Researchers at Oak Ridge National Laboratory (ORNL) have enhanced the flaw detection process in laser powder bed fusion 3D printing, boosting confidence in metal parts produced through this method. This advancement addresses challenges faced by industries like aerospace, defense, and energy when adopting this technology due to difficulties in inspecting printed parts for deep-seated flaws.
ORNL’s new approach merges post-build inspection data with real-time sensor data collected during printing. This combined data trains a machine-learning algorithm to pinpoint flaws. The method consistently identifies flaws approximately half a millimeter in size with a 90% success rate. This in-process flaw detection matches the reliability of traditional, more labor-intensive methods.
Laser powder bed fusion, a prevalent metal 3D printing technique, relies on a high-energy laser to melt metal powder layer by layer, constructing the desired object. Although flaws in materials are anticipated, ORNL’s method offers a more quantified approach to flaw detection in 3D printing, addressing a major obstacle in the industry.
The research, in collaboration with aerospace and defense company RTX, employed CT scans and near-infrared cameras to monitor the printing process, providing critical data to the machine-learning algorithm. Over time, with ongoing training, the algorithm’s accuracy improves, reducing human involvement in inspection.
This ORNL-led breakthrough holds potential for mass production applications, enabling more diverse 3D printed parts and ensuring quality control. As the industry trends towards larger print sizes and faster rates, this method can address inspection challenges for larger, complex parts.
You can read more about the project in the research paper titled “Scalable in situ non-destructive evaluation of additively manufactured components using process monitoring, sensor fusion, and machine learning”, in the Additive Manufacturing journal, at this link.
Come and let us know your thoughts on our Facebook, Twitter, and LinkedIn pages, and don’t forget to sign up for our weekly additive manufacturing newsletter to get all the latest stories delivered right to your inbox.
