Abstract: In view of the problem of inaccurate prediction due to insufficient consideration of the interaction between vehicle-roadstation-grid and other parties in the study of spatial-temporal distribution of charging loads for electric vehicles（EVs）, a prediction model of spatial-temporal distribution of charging loads for EVs based on the universal gravity model is proposed. Firstly, the relationship between the external environment and the energy consumption of EVs is explored, taking into account the road network traffic flow and ambient temperature. Secondly, considering the influence of external environmental factors such as temperature, humidity and radiation on users’ trips, a trip chain model modified by travel intention is obtained. Finally, considering the multi-information fusion, a charging station selection model for EVs based on the universal gravity model is proposed. The results show that the proposed model can take into account the mutual influence of EVs, road networks, charging stations and power grids, accurately calculate the spatial-temporal distribution of charging loads for EVs, and analyze the characteristics of charging loads for EVs in multiple scenarios and regions.