A Forecasting Method for EV Charging Demand Considering Refined Division of Functional Areas

The dynamic spatial-temporal distribution prediction of electric vehicle charging demand is of great significance for the charging operation management in the power grid. To achieve accurate prediction of charging load distribution in urban areas, firstly, the characteristics of different functional areas in the city are considered. Then, the POI data in Baidu Map platform is used to carry on the data mining and modeling, which helps model the division of urban functional areas. Then, a spatial-temporal distribution model of EV charging demand is established based on the theory of Agent-cellular automata (CA) to accurately describe the mathematical model for EVs. The model contains the changes of location and state of charge (SOC) in the drive. Finally, a certain city area is selected as an example to conduct the charging demand load prediction under multiple scenarios. The results show that the proposed model can effectively predict the spatial-temporal distribution of charging demand loads for different date types.