A Framework for Certifying Additive Manufacturing
Thermal, spatial and temporal aspects of additive manufacturing (AM) are extremely complex and geometrically dependent. AM technologies typically result in non-homogeneous microstructures and non-uniform material properties. Capabilities must be developed that can rapidly expand the methods in which information about AM materials is captured, analyzed and used, ensuring the perfect born-certified part every single time.
Oak Ridge National Laboratory is developing new characterization technologies through data analytics that are capable of rapidly extracting information from materials. Using advanced Integrated Computation Materials Engineering (ICME) approaches, the new characterization technologies advance the understanding of AM focusing on process certification rather than the current individual part certification or part family certification.
Within the next few years, ORNL will produce a collaborative, open source and accessible database for materials science mining based on machine learning, algorithms, statistics and modeling.
Open source software that ORNL is creating characterization tools for, such as Dream3D, are expected to drive the next materials revolution, creating a framework for coupling data analytics with advanced manufacturing. The software platform features: data management and tracking; signal processing; computer vision and image processing; n-D data visualization; modeling and simulation; data analytics and machine learning; process optimization; and certification, verification and validation.