考虑异方差性的城市电网电动汽车充电负荷预测

Charging Load Forecasting for Electric Vehicles in Urban Power Grid Considering Heteroscedasticity

随着电动汽车渗透率的上升,充电设施和充电负荷将成为城市电网发展的关键增长点。然而,大量电动汽车集中充电导致负荷波动剧烈,并加剧了负荷序列的异方差性。为此,提出了一种考虑异方差性的电动汽车充电负荷预测模型。首先,通过图形检测法识别评估充电负荷序列的异方差性,综合负荷变化趋势分析可能导致异方差的原因;其次,通过时间序列分解剥离原始序列中的异方差,挖掘负荷分量与气候、温度的关联关系;最后,构建针对异方差性的中长期充电负荷预测框架,融合长序列预测模型,对负荷进行分解与重构。仿真结果表明,所提模型提高了负荷预测的准确性,为城市电网的合理规划与稳定运行提供了重要支持。

With the increase of penetration rate of electric vehicles,the charging infrastructure and the charging load will become the critical growth point of the urban power grid.However,the concentration of charging activities from numerous electric vehicles has led to significant fluctuations in the load and exacerbated the heteroscedasticity of the load sequence.To address this issue,this paper proposes a forecasting model for electric vehicle charging loads considering heteroscedasticity.Firstly,the heteroscedasticity of the charging load sequence is identified and evaluated by using the graphical detection method,which is combined with the load variation trends to analyze possible causes of heteroscedasticity.Secondly,through the time-series decomposition,the heteroscedasticity within the original sequence is separated,and the relationship between load components and climate or temperature is revealed.Finally,a medium-and long-term forecasting framework for the charging load with heteroscedasticity is constructed,which is merged with the long-series forecasting models to decompose and reconstruct the load.The simulation results show that the proposed model improves the accuracy of load forecasting,and provides significant support for the rational planning and stable operation of urban power grids.