Seminar: Dr. Shirin Panahi

March 20, 2025

Model Discovery, Forecasting, and Control in Complex Engineering Systems

Dr. Shirin Panahi, Arizona State University

photo: Shirin Panahi

Monday, March 24, 2025 in MECH 310
2:00-3:00 P.M.
Assistant Professor Candidate

Abstract:
The intersection of Artificial Intelligence (AI) and nonlinear dynamics has opened new frontiers in engineering, enabling breakthroughs in forecasting, control, and model discovery. This talk will explore how AI-driven techniques, including machine learning, can enhance time-series prediction, attractor reconstruction, and digital twin development. The discussed applications are supervised and unsupervised prediction of critical transitions in complex systems such as climate or power grids, and data-driven model discovery using Kolmogorov-Arnold Networks. Through case studies on AI-assisted forecasting of the Atlantic Meridional Overturning Circulation, this talk will highlight AI’s transformative role in engineering. Additionally, more theoretical research on control theory will be presented, where group theory is integrated into Model Predictive Control (MPC) to reduce computational complexity in large-scale systems. This innovation paves the way for more efficient and scalable control solutions in engineering applications. Finally, a vision for an ongoing project on physical AI is presented to address fundamental challenges in mechanical engineering, with broad applications across various engineering disciplines.

Biography:
Dr. Shirin Panahi is currently a postdoctoral research scholar in the School of Electrical, Computer, and Energy Engineering at Arizona State University. She received both her M.Sc. degree in Biomedical Engineering and her Ph.D. in Biomedical Engineering from AmirKabir University of Technology in Tehran, Iran. Before this, she earned a B.Sc. degree in Electrical Engineering from Sadjad University of Technology in Mashhad, Iran. Dr. Panahi’s research portfolio includes control theory, machine learning, and nonlinear systems.