based on FEA (finite element analysis) or FDA (finite difference analysis) are instead promising methods able to fulfill the need of additively manufactured parts producers, provided that the usability and the computational time of such tools be adapted to industrial applications and not limited to research only. Among the benefits of process simulation, time and cost saving in design of new components, alloy development as well as topology optimization are certainly the most interesting [27,28]. Numerical models were developed in literature with different length scale simulation, microstructural and mesoscale modeling aimed at predicting grain growth and defects [29,30] and macro scale modeling aimed at predicting residual stress and distortion of the as-built parts [31,32]. This contribution is aimed at reviewing different PBFPs simulation works found in literature with the main goal to understanding the influence of process parameters on the fabricated material. The complexity of the process requires a multi-scale modelling approach. Since there is not a common definition in literature of the modeling scale, here we define powder-scale models all those models that simulate the power particles behaviour and the microstructure of one track. In general, stress and strains are not calculated in such ‘micromodels’, rather, they are used to investigate the effect of powder density and process parameters on microstructure and defects induced by PBFPs. Layer-scale models are those models that are able to capture the thermal and stress fields of one or few layers and may be useful to optimize the scanning strategy or calculate the inherent strain to transfer to the full-scale models. In this case the powder layer is treated as a homogeneous porous media. Finally, full scale models are developed to predict distortions and residual stresses of the real component (Fig. 4). In the following, the paper will be divided into three main sections. The first one is devoted to summarize the major outcomes deriving from powder-scale models; the second one is aimed at describing the layer-scale models and results. Finally, the third part collects the most important numerical strategies adopted to face the computation of full-scale models and summarizes their main outcomes.