New Algorithm Slashes Inspection Times for Nuclear 3D-Printed Parts
The Oak Ridge National Laboratory (ORNL) has unveiled a breakthrough in the field of nuclear materials inspection with a cutting-edge software algorithm designed to significantly reduce the inspection times for 3D-printed components. This development comes in response to the growing trend within the nuclear industry to utilise additive manufacturing, commonly known as 3D printing, to fabricate intricate components rapidly and domestically.
The new machine-learning-based algorithm devised by ORNL promises to revolutionise the inspection process by reducing the duration required for the inspection of 3D-printed parts used in nuclear applications by as much as 85 per cent. Traditionally, quality assurance for these components involves the use of computed tomography (CT) scans, which employ X-rays to capture comprehensive images of the internal structures to detect any possible defects. The introduction of ORNL’s algorithm significantly enhances the efficiency of these CT scans by enabling accelerated reconstruction and analysis of the images.
This innovative technology is now being adapted for use at the Idaho National Laboratory (INL), where it is set to inspect irradiated materials and nuclear fuel. Researchers at INL highlight the importance of this advancement, noting its potential to improve nuclear safety and efficiency significantly. During a demonstration, INL researchers successfully analysed over 30 3D-printed samples in less than five hours using the algorithm. This is a drastic improvement compared to the older method, which would have taken more than 30 hours for the same task without the new software.
In nuclear research environments like INL, the examination of materials from nuclear reactors is often delayed to safeguard laboratory technicians from radiation exposure during prolonged CT scans. These extended exposure periods can not only compromise the safety of laboratory personnel but also degrade the CT detector, affecting its durability and image precision. By decreasing scan times, the new ORNL algorithm effectively reduces radiation exposure for both staff and equipment, yielding higher-quality data and improving the feedback loop for performance models.
Bill Chuirazzi, the instrument scientist leading INL’s Diffraction and Imaging group, expressed optimism about the algorithm's future applications. He stated, "If we use this algorithm to reduce the scan time for radioactive materials and fuels, it will increase worker safety and the rate at which we can evaluate new materials." Chuirazzi further noted that the advancements could expedite the entire life cycle of innovative nuclear concepts, from their conception to implementation in the power grid.
The endeavour represents a collaborative effort between ORNL and INL, both of which play a significant role in forwarding the development and deployment of new reactor designs. These advancements are seen as crucial steps towards decarbonising the energy sector.
The algorithm was initially crafted by ORNL researcher Amir Ziabari, targeting the optimisation of scans for 3D-printed metal parts. It is currently being fine-tuned for applications concerning radioactive materials and nuclear fuels. This developmental effort receives funding from the US Department of Energy’s Advanced Materials and Manufacturing Technologies (AMMT) programme, which is dedicated to accelerating the commercialisation of breakthrough materials and manufacturing technologies.
Source: Noah Wire Services