Algorithms can correct 3D printing mistakes for different materials, parts and systems

Engineers have developed intelligent 3D printers which can detect and correct mistakes, even if they are in unfamiliar designs or materials such as ketchup or mayonnaise. They do this by learning from other machines’ experiences.

The University of Cambridge engineers developed a machine-learning algorithm that can detect and fix a variety of errors in real time. It can be added easily to existing or new machines to improve their capabilities. The algorithm can also teach 3D printers to print new materials on their own. Nature Communications has published details of the low-cost method.

The 3D printing technology has the ability to revolutionize production, allowing for the creation of customized and complex parts such as medical implants or intricate sweets. It could also change the manufacturing supply chain. It is vulnerable to errors in production, from minor mechanical flaws to complete failures.

Source:
https://techxplore.com/news/2022-08-algorithm-3d-errors-materials.html