考虑电动汽车空间分配的多目标配电网重构优化

电动汽车充电过程在空间呈现的无序性会对配电网重构带来影响。在电动汽车用户倾向于选择距离最近、充电时间最少的充电站进行充电这一基础上,提出一种考虑电动汽车空间分配的配电网重构优化模型。以网络损耗和电压偏移水平为指标建立多目标优化函数,引入电动汽车效用度这一约束条件,对网络开关状态和电动汽车充电负荷空间分配方案同时优化。通过模糊C均值聚类对系统中电动汽车按照空间分布指标进行聚类,将空间分布位置相似的电动汽车聚为一类以降低问题模型维度。然后采用改进遗传算法对模型进行求解以获得综合优化方案。仿真结果表明,聚类能够显著降低计算难度,提高算法收敛速度;在优化模型的引导下,电网在保证自身运行经济性和可靠性的基础上,降低了电动汽车的充电时间和路程,证明了模型和方法的有效性。

Multi-objective Optimization of Distribution Network Reconfiguration Considering Electric Vehicles Spatial Allocation

The disorder of electric vehicles charging process in space will greatly influence the distribution network reconfiguration. On the premise of tending to choose the station of shortest distance and least time for users, a multi-objective optimization model considering space allocation of electric vehicles was established in this context. This model chose network loss and voltage deviation level as optimization goal with utility degree of the electric vehicles as constraint condition to optimize the network switch state and electric vehicles charging load space allocation at the same time. The fuzzy c-means clustering was adapted to cluster electric vehicles in the system according to the spatial index, reducing the dimension of the model effectively. Finally, the model is solved by the improved genetic algorithm to obtain an integrated optimization scheme. The simulation results show that the fuzzy c-means clustering can significantly reduce calculation difficulty and improve the convergence rate of the algorithm, besides, with the guidance of the reconstructed model, the grid ensured its operation economy and reliability, reducing the electric vehicle charging time and distance, proving the validity of the model and method.