Topological Machine Learning Methods for Power System Responses to Contingencies

While deep learning tools, coupled with the emerging machinery of topological data analysis, are proven to deliver various performance gains in a broad range applications, from image classification biosurveillance blockchain fraud detection, their utility areas high societal importance such as power system modeling and, particularly, resilience quantification energy sector yet remains untapped. To provide fast acting synthetic regulation and contingency reserve services grid while having minimal disruptions on customer quality service, we propose new topology-based that depends neural network architecture for impact metric prediction systems. This novel allows one evaluate three types, conjunction transmission lines, transformers, lines combined transformers. We show proposed equipped local measures facilitates more accurate unserved load well amount load. In addition, able learn about complex relationships between electrical properties measurements simulated response contingencies NREL-SIIP system.