基于电动汽车时空特性的充电负荷预测  

A Method of Charging Load Forecast Based on Electric Vehicle Time-Space Characteristics

作　　者：张艳娟[1] 苏小林[1] 闫晓霞[1] 李敏[1] 李丹丹[1] 

机构地区：[1]山西大学,太原市030001

出　　处：《电力建设》2015年第7期75-82,共8页Electric Power Construction

摘　　要：电动汽车充电负荷预测是进行充电设施、电网规划建设以及运行调度控制的基础。电动汽车充电负荷的时空分布具有很强的随机性,在对预测区域空间进行划分的基础上,考虑电动汽车的动态转移特性,对不同功能用地的泊车规律进行分析,预测不同类型电动汽车的空间分布,进而对不同电动汽车充电时间特性的影响因素进行分析,并建立了预测模型。利用蒙特卡洛仿真方法对某市一区域在不同情景下的充电负荷进行计算。结果表明,不同功能区的充电负荷分布特性差异明显,并且采用快速充电方式的比例越高,峰谷差越大,因此可根据预测结果对电动汽车充电时间、充电地点和充电方式进行合理引导,使在满足充电需求的同时,减少充电负荷对电网的影响。Electric vehicle charging load forecasting is the precondition for charging infrastructure construction and charging load coordination control. In this paper,considering the stochastic charging behavior,the parking rules of the different functions areas are analyzed,the spatial distribution of different types of electric vehicles are predicted,and the influential factors are analyzed furthermore. The predicting model is set up as a rusult. The overall daily charging loads at different scenarios are calculated by M onte Carlo simulation. Results have shown that the difference of charging load features for different districts is obvious,and the proportion of using quick charge is higher,the difference of peak and valley is greater. According to the predicted results,not only can we guide charging time and place reasonably,but also provide the basis for charging infrastructure construction.

关 键 词：电动汽车 充电负荷预测 空间动态分布 蒙特卡洛仿真 

分 类 号：TM715[电气工程—电力系统及自动化]