ORNL Targets Faster, Safer Inspection for 3D Printing in Nuclear Applications

For years, the United States’ Oak Ridge National Laboratory (ORNL) has been one of the most active players in additive manufacturing for nuclear energy. In 2020, the lab made headlines for designing the first prototype of a 3D-printed nuclear reactor. Since then, they have printed nuclear reactor components, and in 2022, their researchers created a deep-learning framework that allows them to inspect these additively manufactured parts with greater speed and accuracy. In partnership with Idaho National Laboratory (INL), the two labs accelerated the inspection of 3D-printed nuclear components, and now, they are expanding to inspect nuclear fuels.

Critical to this research is this deep-learning framework created by ORNL. The software algorithm, called Simurgh, checks for flaws in AM parts, which helps them identify metals for 3D printing the next generation of nuclear reactors. It can take decades to verify new materials and manufacturing methods for nuclear reactor components, so the collaboration will accelerate the process.

A 3D-printed part is scanned and analyzed using the ORNL software. (Photo credits: ORNL)

Simurgh uses X-ray computed tomography, which is a type of CT scan, to check the interior quality of 3D printed objects without damaging them. By compiling a series of X-ray images, the internal structure of a 3D-printed part is revealed, identifying weaknesses or printing errors. Typically, scanning the same part from many angles is time-consuming and expensive.