Stochastic Optimal Sizing of Plug-in Electric Vehicle Parking Lots in Reconfigurable Power Distribution Systems

• Published in: IEEE transactions on intelligent transportation systems Vol. 23; no. 10; pp. 17003 - 17014
• Main Authors: Mohammadi-Landi, Meysam, Rastegar, Mohammad, Mohammadi, Mohammad, Afrasiabi, Shahabodin
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.10.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Charging; Distribution networks; Electric power distribution; Electric vehicles; Electrical plugs; Integer programming; Mixed integer; Network reliability; Optimization; Parking facilities; plug-in electric vehicle parking lot; Power distribution; power distribution system losses; Probability density function; Reconfiguration; Reliability; reliability of power distribution system; Sensitivity analysis; Sizing; Stochastic processes; Sustainable transportation; Uncertainty; Vehicle-to-grid; Voltage
• ISSN: 1524-9050; 1558-0016
• DOI: 10.1109/TITS.2022.3166781

Charging demand of plug-in electric vehicles (PEVs) can cause reliability and operational challenges in power distribution systems. The aggregated charging control of PEVs in the parking lots (PLs) may alleviate the challenges, if the place and size of PEV PLs are optimally determined. This paper develops a stochastic framework for finding optimal location and sizing of PLs as well as optimal charging profile of PEV PLs with the vehicle to grid capability in a reconfigurable distribution system. The main aims are to reduce distribution system losses and enhance network reliability, subject to numerous constraints of the power distribution system, PLs, and PEVs. To guarantee the global optimum solutions, the proposed optimization problem is linearized to achieve a mixed-integer linear program model. Furthermore, kernel density estimator (KDE) is presented to model the temporal uncertainties associated with PEV owners' behavior with small number of iterations and no necessary assumptions. Various scenarios and sensitivity analysis are conducted to show the efficacy of the proposed method.