Search

Traditional Extreme Value Statistics (EVS) applied to block maxima sampled anomalies of components produced by Laser-Powder Bed Fusion may produce important inaccuracies in the estimated characteristic defects. In fact, the typical presence of multiple defect types may significantly affect the fitted maxima distributions obtained from different sampling volumes. In this work, we show how the limitations of traditional EVS can be overcome by applying supervised machine learning (ML) algorithms to classify the defects before estimating the maxima distributions for each defect type. The ML-assisted EVS provided maxima distributions unaffected by the different sampling volumes. The obtained maxima distribution lead to robust estimates of exceedance curves and finally, by employing the Shiozawa curve, to robust fatigue strength predictions.

Several factors must be considered within the assessment of parts produced by Additive Manufacturing (AM) such as, for example, heterogeneous microstructure, process-induced defects, surface quality, residual stresses, and dependence on the material’s properties with the building orientation. All these factors severely affect the resistance to static and fatigue loadings of AMed components. Among the possible failure mechanisms, the failure promoted by static loadings when cracks are present is one of the most important failure conditions, in particular for the as-built parts, which might have reduced fracture toughness and ductility compared to the wrought alloy counterpart. In this work, we present a comprehensive approach to the static assessment of AlSi10Mg parts manufactured by Laser Powder Bed Fusion. Two benchmark fracture geometries were designed to investigate the typical AM geometrical features: i) thin walls in tension and ii) notched components in bending. Finite Element analyses of the benchmark specimens showed that an approach based on elastic–plastic fracture mechanics parameters is needed to correctly predict the experimental failures, despite the quasi-brittle behaviour shown by the AlSi10Mg alloy. In view of these results, this paper explores the applicability of the Failure Assessment Diagram (FAD), a tool used for conventional ductile materials, to static assessment of AMed parts. The results show that the assessment approach based on the FAD makes it possible to properly predict the experimental failures of the benchmark specimens.