电动汽车充电站有序充电调度的分散式优化

大量电动汽车无序充电会给电力系统尤其是配电系统的安全与经济运行带来影响甚至挑战。针对集中式优化与控制方法的不足和固定电价策略的缺陷,基于拉格朗日松弛法,将传统的电动汽车充电站有序充电调度集中式优化问题分解为N个子问题(N为需充电电动汽车数量),提出了有序充电调度的分散式优化策略。优化模型以充电站收益最大为目标函数,考虑了用户用电需求、充电时间、变压器容量等约束和充电站分时电价策略。为验证所提方法的有效性,采用蒙特卡洛法模拟电动汽车充电需求,对采用集中式优化和分散式优化策略的有序充电和无序充电情形,以及充电站售电固定电价和分时电价模式下的充电站收益、削峰填谷效果、计算效率等进行仿真计算和分析。结果表明,所提方法相比于无序充电及充电站固定电价策略,可显著提高收益;相比于集中式优化,计算效率更高;充电站采用售电分时电价虽有"填谷"效果,但平抑负荷波动效果并不十分理想。

Decentralized Optimization of Ordered Charging Scheduling in Electric Vehicle Charging Station

The security and economic operation of the power system is impacted or even challenged by the disordered charging of large-scale electric vehicles, especially for the distribution system. Aiming at the defects of centralized optimization and control method and the fixed pricing strategy, a decentralized optimization strategy of ordered charge scheduling is proposed based on the Lagrange relaxation method. The traditional centralized charging optimization problem of electric vehicle charging station is decomposed into N sub problems(N is the number of electric vehicles demanding charging). The maximum profits of the charging station are used as the objective function in the optimization model. The constraints of power demand of users, charging time, capacity of transformer and time-of-use(TOU)price of charging station are taken into consideration. In order to verify the effectiveness of the proposed method, Monte Carlo method is adopted to simulate the charging demand of electric vehicles. The profits of charging station, the effect of clipping load peak and the calculation efficiency under the fixed price and TOU price strategy of the charging station in situations of ordered charging and disordered charging when adopting centralized and decentralized optimization respectively are simulated and analyzed. The simulation results show that the proposed method can significantly increase the profits compared to use the disordered charging and fixed price strategy. By contrast with the centralized optimization, the computational efficiency of the proposed decentralized optimization is much higher. Though the charging station using TOU price has the effect of filling valley, the effect is not too ideal in stabilizing the load fluctuation.