Many-objective optimization for large-scale EVs charging and discharging schedules considering travel convenience

摘要

The uncontrolled charging behaviors of large-scale electric vehicles (EVs) increase the security risk of the power grid and bring a new challenge for the computing ability of the power system. Using vehicle to grid (V2G) technology, most control systems coordinate the power interaction between EVs and power grid by minimizing the load fluctuation and user cost, but their optimization results are often achieved at the expense of reducing personal travel time. EVs should first meet basic travel needs and then obey the scheduling arrangement. Based on this idea, a four-objective optimal control method for EV charging and discharging schedules considering travel convenience is proposed, including minimization of the load fluctuation and user cost and maximization of the flexible travel time and state of charge (SOC). To solve this large-scale many-objective problem, a resource allocation-based preference-inspired coevolutionary algorithm (PICEAg-EV) is presented. Taking the IEEE 33-node system as an example, the simulation and analysis verify the effectiveness of the proposed control strategy and optimization algorithm. The experimental results show that PICEAg-EV outperforms seven popular intelligence algorithms under EV participation rate setting of 10%, 25%, 50%, 100%. Compared with 2- and 3-objective optimization models, the 4-objective optimization model can provide sufficient flexible travel time and a higher SOC for traveling, which is a better match for the user needs.