An approach for minimizing annual energy loss by electric vehicle scheduling to optimal fast charging stations

In order to mitigate the effects of electric vehicles’ (EVs) high power consumption on the distribution network, this study proposes a two-stage approach. First, it uses node traffic density forecasts to select optimal locations for fast charging stations (FCS). Then, to reduce the power demand on the distribution system, it schedules EVs to these FCSs. Considering constraints such as charger number, station capacity, and desired battery state of charge (SOC), the objectives are to minimize annual energy loss, optimize load demand, decrease station development costs (SDC), and maximize charging station (CS) income. The best locations for FCS are determined using a gray wolf optimization (GWO) technique. Rather than allocating EVs to uniform base loads, the study recommends analyzing different initial SOC cases to minimize power demand at FCSs. For EV scheduling, a queuing-averse charging station algorithm is suggested to reduce annual energy loss and increase station income. For various SOC cases relative to the base load scenario, this suggested method achieves an average daily FCS charging time of 648 min, a power loss of 16.8 kW, and an annual energy loss of 9.38 kW.

为缓解电动汽车（EVs）高用电负荷对配电网产生的负面影响，本研究提出一种两阶段优化方法。第一阶段，依托节点交通密度预测结果选取快速充电站（FCS）的最优布局点位；第二阶段，通过将电动汽车调度至各快速充电站，以降低配电网的系统用电需求。本研究在建模过程中考虑充电桩数量、场站容量、目标电池荷电状态（SOC）等约束条件，以最小化年能耗损耗、优化负荷需求、降低充电站建设成本（SDC）并最大化充电站（CS）收益为核心优化目标。快速充电站的最优选址采用灰狼优化算法（GWO）完成求解。相较于将电动汽车分配至均匀基准负荷的常规调度思路，本研究建议针对不同初始荷电状态场景开展分析，以降低快速充电站的用电需求。针对电动汽车调度环节，本文提出一种避排队充电站调度算法，用于降低年能耗损耗并提升充电站收益。相较于基准负荷场景，在所考虑的各类初始荷电状态场景中，所提方法可实现快速充电站日均充电时长648分钟、功率损耗16.8kW，以及年能耗损耗9.38kW的优化效果。