Modeling of PEV Charging Load Using Queuing Analysis and Its Impact on Distribution System Operation Omar Hafez, Student Member, IEEE Department of Electrical and Computer Engineering, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada Ohafez@uwaterloo.ca Kankar Bhattacharya, Senior Member, IEEE Department of Electrical and Computer Engineering, University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada kankar@ece.uwaterloo.ca Abstract— This paper presents a novel approach for modeling the 24-hour charging demand profile of a plug-in electric vehicle (PEV) charging station using queuing analysis. The proposed queuing model considers the arrival of PEVs as a non- homogeneous Poisson process with different arrival rates over the day. A distribution optimal power flow (OPF) model is used to study the impact of the PEV charging load of the charging station on distribution system operation. Various objective functions, such as total feeder losses, energy drawn by the local distribution company (LDC) and total cost of energy drawn by LDC are considered in this paper. Index Terms-- Queuing analysis, Plug-in electric vehicle (PEV), Smart grid, Distribution system, Distribution optimal power flow, ZIP load I. INTRODUCTION As we move toward a greener future, the Plug-in Electric Vehicles (PEVs) have an increasingly important role to play, because of their contribution to emissions reduction from the transportation sector. However, increased numbers of PEVs can have a significant impact on the power distribution system operational performance. The study of the impact of electric vehicle charging profile goes back to the 1980s. Several studies show that the power distribution grid can be significantly impacted by high penetration levels of PEVs. In [1], the effects of electrical vehicle deployment on load management effectiveness is examined and it is noted that electric vehicle charging will likely coincide with the system peak demand and thus, in order to avoid overloading of the distribution feeders, adequate load management schemes need be in place. The objectives of minimizing the load factor and load variance are used in [2] to study the impact of Plug-in Hybrid Electric Vehicle (PHEV) charging on the distribution system. The impacts of PEV fleets on the bulk power system are estimated in [3] by proposing algorithms for the scheduling and dispatch of electric power, by aggregators of PEV fleets, based on realistic vehicle travel patterns from the National Household Travel Survey (NHTS) database [4]. A modeling framework that incorporates the operation and coordinated charging of PEV loads within a three-phase unbalanced, residential, distribution system is proposed in [5] to examine the impact of PEVs on the overall system load profile, bus load profiles, feeder currents, voltages, taps and capacitor switching. Different penetrations of PEVs considering real data of residential load, ambient conditions, and vehicle parameters are examined in [6] to study their effect on the transformer insulation life. Penetration of PEVs into the market is expected to be large in the near future, and with their complex charging behavior, their charging load model need be investigated. Vehicle usage data for 76 vehicles in a one-year period in Winnipeg, Canada is analyzed in [7] to predict PEV charging profiles and their charging infrastructure effect on utility load. The travel patterns of light-duty vehicles in the USA obtained from the 2009 NHTS [4] is used in [8] to estimate the electrical energy and power consumption of such PEVs for two uncontrolled charging scenarios. A modeling methodology is proposed in [9], where detailed vehicle usage patterns are taken into account and statistical distributions of charging energies are produced to develop statistical charging load models of PHEVs considering the US NHTS data [4]. Modeling the PEV charging demand using queuing analysis [10] considering different characteristics is discussed in several studies. A probabilistic constrained load flow problem with wind generation and electrical vehicle demand is presented in [11] where the charging and discharging processes are considered using M/M/∞ queuing model. A mathematical model that covers the spatial and temporal distribution of demand, based on fluid dynamic traffic model and queuing theory is developed in [12] to determine the electrical vehicle charging demand for a rapid charging station. In [13], an electrical vehicle demand model suitable for load flow studies is proposed wherein the electrical vehicle demand is represented as a PQ bus with stochastic characteristics based on the concept of queuing theory. In [14], a single PHEV charging demand model is formulated and queuing theory is used to describe the behavior of multiple PHEVs. The first author acknowledges the funding support received from Umm Al- Qura University, Makkah, Saudi Arabia, to carry out this research. 978-1-4673-8040-9/15/$31.00 ©2015 IEEE