11/25/2020 National Center for Supercomputing Applications
Written by National Center for Supercomputing Applications
MechSE’s Seid Koric, as part of a team of affiliates of the National Center for Supercomputing Applications (NSCA) at UIUC, was recognized this year with another HPC Innovation Excellence Award from Hyperion Research.
Koric previously won two HPC awards, in 2018. The biannual awards recognize achievements by users of high-performance computing technologies. The program's main goals are to help others understand the benefits of adopting HPC and justify HPC investments, especially for small and medium-size businesses; to demonstrate the value of HPC to funding bodies; to expand public support for increasing HPC investments; and to showcase return on investment and scientific success stories involving HPC.
The HPC User Forum Steering Committee performs an initial ranking of submissions, after which domain and vertical experts are called on, as needed, to evaluate them.
“These awards highlight the innovative use of artificial intelligence that is transforming scientific research. They also illustrate the power of the Blue Waters supercomputer in research that benefits society,” said NCSA Director Bill Gropp.
Deep learning models in complex materials design
Researchers from NCSA's Industrial Applications Team and its Center for Artificial Intelligence Innovation (CAII) and an undergraduate collaborator from the Department of Mechanical Science and Engineering (MechSE) at the University of Illinois at Urbana-Champaign received an award for their work bringing together artificial intelligence and structural and meta-material design optimization. The team—NCSA Industrial Applications Director and MechSE Research Associate Professor Seid Koric, MechSE undergrad Hunter Kollmann, NCSA Postdoctoral Research Assistant Diab Abueidda, NCSA Research Scientist Erman Guleyruz, and CAII Affiliate Professor Nahil Sobh—used HPC and GPUs to develop cutting-edge deep learning models to predict optimized structural and meta-material designs while accounting for various nonlinear optimization scenarios thus demonstrating for the first time that it is possible to generate, machine train, and learn from modeling data on HPC. This research leverages NCSA's iForge cluster and Blue Waters supercomputer, and physics-informed AI.
“Our team is humbled and thrilled to have received this award from Hyperion,” said Koric. “Our interdisciplinary research work in the confluence of artificial intelligence and numerical modeling has opened the door for drastically accelerating future design and optimization in highly nonlinear and complex materials, composite manufacturing, materials processing, and similarly computationally demanding workflows in academic research and industry.”