Machine learning for mechanical properties classification in additive manufacturing

In the present paper, the effectiveness of the use of Machine Learning techniques, in particular Deep Learning algorithms, in the analysis of Ti-6Al-4V (Ti64) manufacture is studied; relationships between the values of physical parameters used during the production and mechanical characteristics are defined by means of the analysis of images taken from sections of the specimens, where defects and microstructural discontinuities can be observed. The Deep Learning approach, widely used for image classification and features extraction, also in this case shows promising possibilities, as proved by the implementation results reported.