Multi-objective Optimization Of Signal Timing At Network Congestion Intersection

As the basis of urban traffic signal control system and one of the essential elements of traffic control management, the optimization design of intersection signal timing exerts great influences on solving traffic problems such as relieving congestion, normalizing order, reducing pollution and improving operating efficiency.Based on mixed and crowded traffic of our country, this paper presents two multi-objective signal timing optimization models:one for the single intersection and the other for the network optimization design. The overall research and results are as follows:1. The control parameters of intersection signal timing, the evaluation index of traffic signal control, traffic signal control systems and the user equilibrium distribution network model are introduced respectively.2. Taken mixed and crowded single intersections as research targets, the current paper analyzes the equilibrium of motor vehicle and pedestrian comprehensively. It displays a multi-objective signal timing optimization model that takes average vehicle delay, average pedestrian delay and traffic capacity as optimization indexes, and saturation degree as restriction, and the impact of pedestrian violation on the vehicle delay is also taken into account. The genetic algorithm is also proposed to solve the optimization model. In order to testify the effectiveness of the improved model, this paper, taking symbolic intersections in Chongqing as examples, employs field survey data to upgrade the model.3. In view of the current network traffic, the paper presents a bi-level programming model of signal timing combined with flow distribution. The lower level is the user equilibrium distribution network model and the upper one is a multi-objective signal timing optimization model, which aims at achieving minimum driving time and minimum number of stops. The heuristic algorithm and genetic algorithm are proposed to solve the model. At last, by employing the network of the Chinese character "田" as calculation examples, the current paper displays the integrated timing optimization of intersections in the network. The optimization results reveal that this model effectively reduces the overall delay and number of stops.