Abstract: In order to meet the demands of promoting the charging station planning under the "double carbon" goals, and to solve the problems like long queuing time and long distance between stations caused by the unreasonable charging station planning, a method for charging station planning considering the spatial and temporal distribution characteristics of the electric vehicle charging demands is proposed in this paper. Firstly, the online car trip data are fused with the urban interest point data crawled by the Python to divide the study district into the functional areas to mine for the traveling habits of the users; Secondly, a two-layer road selection model is established considering the driving characteristics of the electric vehicles, and the charging demands are predicted by combining the charging characteristics of the users; Thirdly, a charging station site-setting model is established by integrating the three levels of electric vehicle users-charging stations-grid with a Voronoi diagram for the charging demand coverage, and the optimal locations of the charging stations are determined by improving the inertial weight particle swarm algorithm. Finally, the model is applied to the Second Ring Road area of the city Chengdu in Sichuan province for the feasibility verification.