基于交通出行矩阵的私家车充电负荷时空分布预测

A Temporal and Spatial Distribution Forecasting of Private Car Charging Load Based on Origin-Destination Matrix

针对电动汽车空间负荷预测中充电地点、充电方式不确定性的难题,提出了一种基于交通出行矩阵和云模型的充电负荷时空分布预测方法。首先,通过监测道路流量,反推小区的交通吸引量,动态预测不同地点的停车概率。其次,在选择充电方式时,根据快充、慢充特点,制定用户心理到快充概率之间的转换规则,并在规则中引入云模型,体现用户决策的随机性和模糊性。最后,利用蒙特卡洛法分析计算出不同充电地点的负荷时间曲线,并以某城市中心城区的数据为例,验证了该方法的有效性。计算结果表明,不同小区、不同工作日的交通量变化明显,且充电负荷曲线受交通量变化的影响显著;快充负荷将在一定范围内随机波动,提高慢充失效阈值将减小快充负荷峰值。

To solve the problem of the uncertainty of charging sites and charging modes in forecasting electric vehicle spatial loads, a temporal and spatial distribution forecasting method based on origin-destination （OD） matrix and cloud model is proposed. First, through monitoring road traffic, the traffic attraction volume of the residential areas can be inversely deduced and the parking probability at different locations can then be dynamically predicted. Next, according to the characteristics of fast charge and slow charge, the conversion rules between usersr psychology and fast charging probability can be formulated. Furthermore, the cloud model will be introduced in the rules to reflect the randomness and fuzziness of usersr decision. Last, the load time curves of different charging sites will be analyzed and calculated via applying the Monte Carlo method, whose validity has been verified by data from an urban center city as an example. The calculation results show that the traffic volume changes apparently in different residential areas and different working days, and the charging load curve is significantly affected by the traffic volume change. The results also show that the fast charging load will fluctuate in a certain range randomly. Moreover, increasing the slow charge failure threshold will reduce the fast charging load peaks.