Research On Coordinated Control Optimization Method Of Urban Intersection Group Based On Discrete Event Simulation Optimization

Reasonable optimization of urban intersections coordinated control can reduce the frequency of parking and the delay time of vehicle and increase the capacity of road.Considering that the size of intersection group is larger than that of arterial and the time-space relationship between them is different,this paper should build an appropriate optimization model to optimize the coordination control of the intersection group.However,most of the existing studies are based on analytical optimization or simulation-based optimization to build an optimization model.Most of the analytical optimization uses the delay time or the number of parking to reflect the situation of vehicles at intersections,which is inaccurate.Moreover,the solving algorithm of analytical optimization model is too complex.Most of the simulation-based optimization research assumes the arrival of vehicle flows fits the specific law and ignores the state-dependent of service time.It's hard to reflect the arrival process of vehicles.A discrete event simulation(DES)model is more efficient than the micro-simulation model and more accurate than macro-simulation model,and the simulation-based optimization method can cope with the complexity of the existing optimization algorithm and accurately describe the arrival law of traffic flow.This paper presents an optimization method for coordinated control of urban intersection group based on DES.The optimization objective is the minimum of average travel time and the maximum of the capacity of intersections.Based on the DES model of intersection group,the optimization method of coordinated control of urban intersection group is proposed.This paper includes three contents: constructing the queuing system at intersection,building the DES model of intersection group and studying coordinated control method.The main research as follows:(1)Firstly,this paper constructs a signalized intersection queuing system.Through the analysis of queuing and running process,the intersection is abstracted into three categories: a road queuing system,an entrance lane queuing system and an exit lane queuing system.Considering the characteristics of phase-type(ph)distribution,this paper assumes the vehicle's arriving fits the phase(PH)distribution,the road service time of the intersections is state-dependent and the left-turn and the straight entrance lane service rate are the saturation.Based on the assumption the intersection queuing system is constructed as the theory basis of a DES model and the relevant parameters are calibrated.This is the basis of the DES model.(2)Secondly,this paper builds a DES model of a single intersection and intersection group.Based on the queuing system of intersection,Simulink is used to construct DES models for road,entrance lane and exit lane.Based on this,the intersection and the intersection group DES model is built.The simulation model is validated and analyzed by a case of the intersection in Kunshan.The analysis proves that the simulation model is accurate to some extent and it can simulate the running process of vehicles at intersections.The simulation model is suitable for the intersections with interval which is less than 1000 meters and unsaturated or adjacent saturated.(3)Thirdly,this paper optimizes the coordinated control of intersection group based on DES.In order to analyze the correlation between intersections,this paper uses an estimation model of correlation degree considering intersection interval,traffic flow and signal period.Then,the model takes the minimum average travel time of vehicles and the maximum capacity of intersections as objectives.In order to construct the model,the paper chose the cycle time and the green light time of each intersection as decision variables.In order to solve the multi-objective optimization function,a SPSA algorithm is proposed.An example of intersection group surrounding Kunshan municipal government is presented for comparative analysis.The result shows that the optimized scheme has shorter average travel time and larger capacity of intersection group,and proves the validity of the established model.