Abstract

Deep penetration of distributed energy resources (DERs) calls for improved monitoring of power systems against reliability. For example, the low observability in some distribution grids makes monitoring DERs hard due to limited investment and vast coverage of distribution grids. Past methods proposed machine learning models with limited explainability. However, critical energy infrastructure needs assurance. For such a need, this talk shows how to design assured machine learning for power system monitoring via a twin structure of two learning agents, namely the AI-based and physics-guided models. The twins will collaborate adaptively to minimize the learning error while maximizing physical consistency. The structure ensures good generalization properties of the learned models. Finally, we will show how the ideas grow into several Department of Energy (DOE) projects, showing collaborations between industry and academic members. The talk will end with open future problems in distribution grids.