In order to solve the problems of traditional methods ignoring the trade-off between cost and service level, the coverage is not used as a rigid constraint, and the difficulty of global optimization caused by the spatio-temporal heterogeneity of demand under complex road networks, a new site selection model that integrates dual-objective optimization and coverage constraints is proposed, and the Pareto optimal solution set is designed with the improved whale optimization algorithm (IWOA). The model uses the distance satisfaction function based on cosine and the M/M/s queuing model to quantify the service level, and evaluates the cost by integrating fast/slow charging preferences and peak hours correction by capital recovery factors, and realizes collaborative optimization under the constraints of unique coverage of demand points and steady-state of queuing system. Secondly, the adaptive weight control mechanism, hybrid search strategy and local search enhancement are introduced in the algorithm design to significantly improve the global search ability and convergence efficiency of the algorithm. Finally, based on the POI data and actual road network data in Xi'an, 80 demand points and 64 candidate stations are selected, and the performance of IWOA, PSO, GA and other algorithms is compared through example analysis, and a cost-service-level-coverage three-dimensional decision-making framework is constructed to visually display the impact of constraints on the optimization goal. The results show that the proposed method achieves the optimal trade-off when the coverage rate constraints are 0.8, the service coverage rate is increased to 93.6±1.7%, the full-cycle cost is reduced to 12±2473 million yuan, and the average waiting time of users is shortened to 3.3±0.4 minutes. This study can provide a systematic decision-making framework for solving the problem of lagging charging infrastructure layout, and has important practical value for improving the efficiency of urban electric transportation network.