为了应对环境危机，我国制定2030年前实现碳达峰，2060年前实现碳中和的“双碳”目标，“双碳”目标的实现需要依靠新能源的发展，而电动汽车产业的发展在新能源的发展中起到举足轻重的作用。目前电动汽车的推广应用主要受限于短的车辆里程、长的充电时间以及充电基础设施的不合理布局。由于车辆里程和车辆充电时间在短时间内难以得到突破，那么合理有效的充电基础设施布局成为推动电动汽车产业发展的关键。本文从基于流量需求的角度分模型应用和创新两个层面研究充电基础设施布局规划。在应用层面，本文基于经典的流量捕获选址模型(FRLM)和灵活形式的流量捕获选址模型(FRFRLM)，对上海市嘉定区充电基础设施进行布局规划。针对模型无法在有限时间内有效处理大规模交通网络问题，本文提出了一种在较短时间内解决大规模交通网络问题的算法，可以使得模型能够在短时间内有效处理大规模交通网络问题。针对FRLM和FRFRLM两种模型缺乏对比的问题，本文以上海市嘉定区为案例来探究两种模型之间的异同，结果表明FRLM和FRFRLM在流量捕获以及选址点方面基本相同，但是在处理大规模交通网络问题时，FRFRLM程序运行时间远小于FRLM，这表明基于FRFRLM模型拓展更有意义。最后，针对目前城市真实充电设施建设情况缺乏评价的问题，本文提出了一种能够将真实充电设施影射到交通网络的方法，可以使得现实建设的情况与模型输出进行对比，从而能够对城市当前的充电基础设施建设情况评价并提供一些指导。在创新层面，本文基于FRFRLM考虑充电基础设施的多阶段规划，车辆偏移OD最短路径寻找服务设施以及充电基础设施服务能力有限制的情况下，提出了多阶段规划下同时考虑路径偏移与服务能力的灵活形式流量捕获选址模型(M-CDFRFRLM)，并且以小规模算例用精确算法验证模型的有效性，并对模型中考虑的各种因素进行了灵敏度分析。针对M-CDFRFRLM无法有效求解大规模交通网络问题，本文提出了相应的贪婪算法和遗传算法，并且用小规模算例结果与精确算法结果进行对比分析，实验表明本文提出的贪婪算法和遗传算法求解效果良好。最后，本文用贪婪算法和遗传算法对上海市嘉定区大规模交通网络进行求解。

In order to cope with the environmental crisis, China has committed to achieve peak carbon dioxide emissions before 2030 and achieve carbon neutrality before 2060. The development of electric vehicles plays a critical role in this ambitious strategy. At present, the promotion and application of electric vehicles is mainly limited by short vehicle mileage, long charging time and unreasonable layout of charging infrastructure. Since vehicle mileage and vehicle charging time are difficult to breakthrough in a short period of time, the rational and effective layout of charging infrastructure has become the key factor to promoting the development of the electric vehicle industry.This paper studies the layout of charging infrastructure based on traffic flow demand from the perspective of model application and innovation. At the application level, this paper uses the classic flow refueling location model(FRLM) and the flexible reformulation of flow refueling location model(FRFRLM). To optimize the layout of the charging infrastructure in the Jiading district of Shanghai. To effectively handle the large-scale transportation network this paper proposes an algorithm to solve the charging facility planning problem in the large-scale transportation network in a relatively short time. In view of the lack of comparison between the two models of FRLM and FRFRLM, this paper takes Jiading District, Shanghai as an example to explore the similarities and differences between the two models. The results show that FRLM and FRFRLM are basically the same in terms of flow capture and site selection. However, when dealing with large-scale transportation network problems, the running time of FRFRLM program is much smaller than that of FRLM, which indicates that the model extension of FRFRLM is more meaningful. Finally, in view of the lack of evaluation of the current construction of real charging facilities in cities, this paper proposes an algorithm that can map real charging facilities to the transportation network, which can compare the actual construction situation with the model output, so as to be able to evaluate the construction of charging infrastructure and provide some guidance.At the innovation level, this paper considers the multi-stage planning of charging infrastructure based on FRFRLM, when vehicles deviate from the OD shortest path to find service facilities and the service capacity of charging infrastructure is limited, this paper proposes a multi-stage planning that considers path deviation and service capacity based on the flexible reformulation of flow refueling location model(M-CDFRFRLM), and a small-scale example is used to verify the validity of the model with an accurate algorithm, and a sensitivity analysis is carried out for various factors considered in the model. Aiming at the inability of M-CDFRFRLM to solve large-scale transportation network problems effectively, this paper proposes the corresponding greedy algorithm and genetic algorithm, and compares the results of small-scale examples with the results of the accurate algorithm. Experiments show that our proposed greedy algorithm and genetic algorithm have good results. Finally, we use greedy algorithm and genetic algorithm to solve the large-scale transportation network in Jiading District, Shanghai.