Machine Learning In Situ Quality Control for Metal 3D Printing is being developed by ORNL and RTX.


ORNL and RTX have joined forces to create a new software solution utilizing machine learning for the quality control of metal parts produced through additive manufacturing. The collaboration between Oak Ridge National Laboratory and defense/aerospace giant RTX aims to revolutionize the industry by developing a system that can identify flaws in real-time during the printing process.

The process begins by collecting data using a near-infrared camera and an added visible-light camera during the printing process. These cameras capture detailed images of the parts being printed, which are then analyzed for any potential flaws. Additionally, CT scans are used to inspect the parts once they have been printed. These scans provide further data for analysis.

The combination of data collected from the cameras and the CT scans is used to train an algorithm to identify flaws in subsequent prints. The software is continuously improved through feedback from human operators, ensuring that it becomes more accurate over time.

One of the main achievements of this research is the quantification of the reliability of machine learning-driven quality control. By putting a number value on the level of confidence possible for flaw detection, the researchers have laid the foundation for further advancements in this field. This is crucial for the future scalability of this technique.

Luke Scime, a researcher at ORNL, stated that flaw sizes of about half a millimeter can be detected about 90% of the time. This level of accuracy is a major breakthrough in the industry and will greatly contribute to the qualification and certification of 3D printed parts.

Zackary Snow, another ORNL researcher, highlighted the importance of having a reliable and quantifiable measure of quality control in additive manufacturing. Currently, manufacturers are faced with the challenge of knowing how often to inspect parts due to the lack of a number value for flaw detection. This hinders the qualification and certification process.

The ultimate goal of this collaboration is to achieve CT-level confidence in the quality control of 3D printed parts without the need for CT scans. This would significantly reduce the time and cost involved in post-processing, making additive manufacturing more efficient and cost-effective.

Centralization and standardization of data have become crucial in advancing the additive manufacturing sector. ML-driven in situ quality control is a vital element in this process, as it has the potential to save time, money, and labor in the post-processing stage.

The collaboration between ORNL and RTX aligns with the objectives set forth by ASTM International in its Strategic Guide: Additive Manufacturing In-Situ Technology Readiness Report. The prioritization of automation in quality control for 3D printed parts will accelerate efforts to achieve comprehensive digital traceability in the industry.

By uniting stakeholders in the industry and focusing on automated quality control, there is a greater opportunity to keep counterfeit parts out of strategically important supply chains. This will enhance the overall deployment of additive manufacturing as a reliable and efficient manufacturing method.

The work carried out by ORNL and RTX demonstrates the potential for machine learning and data analysis to transform the additive manufacturing industry. Through continuous improvement and collaboration, the sector can progress towards widespread commercialization and achieve new levels of quality control and certification.

Original source


“Why did the 3D printer go to therapy? Because it had too many layers of unresolved issues!”

Like it? Share with your friends!


Meet the mastermind behind GCode-Guru, a 3D printing wizard whose filament collection rivals their sock drawer. Here to demystify 3D tech with a mix of expert advice, epic fails, and espresso-fueled rants. If you've ever wondered how to print your way out of a paper bag (or into a new coffee cup), you're in the right place. Dive into the world of 3D printing with us—where the only thing more abundant than our prints is our sarcasm.


Your email address will not be published. Required fields are marked *

Choose A Format
Personality quiz
Series of questions that intends to reveal something about the personality
Trivia quiz
Series of questions with right and wrong answers that intends to check knowledge
Voting to make decisions or determine opinions
Formatted Text with Embeds and Visuals
The Classic Internet Listicles
The Classic Internet Countdowns
Open List
Submit your own item and vote up for the best submission
Ranked List
Upvote or downvote to decide the best list item
Upload your own images to make custom memes
Youtube and Vimeo Embeds
Soundcloud or Mixcloud Embeds
Photo or GIF
GIF format