Decentralized charging coordination of plug-in electric vehicles in distribution networks with transformer aging considered

Coordination algorithms are needed for large-scale plug-in electric vehicles (PEVs) charging problems characterizing battery health and distribution transformer aging. It aims at achieving cost efficient and safe operation without excessive centralization of control decisions. However, the presence of transformers causes coupling among PEVs charging behaviors which brings challenge to make decisions in decentralization. To this end, this paper presents an algorithm based on the so-called alternating direction method of multipliers (ADMM) to coordinate multi-period PEV charging problems amidst the coupling relationship among PEVs charging behaviors. The individual costs include not only nominal PEV charging costs, but also a deviation from the difference between updated auxiliary and dual variables. In a setting where the dual variable is a gradient descent of the augmented Lagrangian multiplier, we prove that the algorithm guarantees convergence to the optimal coordination. Simulation results show the optimality and convergence of the proposed method.