Computational efficient approach to compute a prediction-of-use tariff for coordinating charging of plug-in electric vehicles under uncertainty

Abstract

The stochastic nature associated with the baseload (BL) forecasting, photovoltaic (PV) generation and conditions of use of plug-in electric vehicles (PEV) adds new complexity to the definition of PEVs charging coordination strategies. Therefore, a large number of scenarios must be generated to integrate these uncertainties in the definition of a prediction-of-use (POU) tariff that encourages the PEVs to charge at certain times of the day. The main purpose of this article is to analyze the effects of scenario reduction techniques in the determination of an adequate POU tariff that considers the uncertainties associated with BL, PV generation and conditions of use of PEVs. The methodology proposed in this work considers the backward scenario reduction technique to determine the optimal charging power profiles of PEV aggregators through the Distribution System Operator (DSO) coordination. From the PEV optimal charging profiles, the DSO calculates a POU tariff for each aggregator, considering the uncertainties in of the problem. Results show that the use of scenario reduction techniques to determine the POU tariff reduces the computational burden without significantly affecting the obtained results. Finally, the simulation reflects the advantage of integrating PV generation in the distribution system, since using the proposed coordination strategy, the loss of life of the transformer slowed down.