Power electronics is a core technology for future energy systems including data centers, transportation electrification, and distributed energy resources. The performance and functionality of power electronics in these systems critically impact the sustainable development of human society. To leverage the advances in semiconductor devices and the scaling laws of passive components, a promising trend is to minimize the power conversion stress and maximize the passive component utilization through granular power conversion and magnetics integration, while modernizing the modeling and control of fundamental power electronics building blocks.

In pursuit of this vision, this talk will first present the recent developments of power electronics at the chip edge, focusing on advanced power architectures for energy-demanding high-performance computing. Various design considerations for next-generation CPU voltage regulators will be presented. We will then discuss power electronics at the grid edge, exploring the possibilities of using advanced power electronics as sensors and actuators to enhance the future grid's stability and resiliency, ultimately leading to a hierarchical digital-twin model for advanced state estimation and control. Methods for designing an efficient and high control-bandwidth smart inverter for grid impedance measurement will be presented. Finally, we will introduce a large-scale open-source power magnetics research platform – the MagNet (http://mag-net.princeton.edu), and the upcoming international power magnetics modeling challenge – the MagNet Challenge – to transform the modeling and design of power magnetics with advanced data-driven methods, such as artificial intelligence and machine learning.