Researchers at the National Center for Supercomputing Applications (NCSA) and The Grainger College of Engineering, University of Illinois Urbana-Champaign (UIUC), have advanced in stress prediction research through artificial intelligence. They have used deep operator network (DeepONet) implementations in their work to enhance stress response predictions in intricate geometries like those in additive manufacturing. Using the NCSA’s Delta system, results have been achieved faster than through conventional finite element methods.
The said research was conducted via Illinois Computes, a program that provides comprehensive computing and data storage resources. This program has promoted collaborations across different disciplines, blending machine learning with computational mechanics. The Delta system, known for its exceptional GPU computing capabilities, was vital in training deep neural networks and creating training data utilizing Abaqus software.
Two notable publications have resulted from this research. The first, featured in “Computer Methods in Applied Mechanics and Engineering,” presents the innovative DeepONet utilizing a residual U-Net (ResUNet) for simplifying complex geometries. This method is a first in DeepONet architecture, showing superior memory efficiency and flexibility compared to traditional techniques.
The following article details two published papers. One is from “Engineering Applications of Artificial Intelligence” which discusses the novel version of DeepONet, named S-DeepONet. This new version takes advantage of advanced sequential learning methods and offers elevated precision in multi-physics solutions under fluctuating thermal and mechanical payloads.
Said by Iwona Jasiuk, a professor specializing in mechanical science and engineering at UIUC, “Additive manufacturing is a ground-breaking manufacturing technique offering almost unlimited applications.”
The power of DeepONet as a computational tool is in its speed and strength, allowing it to simulate the additive manufacturing process at various spatial and temporal scales. This allows for a deeper understanding of the additive manufacturing process, as well as the process of implementing and monitoring it.
This research signifies not just a significant stride in the use of AI, but also a profound impact on advanced manufacturing processes and digital twins development. The collaboration between NCSA and MechSE brings to light the powerful synergy created by blending multidisciplinary expertise and advanced technology.
Come and let us know your thoughts on our Facebook, Twitter, and LinkedIn pages, and don’t forget to sign up for our weekly additive manufacturing newsletter to get all the latest stories delivered right to your inbox.
“Why did the 3D printer go to therapy? Because it had too many layers of unresolved issues!”
0 Comments