Data-driven and ML approaches for enabling resiliency with high DERs

The trend of more frequent power outages can be directly linked to the increase in adverse events (e.g. cyber-attack, wildfires, hurricanes). New sets of spatiotemporal information (e.g. network GIS data, cyber data, additional sensor data from smart meter/ micoPMU/ smart transformers, terrain topography, vegetation indices, weather measurements and forecast models) can be processed using data-driven machine learning (ML) approach to predict the risk of adverse impacts and enhancing situational awareness. Also, edge devices and DERs provide new possibilities and challenges for direct or indirect integration into advanced distribution management system. For decision support, proactive or corrective mitigation measures need to be explored utilizing services from legacy devices as well as edge devices. Decision support can be further enhanced by efficient data processing and NLP driven digital assistant for operators given stressful environment in the control room. This panel examines coordinated implementation of data-driven situational awareness and decision support to improve the operational resiliency. Panel will also discuss educational aspect in DER-rich environment for preparing next generation workforce related to these cyber-physical challenges.