Optimization Method Of Traffic Signal Coordinated Control Considering Regional Effect

The coordinated control of urban traffic signals is an important means to alleviate traffic network congestion and improve the operation efficiency of the traffic network.In the meantime,the coordinated control of urban traffic signals not only studies the signal control of trunk roads or single intersections in the road network,but also considers the overall operation of the road network effectiveness.Consequently,it is completely difficult to analyze and study the entire traffic network directly because of the enormous urban traffic network system.Subnetworks partition of regional traffic signal control is an effective solution to solve this problem.According to the divided sub-networks,when signal timing is applied to the intersections in each sub-networks,it is often more likely to neglect the problem of congestion of sub-network boundary intersections.At the same time,it is easy to divide the regional intersections into trunk lines or single intersections for independent signal optimization control,and neglect the regional effect of the traffic network.Therefore,this thesis conducts research on traffic signal control based on the above three aspects.The main work of the thesis is as follows:(1)Aiming at the difficulty of effectively dividing the complex traffic network by the existing partitioning method,an improved Newman fast algorithm suitable for traffic subnetwork partitioning is proposed.Because Newman fast algorithm is a local search-based algorithm,the result set is often a local optimal solution.In order to overcome this drawback,this thesis proposes a Newman fast algorithm based on community contribution degree for traffic network.Although the result of the search may not be the global optimal solution,it can overcome the disadvantages of Newman’s fast algorithm local search to some extent.The experimental calculation results show that the algorithm can effectively enhance the modularity value of the network after partitioning and improve the rationality of the partitioning results.(2)In view of the deficiencies in the existing research,the problem of sub-network boundary congestion is often neglected.This thesis proposes a signal coordination control optimization model to prevent sub-network boundary congestion.The optimization model takes the vehicle delay and signal period difference between adjacent sub-networks as the optimization target,and constructs a multi-objective optimization model for coordinated control between sub-networks.The effectiveness of the optimization model and scheme is verified by experiments.(3)At present,most researches of the regional signal coordination control are to divide the sub-networks into trunk lines or single intersections,perform trunk optimization and single point optimization respectively.In order to overcome the shortcomings of this method,this thesis proposes a regional delay model,which mainly considers the vehicle delay in the whole region,rather than a certain trunk line or an intersection,which can better reflect the regional effect.The optimization model not only considers the different delays of the vehicle,but also considers the fluctuation delay caused by the random arrival of the vehicle.The simulation results show that the optimization model has better optimization performance under non-low saturation traffic conditions.