EAGER: Novel Strategies for Large-Scale Charging Control of Electric Vehicles Under Power System Stability Constraints

Electric vehicles (EVs) are envisioned to be an integral component for the US power distribution system in the coming years, raising questions on how the control logic, scheduling, and pricing of EV charging may adversely impact the dynamical properties and stability of the distribution grid. If the number of EVs keep increasing, and if EV owners keep following the same control and pricing mechanisms as they do now, that may encourage a large fraction of drivers to charge their cars at certain specific times of the day thereby causing overloading problems in the grid. Grid operators, therefore, must find practical ways for incentivizing EV customers to settle for a slightly lesser charging demand than what they want for the same charging duration, and thereby control charging patterns across neighborhoods and cities so that instability issues in the local distribution grid can be prevented. The objective of this EAGER project is to take a step forward towards addressing this practical challenge. Both model-driven and data-driven optimization and control algorithms will be developed to translate EV charging information submitted by drivers to suitable routing and scheduling routines that come with guaranteed small-signal stability and voltage stability of the grid. The intellectual merit of this project lies primarily on evaluating massive-scale EV integration and developing a new understanding of power electronics-based control mechanism based on the principles and applications of dynamics and control systems theory to help prevent instability in electric power grid. The broader impacts include bridging a long-standing gap between control theory and vehicular power electronics, integrating research results with power system courses, organizing workshops and conference tutorials, and collaboration with industry.

This project will address two main tasks. The first task will be to understand the fundamental relationship between charging currents drawn through DC fast chargers and the stability margins of the grid where these chargers are installed. Optimization and control algorithms will be developed to maximize these margins in return of monetary incentivization offered to EV owners. In some cases, this relationship may not follow entirely from the physical knowledge of the charging circuits, in which case machine learning based methods will be used. The second task will address scenarios where some EV owners may bias the optimization problem by submitting inaccurate information about their charging demands to maximize their individual incentives. Strategies will be developed to eliminate such biases using ideas from game theory, optimal control, and adaptive dynamic programming. The study will promote many new directions of theoretical and experimental research for tomorrow’s energy networks and their integration with transportation networks.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Award Number: 2533039
Principal Investigator: Aranya Chakrabortty
Funds Obligated: $299,961
State: NC