US labs collaborate to speed up materials inspections
A new algorithm has reduced the time needed to inspect 3D-printed parts for nuclear applications by 85% and could lead the way to faster, safer inspections of irradiated materials and nuclear fuel, speeding up the development and deployment of new nuclear technology.
_75348.jpg)
The quality of complex nuclear parts made by additive manufacturing - 3D printing - techniques is typically verified through computed tomography (CT) scans, which use X-rays to capture images and detect weaknesses or errors in the internal structure. Oak Ridge National Laboratory (ORNL) developed the algorithm - which it describes as a deep-learning framework - to produce faster, more accurate scans of 3D-printed metal parts by using machine-learning to rapidly reconstruct and analyse the images.
This cuts down the cost, time, and number of scans needed to perform an inspection, reducing costs significantly, according to ORNL lead researcher Amir Ziabari. "And the quality is higher, so the post-processing analysis becomes much simpler," he said.
Shorter scans would also mean less radiation dosage per scan, improving safety for technicians as well as lessening wear on detectors from radiation during long X-ray CT scans.
Researchers at Idaho National Laboratory (INL) applied the algorithm to analyse more than 30 3D-printed sample parts in less than 5 hours of scan time - it would have taken more than 30 hours to complete each scan without the software. The software is now being "trained" so it can be used with radioactive materials and fuels.
"If we use this algorithm to reduce the scan time for radioactive materials and fuels, it will increase worker safety and the rate we can evaluate new materials," said Bill Chuirazzi, an instrument scientist and leader of INL’s Diffraction and Imaging group.
The software, which is being supported through the US Department of Energy's Advanced Materials and Manufacturing Technologies programme, could be applied in many fields including defence, auto manufacturing, aerospace and electronics printing, as well as non-destructive evaluation of electric vehicle batteries, ORNL said. The framework is already being incorporated into software used by commercial partner ZEISS within its machines at the Department of Energy's Manufacturing Demonstration Facility at ORNL.
_58601.jpg)
