Physics-aware and Risk-aware Machine Learning for Power System Operations (Slides)

Recent years have witnessed rapid transformations of contemporary advances in machine learning (ML) and data science to aid the transition of energy systems into a truly sustainable, resilient, and distributed infrastructure. A blind application of the latest-and-greatest ML algorithms to solve stylized grid operation problems, however, may fail to recognize the underlying physics models or safety constraint requirements. This talk will introduce three examples of bridging physics- and risk-aware ML advances into efficient and reliable grid operations. First, we develop a topology-aware approach using graph neural networks (GNNs) to predict the price and line congestion as the outputs of real-time optimal power flow problem. Building upon the underlying relation between prices and topology, this proposed solution significantly reduces the model complexity of existing end-to-end ML methods while efficiently adapting to varying grid topology. Second, we put forth a risk-aware ML method to ensure the safety guarantees of data-driven, scalable reactive power dispatch policies in distribution grids. The resultant policies can directly account for the statistical risks on prediction error to attain guaranteed voltage violation performance. Last, we consider a reinforcement learning framework for managing a large number of dynamical, flexible energy resources such as electrical vehicles, and demonstrate the need to simplify the system representation through physics-aware state/action aggregation.