基于离散事件仿真的城市交叉口群协调控制优化方法  

Coordinated Control Optimization Method of Urban Intersection Groups Based on Discrete Event Simulation

作　　者：方岩 齐心 李意[1] 赵斌[1] FANG Yan;QIN Xin;LI Yi;ZHAO Bin(School of Traffic and Transportation,Southwest Jiaotong University,Chengdu 610031,China)

机构地区：[1]西南交通大学交通运输与物流学院,成都610031

出　　处：《综合运输》2020年第6期61-68,共8页China Transportation Review

摘　　要：为描述交叉口车辆到达的时变性及路段服务车辆的状态相关性,本文提出了交叉口Mt/G (x)/C/C型排队网络系统。同时为解决现有关于城市交叉口群协调控制的解析优化模型及算法过于复杂的问题,提出基于离散事件仿真的城市交叉口群协调控制优化方法。该提出的方法以车辆平均行程时间最短和交叉口群通行能力最大为优化目标,以信号公共周期、绿灯时长、交叉口饱和度及绿信比为约束条件,并基于同步扰动逼近(SPSA)算法收敛速度快的优势,采用该算法求解优化模型。为验证模型有效性,选取昆山市一交叉口群进行仿真分析。结果表明,仿真模型准确性较高、稳定性较好。同时,采用基于离散事件仿真优化的城市交叉口群协调控制优化方法,车辆的平均行程时间较小且交叉口群通行能力较大。To describe the time-varying arrival of vehicles at intersections and the state-dependent service time of vehicles on road sections,a Mt/G(x)/C/C queuing system was proposed.To cope with the complexity of the existing optimization models and algorithms for coordinated control of urban intersection groups,an optimization method based on discrete event simulation was proposed.Taking the minimum of average travel time and the maximum of the capacity of intersections as optimization goals,and constrained by the common signal cycle,green time,intersection saturation,and green ratio,the optimization method for coordinated control of urban intersection groups was proposed.Due to the fast convergence rate,the simultaneous perturbation stochastic approximation(SPSA)algorithm was adopted to solve the optimization model.The discrete event simulation model and the simulation-based optimization model were validated and analyzed by a case of the intersection group in Kunshan.The results show that the discrete event simulation model is accurate and stable.Meanwhile the simulation-based optimized scheme has shorter average travel time and larger capacity of the intersection group.

关 键 词：交通信号控制 排队系统 离散事件仿真 仿真优化 SPSA 

分 类 号：U491.23[交通运输工程—交通运输规划与管理]