Electric vehicle-distribution grid charging load forecasting strategy considering spatial and temporal distribution and dynamic traffic information

The charging process of electric vehicles (EVs) connected to the distribution network is highly random and results in significant load fluctuations, which is not conducive to accurate load forecasting and the efficient operation of the distribution network. This paper investigates the load prediction and its impact on the operation state of the distribution network when charging into the distribution network under different scenarios. First, the coupling and interaction characteristics of the road network, distribution network, and EVs are analyzed, followed by the establishment of a real-time road network-distribution network interaction model containing the spatial and temporal stochastic distribution characteristics of EVs and dynamic traffic information. Second, based on the characteristics of various EV types, travel time, spatial and temporal distribution, and initial SOC, this paper establishes a charging path planning and charging residual time calculation model that takes ambient temperature, vehicle speed, and power consumption into account, and realizes the load prediction of the EV charging process. This paper concludes by simulating and validating the established EV load prediction model and its impact on the distribution network using an example. The results demonstrate that the load prediction model developed in this paper can accurately predict the charging load of electric vehicles (EVs) under the influence of spatial and temporal distribution characteristics and dynamic traffic information in various geographic regions, and provide technical references for the planning of charging stations and distribution networks.