(Nanowerk News) In additive manufacturing, also known as 3D printing, a 3D printer deposits material onto a surface or mold, letting it solidify layer by layer until you get a 3D object – this simple as a cube made of plastic or as complex as a metal jet engine component – takes shape.
Unlike subtractive manufacturing, which involves carving an object out of a larger piece of material, additive manufacturing only uses the amount of material needed to create a piece. This cuts down on waste and costs, making it particularly attractive for prototyping purposes. It also allows users to easily create parts with complex shapes.
Precision and quality are essential in additive manufacturing, especially when used to create heat-resistant metal parts for applications such as jet engines, rockets, or other high temperature environments. A team of two engineers from the University of Arizona are using machine learning methods and $ 750,000 in funding from NASA to monitor and mitigate defects that arise in additive manufacturing. Mohammed Shafae of Systems Engineering and Industry (link is external) and Andrew Wessman of Materials Science and Engineering (link is external) collaborate with Lockheed Martin Space, Open Additive LLC and CompuTherm LLC.
“Andrew and Mohammed are using their combined experiences in materials science and systems engineering to examine additive manufacturing from the microscopic level to the large-scale systems level,” said David W. Hahn, Craig M. Berge Dean of the College engineering. “Cutting-edge manufacturing is one of the research areas of the college, and it is a prime example of an interdisciplinary effort to advance the field and keep AU at the forefront. ”
Different faults, different problems
There are two broad categories of defects that can occur in additively manufactured products.
Process defects are physically visible aberrations that occur when something goes wrong in the printing process. For example, two coats may not adhere properly, or there may be a hole or crack in the material.
Material defects are variations in the chemistry or arrangement of atoms that are only visible with high resolution microscopes. The complexity of many additively manufactured parts can make it difficult to detect these defects using common inspection methods. Material defects can occur if one layer cools further and another hot layer is placed on it. The temperature of the first layer could increase and changing the cooling process could alter the properties of the part. For example, the metal might become brittle or less able to withstand stresses.
“You can think of how dangerous it would be if the part was used in a jet engine or a rocket.” Shafae said. “The types of faults that we are focusing on are faults that will cause the material to behave differently than expected. ”
Machine learning and the fourth industrial revolution
Shafae and Wessman will use a sophisticated sensor system, combined with thermal cameras and high-speed localized cameras, to monitor the 3D printing process and identify when and where faults occur. They will apply machine learning methods to the data and develop a model that can predict faults when they occur. This will allow scientists to take corrective action to avoid faults or terminate a process before wasting more time and materials. Research in this area typically uses a single type of sensor to detect specific categories of defects, but this work takes the concept a step further.
“We will really try to learn how these separate categories of faults can relate to each other, because sometimes process faults can be the main cause of hardware faults,” Shafae said.
The machine learning element of this project is essential: A product is only as strong as its weakest point, and industrial-scale additive manufacturing processes often generate several terabytes of data that no researcher could. to sourt out. Removing this step from human hands allows for a much more in-depth analysis of the process.
This combination of augmented intelligence with the merging of the digital, physical and biological worlds forms the cornerstone of the University of Arizona’s strategic plan. The combination of data processing, process optimization, materials analysis and machine learning makes it an excellent representation of the Fourth Industrial Revolution.
“This is really an example of what people need to do to access Industry 4.0, which is basically using data to improve processes and make sure they work the way you want them to,” said Wessman. . “By improving the quality of the manufacturing process, you can know what you have is a good product the moment you take it off the machine. ”