Researchers at the Oak Ridge National Laboratory (ORNL), in Tennessee, USA, are developing artificial intelligence (AI) software for powder bed 3D printers that assesses the quality of parts in real time.
Powder bed printers build objects by using a laser to fuse together deposited layers of powder material. During printing, however, problems such as uneven distribution of the powder, spatters, insufficient heat, and some porosities can result in defects at the surface of each layer.
The new AI software ‘Peregrine’ is designed to identify and control such surface-visible defects by using a convolutional neural network to analyse images captured from cameras installed on the printers at high speeds.
By taking into account the composition of edges, lines, corners and textures in the images, Peregrine is able to detect any anomalies that may affect the quality of parts, and automatically alert operators so adjustments can be made in real time.
Described in the journal Additive Manufacturing, the ORNL researchers intend the software to be machine-agnostic, meaning it will be installable on any powder bed system. So far the software has been tested successfully on seven powder bed printers at ORNL, including those using electron beam melting, laser powder bed, and binder jetting technologies.
Standard cameras ranging from 4 to 20 megapixels were installed in the printers, enabling images of the print bed to be captured at each layer. The software produces a common image database that can be transferred to each new machine in order to train new neural networks quickly.
The software has been tested as part of the Transformational Challenge Reactor (TCR), a demonstration programme pursuing the world’s first additively manufactured nuclear reactor. The microreactor will be developed with newer materials in less time at a lower cost than other nuclear reactors, ensuring the future of this important carbon-free energy source.
'For TCR in particular, you could have a scenario in which the regulator will want detailed data on how a part was manufactured, and we can provide specs with the database built using Peregrine,' said Luke Scime, ORNL's principal investigator for Peregrine.
'Establishing correlations between these signatures collected during manufacturing and performance during operation will be the most data-rich and informed process for qualifying critical nuclear reactor components,' said Kurt Terrani, TCR programme director. 'The fact that it may be accomplished during manufacturing to eliminate the long and costly conventional qualification process is the other obvious benefit.'
Peregrine detects an anomaly in a component being additively manufactured on a powder bed printer. (Image: Luke Scime/ORNL, U.S Dept. of Energy)
As the monitoring system has developed, the researchers have been able to combine the captured image data with data from other sources such as the printer’s log files, the laser systems and operator notes, allowing parts to be uniquely identified and statistics from all parts tracked and evaluated.
The AI software was developed at the Manufacturing Demonstration Facility (MDF) at ORNL, a US Department of Energy user facility that works closely with industry to develop, test and refine nearly every type of modern advanced manufacturing technology. In other process control work, MDF researchers are currently developing methods to monitor for defects on the subsurface of builds and to detect porosity that may form in deeper layers, including the use of photodiodes and high-speed cameras.
Supporting the factory of the future
The Peregrine software supports the advanced manufacturing ‘digital thread’ being developed at ORNL that collects and analyses data through every step of the manufacturing process, from design to feedstock selection to the print build to material testing.
'Capturing that information creates a digital "clone" for each part, providing a trove of data from the raw material to the operational component,' said Vincent Paquit, who leads advanced manufacturing data analytics research as part of ORNL’s Imaging, Signals and Machine Learning group. ‘We then use that data to qualify the part and to inform future builds across multiple part geometries and with multiple materials, achieving new levels of automation and manufacturing quality assurance.’
The digital thread supports the factory of the future in which custom parts are conceived using computer-aided design, or CAD, and then produced by self-correcting 3D printers via an advanced communications network, with less cost, time, energy and materials compared with conventional production. The concept requires a process control method to ensure that every part rolling off printers is ready to install in essential applications like cars, airplanes, and energy facilities.