Distributed Convex Optimization for Electric Vehicle Aggregators

Abstract

One of the main challenges for electric vehicle (EV) aggregators is the definition of a control infrastructure that scales to large EV numbers. This paper proposes a new optimization framework for achieving computational scalability based on the alternating directions method of multipliers, which allows for distributing the optimization process across several servers/cores. We demonstrate the performance and versatility of our framework by applying it to two relevant aggregator objectives: 1) valley filling; and 2) cost-minimal charging with grid capacity constraints. Our results show that the solving time of our approach scales linearly with the number of controlled EVs and outperforms the centralized optimization benchmark as the fleet size becomes larger.