- This event has passed.
LANS Seminar
Seminar Title: What We are Building in GridAI: GridMind, LUMINA, and APPFL
Speaker: Kibaek Kim, Computational Mathematician, Laboratory for Applied Mathematics, Numerical Software, and Statistics, Mathematics and Computer Science Division, Argonne National Laboratory, and a Senior Scientist at-Large, University of Chicago Consortium for Advanced Science and Engineering
Date: Thursday, March 12, 2026
Time: 2:30 PM-3:30 PM (In-Person)
Location: Hybrid, Bldg. 240, Conference Room 4301
Description: GridAI is one of the DOE Genesis Mission seed Model Team projects, with a focus on advancing the electric grid operations and planning with AI. This seminar gives a technical, work-in-progress view of what we are building at Argonne for GridAI, organized around a concrete use case: contingency analysis under a hurricane event on a synthetic Texas system. I will briefly introduce GridMind as the agentic workflow component that coordinates tools, models, and simulations, then focus mainly on LUMINA: the foundation-model training and inference pipeline, what we observe empirically in experiments, and how we integrate surrogate inference with selective calls to high-fidelity solvers when needed. I will also touch on APPFL and recent results that enable distributed training/adaptation across HPC facilities. I will close with a forward-looking vision for GridAI with these capabilities: scalable, AI-enabled grid analysis that moves from fragmented studies to proactive risk identification and end-to-end decision support in collaboration with other labs and partners.
Bio: I am a Computational Mathematician in the Laboratory for Applied Mathematics, Numerical Software, and Statistics within the Mathematics and Computer Science Division at Argonne National Laboratory, and a Senior Scientist at-Large at the University of Chicago Consortium for Advanced Science and Engineering. My research focuses on federated learning algorithms and open-source software, agentic workflow orchestration for large-scale scientific and engineering studies, and modeling and numerical algorithms for optimization on high-performance computing systems and GPUs. My work is applied to electric power system operations and planning (e.g., large-scale contingency analysis and fast surrogate screening), healthcare, and other scientific domains of interest to the U.S. Department of Energy. Before joining Argonne, I obtained a Ph.D. degree in Industrial Engineering and Management Sciences from Northwestern University. I am a recipient of the DOE Early Career Research Program award. I serve as an associate editor for Mathematical Programming Computation and Naval Research Logistics, and as a board member for the COIN-OR Foundation and IISE Energy Systems.
Please note that the meeting URL for this event can be seen on the cels-seminars website which requires an Argonne login.
See all upcoming talks at https://www.anl.gov/mcs/lans-seminars