According to the US Ministry of Energy (US -Doe), researchers have used diagnostic tools and machine learning to develop a new method for recognizing and predicting defects in 3D printing materials. The method uses various imaging and machine learning techniques to identify the generation of pores in real time with almost perfect accuracy. The researchers will soon develop detection technologies that can recognize other types of defects that occur during the additive manufacturing process with the aim of creating a system that not only captures defects, but also enables repairs during the AM.
Many industries rely on metal additive production in order to quickly build complex parts and components-to build everything from rocket motifs to pistons for high-performance cars and tailor-made orthopedic implants. New advanced diagnostic tools for recognizing and potentially repairing defects will expand the use of AM in aerospace and other industries that rely on high-performance metal parts.
One of the main errors in laser powder fusion is the formation of keyhole pores. These pores or structural defects can affect the performance of the printed parts. Many 3D printing machines have thermal imaging sensors that monitor what is built up, but can overlook the formation of pores. The only way to grasp pores directly in the dense metal is to use highly intensive beams such as those of Advanced Photon Source, a Doe Office of Science User Facility.
Here the researchers correlated the X-ray images of the rehearsal interior and the thermal images of the melting pool and found that the formation of a keyhole pore on the surface of the material generates a different signal that can be recognized by thermal cameras. First they trained a machine learning model with X -ray images to predict the formation of pores with only thermal images. Then they tested the ability of the model to decipher the complex thermal signals and predict pore production in non -marked samples. The researchers found that the approach could recognize the exact moment in which a pore formed during the pressure process on time scales of less than milliseconds.