The potency of defects on fatigue of additively manufactured metals

Given their preponderance and propensity to initiate fatigue cracks, understanding the effect of processing defects on fatigue life is a significant step towards the wider application of additively manufactured (AM) parts. Here a novel machine learning (ML) based fatigue life prediction approach (Wu-Withers model) has been developed to relate the applied stress and the projected area, morphology, and location of the critical defects, identified post-mortem, to the fatigue life of laser powder bed fused AlSi10Mg alloy. It was found that an Extreme Gradient Boosting model was able to predict the fatigue lives with high accuracy with the importance of these characteristics in limiting fatigue life ranked in the order given above. The model was able to predict the very different lives of samples tested parallel and perpendicular to the build direction in terms of these four parameters indicating that microstructure was of minor importance. In particular the large projected area of the defects on the crack plane when testing parallel to the build direction was found to be primarily responsible for the shorter lives observed for this testing orientation. The ML model compared well with the more general two parameter (stress and projected area) Murakami model and even more closely with a four parameter (stress, area, shape and location) model for fatigue supporting their empirical dependencies.