Rolling Multi-Period Optimization to Control Electric Vehicle Charging in Distribution Networks

• Published in: IEEE transactions on power systems Vol. 29; no. 1; pp. 340 - 348
• Main Authors: O'Connell, Alison, Flynn, Damian, Keane, Andrew
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2014; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Availability; Battery; Electric vehicles; Electricity; Jacobian matrices; load flow; load management; Load modeling; Optimization; Power cables; Sensitivity; smart grids
• ISSN: 0885-8950; 1558-0679
• DOI: 10.1109/TPWRS.2013.2279276

The integration of electric vehicles (EVs) poses potential issues for low voltage (LV) distribution networks, such as excessive voltage deviations and overloading of equipment. Controlled EV charging is seen as one possibility for reducing, or even eliminating, these issues. The implementation of controlled charging schemes, in particular centralized schemes, can require forecasts of a number of variables, e.g., household loads, EV availability, and battery requirements. The unpredictability of individual customer behavior may, however, lead to large variations between forecast and realized behavior. This work presents a multi-period, unbalanced load flow and rolling optimization method, which focuses on controlling the rate and times at which EVs charge over a 24-h time horizon, with a minimum cost objective, subject to certain constraints. Inputs are updated and a new optimization is performed at each time step so that deviations from the initial forecast can be readily accounted for.