EUROPEAN CONGRESS AND EXHIBITION ON ADVANCED MATERIALS AND PROCESSES - EUROMAT 2021

Recent years have marked an increased interest in Additive Manufacturing (AM) processes and their applications. This manufacturing route presents advantages over conventional ones as it allows the production of geometrically complex parts in almost their net shape. This minimizes the required post-processing and thus raises opportunities of decreased material and time costs. However, it remains somewhat restricted due to a limited number of alloys that are adapted to AM-processes— or ‘printable alloys’. The aim of this study is to propose a computational method based on bayesian machine learning combined with a thermodynamic approach (CALPHAD) integrated in a multi-objective genetic algorithm to design AM-optimized alloys. In this context, several material characteristics influencing the defects commonly observed in AM-fabricated parts, such as solidification cracking, porosity, balling, residual stresses and distortions, were taken into account along with final strength. Specifically, criteria regarding phase stability, solidification conditions and several thermal and surface properties were defined. The proposed models link material composition to various properties by using data sets constructed from published literature and industrial material datasheet. Its application to design improved grades of austenitic stainless steels will be shown and discussed.