This paper explores the location selection problem of electric vehicle charging stations, aiming to effectively meet charging demands within urban areas. The research utilizes data from Ride‐hailing order data as the study data. Through data cleaning, processing, and matching, the city is divided into grids, and the charging demand of each grid is analyzed. To capture the distribution characteristics of demand points, we use the mini-batch k‐means algorithm to perform clustering analysis of pick‐up and drop‐off orders. Based on the clustering results, we calculate the average distance between each grid and the clustering center and use it as the distance weight. Combined with the order quantity, we calculate the comprehensive weight of each grid. We use the particle swarm algorithm to solve the charging station location selection problem. This algorithm determines the appropriate charging station location by prioritizing grids with higher comprehensive weights. By prioritizing grids with higher comprehensive weights, we establish charging stations in areas with high demand and close to the clustering center, thereby improving the utilization rate of charging stations. We also analyze the applicability and limitations of the particle swarm algorithm and visualize the location selection results. This study provides a simple and efficient solution method for the location selection problem of urban electric vehicle charging stations based on actual order data, which helps provide a reference for urban planners and charging facility operators, promotes the rational layout of electric vehicle charging facilities, and further promotes the popularization and sustainable development of electric vehicles.