The road section information and traffic flow information of urban road traffic are vital to the safe and efficient operation of road traffic. During the traffic peak period, by controlling the key road sections, we can achieve the entire road traffic network fully controllable. In order to find the key road sections in the road network, this paper transforms the intersection-node model of the road network into the section-node model of the road network. Based on the road section information and traffic flow information, a congestion coefficient is proposed to measure the vehicle congestion degree of the road traffic network, and it is used as the edge weight of the section-node model. Finally, the key road section identification algorithm is used to identify the key road sections of the road traffic network. Taking the main urban roads in Huanggu District of Shenyang as an example, a network model with congestion coefficient as weight is established. The number of key road sections identified according to the method in this paper is 14, accounting for about 14.3% of the total sections of the road network, which has low control cost, and most of them are from north to south and from west to east. Among them, 8 road sections are distributed in the top five of the real-time road congestion ranking in Huanggu District, accounting for about 57.1% of the total number of key road sections. This shows that the key road sections given in this paper are more distributed in the road sections with congested traffic conditions, which is in line with the actual situation.