Project 1: Digital Twin
Digital Twin technology provides an integrated platform with the “cyber” and “physical” parts to replicate a real engineering complex system, which provides a perfect platform for the research development for electrical grids. In this project, the team will focus on the research of a digital twin system, called Autonomous Network Guardian for Electrical systems (ANGEL), with the integration of power grid physics, modeling and simulation, scenario design, performance analysis and visualization. ANGEL will be able to analyze and emulate complex grid behaviors under different conditions. This project will also integrate the latest machine learning and data mining techniques to be able to analyze the grid behaviors and responses, and therefore improving grid reliability and resiliency.
Project 2: Early Warning System
Cyber-physical events in the smart grid result in a vast variability of grid behaviors, which are challenging in the identification of causes and the evaluation of risks. In this project, the team will focus on the research of an early warning system that is capable of analyzing the spatial-temporal complexity and functioning with a short response timeframe. This must rely on a thorough understanding of both grid behaviors and potential threats. In the past decade, we have witnessed multiple major blackouts resulting from cascading failures triggered by only a small number of contingencies or sabotages. Based on the comprehensive analysis of the electrical grid characteristics, the team aims to establish a proactive early warning system for the electrical power grid wiht the following major research components: analyzing spatial-temporal grid behaviors attributable to massive blackouts; profiling prominent coordinated attack threats targeting the cyber-physical vulnerabilities; and developing data-driven methods with machine learning for early warning detection and grid behavior analysis.
