Scholars at the FAMU-FSU College of Engineering are harnessing artificial intelligence to develop innovative tools designed to detect flaws in powder-based 3D printing as part of a $2.2 million initiative backed by the Air Force Office of Scientific Research, abbreviated as AFOSR.
Additive manufacturing, commonly referred to as 3D printing, has the capability to revolutionize production and prototyping processes. The powder-based technique is a method in which powdered materials are fused to create solid functional components.
Under the guidance of Hui Wang, an associate professor in the Department of Industrial and Manufacturing Engineering, engineers from FAMU-FSU will partner with researchers from Pennsylvania State University and HP Inc. to utilize combinatorial generalization, or CG, to enhance the precision of models intended for defect prediction in 3D printing.
“Combinatorial generalization embodies a computational reasoning capability that bolsters next-generation AI by emulating human recognition processes to generate new insights from data,” Wang stated. “It’s thrilling to apply this cutting-edge technology to tackle complex manufacturing challenges.”
The focus of the project is on binder jetting, a 3D printing methodology that employs a liquid binder to adhere powdered materials together. This technique is crucial for producing high-performance components with complex designs and can achieve higher productivity levels compared to other metal printing techniques.
IMPORTANCE
Powder-based manufacturing is utilized across numerous sectors, including aerospace, automotive, healthcare, and military applications for crafting intricate metal components.
A critical issue in powder-based production is comprehending how materials can become porous, or filled with minuscule holes, a frequent defect resulting from interactions between powder spreading and powder droplets.
To address this challenge, the researchers will employ the evolving AI tool known as combinatorial generalization, which enables the AI to draw conclusions about new scenarios based on fragmented knowledge acquired from diverse data sources.
An instance of CG in humans could be a designer who takes cues from existing work to fabricate something entirely novel, or a physician who reviews a patient’s symptoms and medical history to identify patterns indicating a specific ailment, even if they have not encountered that exact situation before. This method exemplifies how existing knowledge and patterns can foster innovative solutions in additive manufacturing.
By identifying patterns and connections, the project aims to introduce groundbreaking CG tools for examining defects that are typically elusive. The anticipated advantages may extend beyond manufacturing, featuring applications in system maintenance and cybersecurity.
“This innovation has the potential to democratize manufacturing, simplifying the transition for novice manufacturers from laboratory trials to full-scale production,” Wang remarked.
COLLABORATION AND STUDENT ACHIEVEMENT
The research conducted by Wang’s team is greatly enhanced by graduate students, postdoctoral researchers, and faculty collaborators Tarik Dickens and Rebekah Downes from the Department of Industrial and Manufacturing Engineering. Jun Zeng, who heads the Physics AI engine group in the additive manufacturing R&D Lab at HP Inc., along with Penn State University Professor Jingjing Li, are also partners in this initiative.
Funding from the AFOSR will enable the researchers to acquire vital experimental materials and small-scale testing apparatus for the college. The Penn State research team is contributing essential simulations and XCT data modeling to support FAMU-FSU engineering researchers.
In collaboration with HP Inc., the initiative is set to provide crucial training and development opportunities, including access to data from HP’s 3D Printing and Digital Manufacturing Center of Excellence, advanced computing resources, and student internships. Furthermore, doctoral and undergraduate students from FAMU-FSU have participated as interns at HP.
FORWARD LOOKING
This project exemplifies the potential of interdisciplinary collaboration, merging expertise in industrial and manufacturing engineering with the innovative capabilities afforded by artificial intelligence.
“AFOSR initiatives in computational mathematics and physical sciences are actively pursuing innovative solutions in machine learning for scientific exploration, and our project aligns seamlessly with this vision,” Wang elaborated. “We anticipate that our findings will significantly enhance the Department of Defense researchers’ ability to utilize a diverse range of data sources effectively and accelerate the development of new manufacturing techniques and materials. It’s exhilarating to consider the future possibilities and the direction this research may take us.”
The article FAMU-FSU College of Engineering researchers use artificial intelligence to enhance defect detection in powder-based 3D printing first appeared on Florida State University News.