A load calculation method considering the charging probability of EV users and what kind of power charging pile is selected is given in the urban multi-use scenario. Firstly, the temporal and spatial distribution characteristics of private electric vehicles in urban power consumption scenarios are considered from a macroscopic perspective. Secondly, The main factors affecting the charging decision are analyzed from the user's point of view. Three variables, namely electricity price data, battery SOC and parking duration, are input to fuzzy inference system to generate the user's charging probability to describe user's charging willingness. Dynamic programming theory is used to determine what kind of power charging pile to charge. Finally, based on Monte Carlo idea, a spatial-temporal load prediction model for charging private electric vehicles in urban multi-electricity scenarios is given. By examples, it shows that the model can be used to analyze and calculate the charging load of private electric vehicles of city flexibly.