Since electric vehicle users have difficulty finding a balance between charging time and charging location, it is impossible to accurately grasp when and where to perform charging behavior. This paper first selects the driving condition characteristic pa-rameters that can accurately reflect the unique driving characteristics of users in the actual road, and uses the working condition identification method to construct the remaining electricity estimation model of electric vehicles under real-time dynamic road conditions to judge when there is charging demand in the travel process. When there is charging demand, the electric vehicle waiting time model is established by predicting the number of arriving vehicles of the charging station, and the effective charging period is planned for the user as the basis for selecting the charging location; considering the coupling relationship between the charging time and the charging location, From the user"s point of view, under the constraint of the remaining battery power, the electric vehicle charging path model with the best combination of the user"s travel distance, travel time and charging cost is con-structed and applied to the actual traffic network in Beijing. In the region, the ant colony algorithm is used to verify the simulation.