Webinar Abstract

Artificial intelligence is widely expected to transform power electronics design and condition monitoring, yet most existing approaches still depend on large, carefully labelled datasets that are expensive or impossible to obtain in practice. Converter design requires extensive simulations and experiments, while degradation and fault data are naturally scarce, imbalanced, and strongly affected by changing operating conditions. This webinar presents two complementary approaches to overcome these barriers: an active-learning framework that identifies the most informative samples and reduces the data required for reliability-oriented converter design, and a domain-specific time-series foundation model that learns from large amounts of unlabelled operational data to improve condition monitoring under limited labels and unseen operating conditions. Together, these studies show how AI can move beyond conventional data-driven modelling toward more efficient design, stronger generalization, and more reliable power electronic systems.

Speaker Bio

Yi Zhang is an Assistant Professor in the Department of Electrical and Electronic Engineering at The Hong Kong Polytechnic University. His research focuses on reliable and intelligent power electronics, including condition monitoring, wide bandgap power devices and packaging, thermal and reliability modelling, and artificial intelligence for power electronics. He received his Ph.D. degree from Aalborg University, Denmark, and was awarded an International Postdoctoral Fellowship from the Danish research council (DFF) in collaboration with Aalborg University and MIT. He has conducted research at EPFL, RWTH Aachen University, and Georgia Tech through postdoctoral or visiting appointments. He received the IEEE Power Electronics Society Ph.D. Thesis Talk Award in 2020, the 1st (2021) and the 2nd (2025) Prize Paper Award from IEEE Transactions on Power Electronics, the 2nd Prize Paper Award from IEEE Transactions on Industry Applications in 2025. He currently serves as an Associate Editor for IEEE Transactions on Power Electronics and editorial roles for several journals in the power electronics field.

Register here: https://ieee.webex.com/weblink/register/r78052e2b09b659d93ad5405db28c1375