Controlled Charging of Electric Vehicles to Minimize Energy Losses in Distribution Systems

Abstract

Abstract: Controlled electric vehicle (EV) charging scenarios are proposed, each characterized with an algorithm and associated computational and communication requirements, to be adopted by an EV aggregator or system operator. The proposed scenarios include uniform, random, conditional-random, and valley-filling charging scenarios. Different from previous studies, this paper focuses on easy-to-implement charging scenarios. Further, a modeling framework is presented to investigate the impact of the proposed charging scenarios on energy losses in distribution systems, as compared to an uncontrolled charging scenario and a reference scenario with no EVs. The modeling framework considers uncertainties involved in the behavior of low voltage customers and EV charging loads. It is applied to a distribution system for various case studies, including different penetrations and combinations of EVs with various characteristics. As seen by the results, although the valley-filling charging algorithm represents the optimal solution from the perspective of energy losses, the uniform charging scenario emerges as a quasi-optimal solution, having lower computational and communication requirements, which makes it easier to be adopted by EV aggregators or system operators. Further, with appropriately selected coefficients, the conditional-random charging algorithm can also exhibit a performance comparable to that of the uniform charging algorithm.