Smart charging of PEVs penetrating into residential distribution systems

Abstract

This paper presents a novel modeling framework for the analysis of Plug-in Electric Vehicle (PEV) charging in unbalanced, residential, distribution systems. A Smart Distribution Power Flow (SDPF) framework is proposed to determine the controlled or smart charging schedules and hence address the shortcomings of uncontrolled charging. The effect of peak-demand constraint imposed by the Local Distribution Company (LDC) is also studied within the SDPF framework for the smart charging scenarios. Uncontrolled versus smart charging schemes are compared for various scenarios, from both the customer's and the LDC's perspective. Various objective functions, such as energy drawn by the LDC, total feeder losses, total cost of energy drawn by LDC and total cost of PEV charging are considered. Studies are carried out considering two sample systems i.e., the IEEE 13-node test feeder and a real distribution feeder. Analyses are also presented considering a probabilistic representation of the initial state of charge (SOC) and start time of charging for various scenarios to take into account the difference in customers' driving patterns. The results show that uncontrolled charging of PEVs results in increased peak demand, low node voltage levels, and increased feeder current magnitudes. On the other hand, the SDPF framework provides very satisfactory operating schedules for the overall system including smart PEV charging.