摘要
随着电动汽车的快速推广应用,对电动汽车充电负荷需求实现较准确的预测也逐步成为电动汽车领域研究的热点。为解决电动汽车大量推广应用后的电网负荷平衡问题,在分析电动汽车时空特性分布的基础上,提出了一种基于自适应动态三次指数平滑法的电动汽车负荷预测模型。该预测模型对传统的三次指数平滑模型进行改进,并根据误差绝对值之和最小的原则及时调整,以获得理想的平滑系数,再对后续的年度最大日电动汽车充电量和不同小区的典型日电动汽车充电负荷进行预测。以某工业小区和居民小区电动汽车负荷分布为例进行仿真,仿真结果表明所提方法误差较小,给出的充电桩规划建议也切实可行。
With the rapid development and application of electric vehicles, the accurate charging load forecasting of electric vehicles is becoming a research focus. In order to handle the grid load balance with the electric vehicles expansion, this paper proposes a self-adaptive and dynamic forecasting method based on electric vehicle features of spatiotemporal distribution. The forecasting model improves the traditional cubic exponential smoothing model, and the ideal smoothing factor is changed in time by the principle of minimum absolute value error; then electric vehicle charging quantity and charging load forecasting from different areas on typical day can be done with the new method. Load distribution of electric vehicles in an industrial area and residential quarter is simulated, and the result shows that the method can reduce errors and its suggestions on charging poles are practical and feasible. The simulation results show that the proposed method is feasible.
出处
《浙江电力》
2016年第12期15-20,共6页
Zhejiang Electric Power
基金
国网浙江省电力公司群众性科技创新项目(5211JX1500KM)
关键词
电动汽车
负荷平衡
时空特性
自适应
负荷预测
electric vehicle
load balance
features of spatiotemporal
self-adaptive
load forecasting
