From reduced-order modeling to scientific machine learning: How computational science is enabling the design of nextgeneration engineering systems

 

Abstract:

For the last six decades, engineering design has been advanced through the synergistic and principled use of theory, experiments and physics-based simulations. Our increased ability to sense, acquire and analyze data is clearly a game-changer. From data analytics and machine learning to digital twins, a powerful new set of engineering design tools are beginning to emerge. Yet, we must be careful not to overlook the limitations of datadriven approaches, especially since our engineering data -- even if "big" -- are almost always sparse, noisy and indirect. This talk will discuss the importance of bringing together the perspectives of data-driven learning and physics-based modeling, with a particular focus on methodologies for reduced-order modeling and predictive digital twins.

 

Biography:

Karen E. Willcox is Director of the Oden Institute for Computational Engineering and Sciences, Associate Vice President for Research, and Professor of Aerospace Engineering and Engineering Mechanics at the University of Texas at Austin. She is also External Professor at the Santa Fe Institute. At UT, she holds the W. A. “Tex” Moncrief, Jr. Chair in Simulation-Based Engineering and Sciences and the Peter O'Donnell, Jr. Centennial Chair in Computing Systems. Prior to joining the Oden Institute in 2018, she spent 17 years as a professor at the Massachusetts Institute of Technology, where she served as the founding Co-Director of the MIT Center for Computational Engineering and the Associate Head of the MIT Department of Aeronautics and Astronautics. Prior to joining the MIT faculty, she worked at Boeing Phantom Works with the Blended-Wing-Body aircraft

design group. Willcox's research has produced scalable computational methods for design of next-generation engineered systems, with a particular focus on model reduction as a way to learn principled approximations from data and on multifidelity formulations to leverage multiple sources of uncertain information. She currently has funded projects supported by the US Air Force Office of Scientific Research, Air Force Research Laboratory, ARPA-E, Department of Energy, Lockheed Martin, NASA, Sandia National Laboratories, and the Texas Higher Education Coordinating Board. Willcox currently leads several multi-institution research teams: she is Co-director of the Department of Energy AEOLUS Multifaceted Mathematics Capability Center on Advances in Experimental Design, Optimal Control, and Learning for Uncertain Complex Systems; she leads an Air Force MURI team on Machine Learning for Physics-based Systems; and she leads the Rise of the Machines team developing robust, interpretable, scalable, efficient methods for digital twins under the Department of Energy AI and Decision Support for Complex Systems program.