Grant: Discovery Grants Program - Individual

As the global climate changes, the frequency and severity of extreme weather events are increasing. These events cause prolonged power interruptions, depriving the electricity for critical infrastructure providing healthcare, security, transportation, water supply, and more. Prolonged power outages have incurred billions of dollars in economic costs and in certain instances, have claimed human lives. It is unrealistic to make electricity grids invulnerable to high-impact low-probability events such as severe weather events, cyber-physical attacks, geomagnetic storms, and human errors. However, electricity grids can be made more resilient by increasing their ability to withstand disruptions and reduce the magnitude and duration of outages. For this purpose, fortification of network infrastructure is an essential but expensive first step conducted as part of grid planning. Enhancing resilience also involves preventive response hours ahead of an event, emergency response during the event, and rapid restoration after the event. The long-term goal of this research is to develop smart tools and techniques to support these latter tasks. When an event can be forecasted, proactive operational actions can be taken to reduce the impacts and the risk of complete failure. These actions involve optimally scheduling controllable resources such as energy storage, electric vehicles, and space/water heating, to help safely ride through the event. Also, if there is a risk of cascading failures, the grid can be partitioned into self-sustaining islands. The proposed research will develop improved algorithms to optimally schedule resources and intelligently determining islands to minimize outages and support restoration. Current approaches inadequately account for the dynamics of modern grids with a high penetration of renewable generation. Wide area monitoring enables coordinated emergency response to prevent blackouts. However, existing wide-area control strategies typically focus on a single phenomenon, consider normal pre-event operation, and assumes availability of system-wide measurements. This research will develop new emergency control algorithms to handle systems in a deteriorated state, multiple instabilities, and partial loss of measurements. When a catastrophic failure occurs, the ability to rapidly restore the grid is a key aspect of resilience. Networkable microgrids are emerging as a means of enhancing resilience by allowing the sharing of generating and storage resources to support fast restoration. Since communication systems can also be affected by the same events, this research aims to develop decentralized and adaptive approaches to grid restoration. The research will also explore methods to ensure fair restoration plans that avoid longer outages for marginalized communities. The research program will train 11 HQPs. Its outcomes can aid power industry to enhance grid resilience, thereby improve public well-being during catastrophic events.