Dec 12 (Mon) @ 2:30pm: "Control and Estimation in Network Systems," Kevin D. Smith, ECE PhD Defense

Abstract

Networks are ubiquitous in natural and engineered systems, from critical infrastructures like power grids and water distribution systems, to contact networks in epidemiological models. Managing these systems requires a broad array of tools to monitor the configuration and state of the network, identify optimal operating points, and design controls. This dissertation examines a collection of topics broadly related to this theme. In this talk, we first study the problem of safety-critical control in networks of grid-forming inverters. Coupling a physically-meaningful Lyapunov-like function with an optimization approach to identifying forward-invariant sets, we propose a method to certify that a post-fault trajectory achieves frequency synchronization while respecting safety constraints. Second, we propose a machine learning model to predict edge flows in infrastructure networks. Using an implicit neural network that incorporates two fundamental physical principles to estimate flows on unlabeled edges, we significantly outperform previous state-of-the-art models.

Bio

Kevin D. Smith is a PhD candidate in the Department of Electrical and Computer Engineering at UCSB. He joined UCSB in 2017, having been awarded an NSF IGERT traineeship to study network science, and has subsequently been a member of Francesco Bullo's group. His research interests are broadly related to dynamics, control, identification, and state estimation in network systems, in particular power grids and other infrastructure networks.

Hosted by: Prof. Francesco Bullo

Submitted by: Kevin Smith <kevinsmith@ucsb.edu>