Ti-6Al-4V LASER POWDER BED FUSION PROCESS ENGINEERING

In recent years, additive manufacturing (AM) has taken increasing interest in many industrial sectors such as aerospace and biomedical thanks to its freedom of design. Thus allowing the production of complex and highly customized parts. Moreover, AM leads to a new way of thinking during the design phase, making possible new design strategies such as part consolidation and topology optimization to increase the functionality of the part. Among all the AM technologies, laser powder bed fusion (L-PBF) is one of the most suitable to produce metallic parts. In fact, L-PBF is widely adopted in industries for the fabrication of Ti-6Al-4V parts which usually find applications in the aerospace and biomedical fields. This is due to the high strength/weight ratio and corrosion resistance of the Ti-6Al-4V alloy. Despite all these advantages, parts produced with L-PBF can result in poor surface quality or mechanical properties. Thus, the Ti-6Al-4V L-PBF process engineering is of fundamental importance.In this dissertation, challenges for the fabrication of parts with the L-PBF process such as lack of fusion, gas pore, residual stress, and part distortion are presented. Following, the effect of process parameters on mechanical properties and part quality has been analyzed. Finally, prediction tools such as numerical simulations and machine learning are shown for the Ti-6Al-4V L-PBF process engineering. In order to investigate the influence of process parameters on L-PBF parts, mechanical tests were carried out together with microstructure analysis. Thus allowing a better understanding of the printing process. Moreover, reverse engineering techniques such as 3D acquisition systems were of fundamental importance to validate the results obtained from the numerical simulation. Overall, the results obtained from this thesis work suggest that: (i) build orientation is a key parameter when designing the L-PBF process due to its influence on thermal exchanges; (ii) line energy density alone is not enough to predict the success of the printing process, especially in the case of the warping phenomenon; (iii) machine learning can be an effective tool to predict the quality of the printed part when several process parameters have to be taken into account; (iv) numerical simulation can be used to study the development of residual stress during the L-PBF process and predict the part distortion.