For the purpose of reducing the construction cost of EV charging station and ensuring the charging demand of users, it is imperative to scientifically plan the site selection of charging station. Based on POI data, K-means clustering was used to obtain demand centers. Combined with the existing resource basis of urban public parking lots, a site selection optimization model with the goal of “minimizing the construction cost of charging stations, maximizing the satisfaction of users’ distance, and maximizing the coverage of demand” was constructed. This model balances economic cost, user convenience and service efficiency through multi-objective collaborative optimization. Taking Hongshan District of Wuhan City as an empirical case, the model solution results show that building charging stations based on the existing parking lots can significantly reduce the investment in land and infrastructure, which is more in line with the current situation. Compared with the conventional single-objective site selection method, this model effectively avoids the limitations of cost or convenience orientation, achieves a balance of interests among the government, operators and users, and provides a practical reference for the sustainable planning of urban charging infrastructure.