The traffic flow at short-distance intersections on urban arterial roads is large,and the traffic flow is limited by the length of the entrance road,which is prone to queue overflow. This traffic problem affects the vehicular traffic at the upstream intersection. In this paper,from the perspective of traffic flow operation and traffic wave propagation,the process of vehicle arrival-queuing-departure in short-distance entrance lanes was analysed;the concept of stranded queuing was put forward,and the models of maximum queuing length and stranded queuing length were established. According to the queuing length model and the vehicle arrival and departure diagram,a vehicle average delay model suitable for short-distance intersections was established. Taking the minimum delay of vehicles at the intersection as the goal,the maximum queuing length of the short-distance lane does not exceed the length of the road section and preventing the vehicles from queuing twice as the constraints,an optimization model of signal control at the short-distance intersection was established. Genetic Algorithms are used in the solving process. The intersection of Furong Road and Chengnan Road in Changsha was used as an example for application. The VISSIM software is used for the simulation verification before and after signal optimization. The results show that:the optimized signal control scheme reduces the average queuing at short-distance entrances by 36.9%,the maximum queuing by 42.3%,and the average number of stops at intersections by 21.2%. The number of passing vehicles increased by 19.8%,and the average vehicle delay decreased by 25.8%,which proves that the method proposed in this paper can effectively prevent the queue overflow phenomenon at the short-distance entrance,and improve the traffic capacity and operation efficiency of the intersection.