The development of smart city is of strategic significance to the realization of modern society in China, and rail transit network is an important part of urban development. Currently, China's urban rail transport is at the bottleneck stage, and many medium-sized cities are facing a sudden heavy passenger congestion situation. Based on the next-generation innovation of the knowledge society, smart cities will maximize the use of new generation information technology in all fields of the city, and realize advanced urban informatization that realizes deep integration of informatization, industrialization, and informatization and helps to reduce urbanization and "disease in big cities." Improve the quality of urbanization, realize refined and dynamic management, and enhance the effectiveness of urban management and improve the quality of life of citizens. In order to ensure the safety of passengers, passenger flow control and optimization and upgrading of rail transit network become urgent. However, there are few researches in this field in China, so this paper will focus on the characteristics of sudden passenger flow and congestion control based on smart city rail transit network. First, in this paper, we systematically explained the relationship between smart cities and smart transportation and analyzed the need for sudden passenger flow control in urban railway transportation according to its characteristics. According to a study on the characteristics of sudden passenger flow, this paper believes that passenger flow comes primarily from large-scale activities, commuters, and individual emergencies, but is still predictable and controllable. For this reason, this paper studies the control scheme of sudden passenger flow congestion, which fully combines the characteristics of sudden passenger flow, and according to the characteristics of the treatment. This scheme is safe and reliable and has the advantage of controlling congestion from the source. In order to further test the actual effect of the passenger flow control scheme, the simulation experiment is carried out in the end of this paper. According to experimental data analysis, the passenger flow control method of this paper can effectively improve the existing rail transport congestion control in the case of sudden large-scale passenger flow, and the effect is remarkable, and most of them. Suitable for urban rail transport networks.