Microstructure Evolution, the Influence on Material Properties, and the Optimized Processing Parameters Search in Additive Manufacturing

Additive manufacturing (AM) still needs to address many challenges due to its multi-physical processes in various materials systems or multiple situations that need to be applied before implementation in more fields and replacing more places of traditional manufacturing. Specifically, the primary aim of this investigation is to study the microstructural changes that affect material properties, such as elastic modulus and Poisson's ratio, and then how to search for the optimized processing parameters. These changes affect the material's performance, including residual stress, fractures, etc. To achieve this, the characterization of the microstructure of materials, mainly the surface/textures, grain size, and defects, if necessary, is of great importance. The texture and grain size simulation for multi-phase materials systems based on accurate physical stimuli modeling of processing is conducted. The influence of microstructural evolution on material properties such as elastic modulus strength is measured. Currently, although the process-structure-property prediction relationship has been built with analytical models, there’s still no simulation approach established for searching for optimized processing parameters, which is, in fact, the realistic goal of the industry. In this work, the inverse method is utilized for analytically finding the optimized processing parameters with some general machine learning model. The time-consuming for computational cost has been enhanced further. Several paradigms are constructed to optimize manufacturing processes, utilizing combined advantageous analytical and data/machine learning or semi-analytical frameworks. This study bridges the gap between micro and macrostructures and the properties of materials in AM, potentially revolutionizing the industry and inspiring a new philosophy for science.