A Fully Distributed Cooperative Charging Approach for Plug-In Electric Vehicles

Abstract

The flexibility of plug-in electric vehicles (PEVs) in shifting their charging schedules can be utilized to reduce charging costs. Given the ever-increasing adoption of PEVs and their geographical spread, coordinated charging schedules could be enabled by distributed algorithms. Here, we propose a fully distributed solution for PEVs cooperative charging (PEV-CC) problem. The PEV-CC minimizes the charging costs for a PEV fleet whilst considering limitations of PEVs and charging infrastructure. The PEV-CC is a convex multi-time step problem and a receding horizon is employed to integrate feedback into the decision-making process. Driving uncertainties are accounted for by considering multiple driving scenarios for individual PEVs. Our distributed iterative procedure achieves a distributed solution of the underlying convex optimization problem through local computations and limited communication. The algorithm is designed to reach an agreement on a price signal among PEVs over the course of iterations, while local PEV constraints are enforced at each iteration. Therefore, each iteration yields a feasible solution for the PEV-CC problem. The performance of our proposed algorithm is evaluated on a fleet of PEVs as a test case.