A Hierarchical ADMM Based Framework for EV Charging Scheduling

Abstract

Electric vehicles (EVs) are controllable loads from which distribution grid operator can benefit in order to minimize the load profile variations. In this paper, we proposed a hierarchical distributed optimization framework such that EV management system (EVMS), as a part of distribution grid management system, minimizes the load variance of the grid in communication with the EV aggregators which control EV charging load of the distribution system feeders. The hierarchical distributed framework, based on alternative direction method of multipliers (ADMM), increases the scalability of the EV charging infrastructure while decreases computational burden. In our proposed approach, each EV aggregator schedules the EV charging profiles of its feeder in a distributed fashion which avoids sharing the EV owners' desired charging profile information and enables privacy preserving. To show the performance of our approach, we apply it to a case study with 100% EV penetration, including 4 feeders and 60 EVs, and show how the load variance of the system and charging cost of individual EVs decrease.