Scientific machine learning (SciML) has emerged as a promising field for data-driven modeling of complex fluid flows, particularly those encountered in advanced power generation and propulsion concepts. However, these (and other) aerospace applications impose stringent requirements on SciML models, necessitating solutions that address both complex physics and geometry. To realize the vision of data-driven modeling in aerospace applications, SciML approaches must be (a) physically interpretable, (b) inherently compatible with complex data representations, and (c) scalable.

Shivam Barwey,
Argonne National Laboratory
In this talk, I discuss how SciML models can be designed to meet these requirements and provide some perspective on how hardware-oriented machine learning can enable transformative advancements in fluid simulation capabilities for high-speed power and propulsion applications.
Shivam Barwey is the Argonne Energy Technology & Security (AETS) Named Fellow at Argonne National Laboratory. He received his Ph.D. in aerospace engineering at the University of Michigan. His research interests lie at the intersection of machine learning, computational fluid dynamics, and high-performance computing, with a focus on the numerical modeling and simulation of high-speed reacting flows relevant to power and propulsion applications.