Researchers at the University of Illinois have developed a new way to detect defects in additively manufactured parts. One of the most important jobs in any factory is determining whether manufactured parts are defect-free. Finding defects is especially difficult with additive manufacturing, because it can create parts with complex 3D shapes and significant internal features that are not easily observable.
This new technique uses deep learning to make identifying defects in additively manufactured parts much easier. To build the model, the researchers used computer simulations to generate tens of thousands of artificial defects that exist only in a computer. Each computer-generated defect varied in size, shape, and location, allowing the deep learning model to train on a range of possible defects and recognize the difference between defective and non-defective parts. The algorithm was then tested on real parts, some of which were defective, some of which were not. The algorithm was able to successfully identify hundreds of defects in real, physical parts that the deep learning model had never seen before.
“This technology addresses one of the toughest challenges in additive manufacturing.” “The new technology is a game changer,” said William King, professor of mechanical science and engineering at the University of Illinois and project leader.“Using computer simulation, we can very quickly build machine learning models that identify defects with high accuracy. Deep learning allows us to accurately detect defects that were previously undetectable by computers.”
This study Intelligent Manufacturing Journal ““Detecting and Classifying Hidden Defects in Additively Manufactured Parts Using Deep Learning and X-Ray Computed Tomography” X-ray computed tomography was used to inspect the inside of 3D components with invisible internal features and defects. 3D components are easy to create with additive manufacturing, but difficult to inspect when critical features are not visible.
More information: www.mechse.illinois.edu
Homepage Link
