The Department of Energy’s Oak Ridge National Laboratory (ORNL) has made a new set of additive manufacturing (3D Printing Database) data publicly available. This release aims to help both industry professionals and researchers improve the quality of 3D-printed components. The comprehensive datasets can significantly enhance efforts to verify the quality of 3D-printed parts using in-process data, eliminating the need for costly and time-consuming post-production analysis.
For over a decade, data has been collected at DOE’s Manufacturing Demonstration Facility (MDF) at ORNL. This facility has been at the forefront of early-stage research in advanced manufacturing, resulting in a wealth of information on 3D printer performance. ORNL’s extensive experience in pushing the boundaries of 3D printing with new materials, machines, and controls has enabled the creation and sharing of these valuable datasets. The latest dataset is now freely accessible online.
Traditional manufacturing benefits from centuries of quality-control practices, while additive manufacturing, a newer technique, often relies on expensive evaluation methods like destructive mechanical testing or non-destructive X-ray computed tomography. These methods, though informative, are challenging to apply to large parts. ORNL’s 3D printing datasets can be utilized to train machine learning models for improved quality assessment of various components.
“We are providing trustworthy datasets for industry to use toward certification of products,” said Vincent Paquit, head of ORNL’s Secure and Digital Manufacturing section. “This data management platform is designed to tell a complete story around an additively manufactured component. The goal is to use in-process measurements to predict the performance of the printed part.”
About the 3D Printing Datasets
The 230-gigabyte dataset includes the design, printing, and testing of five sets of parts with different geometric shapes, all made using a laser powder bed printing system. It offers access to machine health sensor data, laser scan paths, 30,000 powder bed images, and 6,300 material tensile strength tests.
This 3D Printing datasets is the fourth and most extensive in a series of AM datasets released by ORNL. Previous datasets have focused on parts made with electron beam powder bed and binder jet printing at the MDF. These datasets can be searched for specific information needed to understand rare failure mechanisms, develop online analysis software, or model material properties.
The MDF, supported by DOE’s Advanced Materials and Manufacturing Technologies Office, is a national consortium collaborating with ORNL to innovate and transform U.S. manufacturing.
ORNL researchers demonstrated the application of these datasets by training a machine learning algorithm with in-process measurements. Paired with high-performance computing methods, this algorithm can reliably predict the success of mechanical tests, reducing errors in predicting a part’s tensile strength by 61%.
Correlating in-process measurements with the final product is essential for determining when additional testing is necessary. “This is a key enabler to additive manufacturing at industry scale, because they can’t afford to characterize every piece,” Paquit said. “Using this data can help capture the link between intent, manufacturing, and outcomes.”
The data was generated as part of the Advanced Materials and Manufacturing Technology Program, funded by DOE’s Office of Nuclear Energy. These smart manufacturing approaches accelerate the development, qualification, demonstration, and deployment of advanced manufacturing technologies, enabling reliable and economical nuclear energy.
UT-Battelle manages ORNL for the Department of Energy’s Office of Science, the largest supporter of basic research in the physical sciences in the United States. For more information, please visit energy.gov/science.
