Electric vehicles (EV) show huge potentials in terms of reducing greenhouse gas emissions.However, the integration of EV into power systems may also bring challenges, such as demandincrease during peak hours. Therefore, charging behaviors of EV should be scheduledappropriately.This paper proposes an optimization model to address the charging management problem of EV.The EV charging management problem is formulated as a stochastic linear programming model.The objective of the model is to minimize the distance between the EV total charging demandand a pre-defined reference demand curve within a one-day period. With this objective, the truetask of the model depends on the setting of the reference demand, which makes the model easilyapplicable for different purposes. These reference demands might focus on integrating localrenewables, arbitrage trading on different electricity markets or load management from the localgrid perspective. As future arrivals of EV are uncertain but contribute to the total charging demandin future periods, the model considers these uncertainties by a stochastic term in their arrival time,departure time and electricity demand (i.e. state of charge). Hence, EV usage patterns aresimulated with inhomogeneous Markov chains and scenario reduction technique is applied toreduce the number of scenarios considered and to guarantee feasible computing times. Thesimulation results demonstrate that the controlled charging strategy outperforms the uncontrolledcharging in terms of demand decrease during peak hours and that the proposed model managesto distribute the EV charging demand throughout the day.
展开▼
