Join us in welcoming Wentao Tang, assistant professor of Chemical Engineering at NC State, as he discusses dissipative properties for control. Alums and friends of the program are always welcome.
https://ncsu.zoom.us/rec/share/ecWLi8mYzUZxWejM624xUWQv7zNIJNWt4b9_wy1bFoL_YEz6kXx4SZjmKAcuUt-h.qnbu3mqcJJYyu6p7
Access Password: z&H#W56!
Learning of Dissipative Properties for Control
While machine learning techniques have been increasingly used for the control of nonlinear systems, discovering or preserving the structured properties underlying the nonlinearity is an issue of fundamental importance. From a thermodynamics point of view, conservation and dissipation are the universal laws governing engineering systems, and the learning of such properties from data not only facilitates the design of controllers with guaranteed performance but also provides physical insights and interpretability. In this talk, I will introduce the dissipative systems theory and its origin from irreversible thermodynamics, as well as its connections to convex optimization and dynamic optimization. I will also present some formulations of dissipativity learning for nonlinear systems, with elementary theoretical analysis and applications to reaction and separation processes, from my recent research.
Wentao Tang is an Assistant Professor in Chemical and Biomolecular Engineering at NC State University. He received his B.S. in Chemical Engineering and B.S. in Mathematics and Applied Mathematics from Tsinghua University in 2015, and Ph.D. in Chemical Engineering from University of Minnesota in 2020. He was a process control engineer at Shell from 2020 to 2022 prior to joining NC State. His research focuses on developing data-driven model-free control methods for problems in nonlinear control theory, the analysis of optimization algorithms as dynamical systems, as well as potential applications in chemical processes.
