为了提升城市纯电动公交换电效率，降低充电站建设和运行成本，研究了电池交换式纯电动公交充电站选址定容优化问题。首先，考虑纯电动公交车换电站运行保障能力，建立了换电设施数量和电池储备容量模型，获得了换电工位和最优电池配置数量。然后，基于纯电动公交换电需求，利用排队理论对充电站运行工况建模，设置惩罚因子来保证充电站服务质量，建立了以服务半径、服务强度和供需平衡等为约束，以年总成本最小为目标的选址定容模型，应用GA、PSO和PSO-GA算法进行求解。最后，对选址定容模型敏感性分析，以获得充电速率、额定行驶里程等参数对选址结果的影响。案例应用结果表明：PSO-GA算法较GA和PSO算法在目标函数和收敛速度方面更优，充电站最优数量为30座，充电桩数量为943个，最低总成本为1 215.142 9万元。纯电动公交车额定行驶里程与建站数量、运输成本和建设成本呈负相关；充电速率与建站数量和建设成本呈负相关，与运输成本呈正相关，提高充电速率会加快电池老化，降低电池使用寿命。研究结果可为城市纯电动公交车充电站合理规划和运营提供理论依据。

In order to improve the electricity exchange efficiency of urban pure electric bus and reduce the construction and operation costs of charging stations, this paper studied the optimization of urban battery exchange pure electric bus charging station siting and capacity selection. Firstly, considering the operation guarantee capacity of pure electric bus exchange stations, the study established a model of the number of exchange facilities and battery reserve capacity and obtained the number of exchange stations and optimal battery configuration. Then, based on the pure electric public exchange demand, the charging station operation conditions were modeled by using queuing theory, and penalty factors were set to ensure the service quality of charging stations. A site selection and capacity model with service radius, service intensity and supply and demand balance as constraints and minimum annual total cost as the object was established, and GA, PSO and PSO-GA algorithms were applied to solve it. Finally, a sensitivity analysis of the siting and capacity model was performed to obtain the effects of parameters such as charging rate and rated driving range on the siting results. The results of the case application show that the PSO-GA algorithm is better than the GA and PSO algorithms in terms of objective function and convergence speed, and the optimal number of charging stations is 30, the number of charging piles is 943, and the lowest total cost is 12 151 429 000 RMB. The rated driving range of pure electric buses is negatively correlated with the number of stations, transportation cost and construction cost; the charging rate is negatively correlated with the number of stations and construction cost and positively correlated with transportation cost, and increasing the charging rate will accelerate battery aging and reduce battery life. The research results can provide theoretical basis for the reasonable planning and operation of urban pure electric bus charging stations.