文章编号：1672-5328（2013）03-0027-07

成卫1，袁满荣2，姬利娜3
（1.昆明理工大学云南省高校智能交通系统工程研究中心，云南昆明650093；2.昆明市公安局交通警察支队，云南昆明 650023；3.安阳工学院，河南安阳455000）

摘要：在现有交通资源下，利用 交通信号的动态调控缓解交通拥 堵是一种行之有效的方式。首先 探讨了道路交叉口信号控制的空 间和时间优化思路，在时间优化方 面提出基于粒子群算法的信号配 时优化模型。以昆明市学府路为 例，在分析大量交通流数据的基础 上，根据三相交通流理论，对交通 状态进行划分并提出有针对性的 控制策略。将信号配时优化模型 应用于学府路3 个相邻的关键交 叉口。交通仿真和方案试运行结 果显示，优化前后同步流状态下交 叉口延误平均降低21.0%，车辆排 队长度平均降低12.4%；堵塞状态 下交叉口延误平均降低32.0%，车 辆排队长度平均降低24.9%。这 一结果表明该模型在道路交叉口 信号配时优化中具有合理性和有 效性。

关键词：交通控制；道路交叉口；信 号配时；粒子群；优化模型；控制策 略；仿真评价

中图分类号：U491

文献标识码：A

Signal Timing Model Based on Particle Swarm Optimization:ACase Study of Kunming

ChengWei1, Yuan Manrong2, Ji Lina3
(1. Intelligent Transport System Center of Higher Institutions, Kunming University of Science and Technology, Kunming Yunnan 650093, China; 2. Traffic Police Detachment, Kunming Public Security Bureau, Kunming Yunnan 650023, China; 3. Anyang Institute of Technology, Anyang Henan 455000, China)

Abstract：Dynamic signal control is an effective method to relief traffic congestion, which does not require additional resources. By discussing the signal control improvement in space and time, this paper first develops a signal timing model based on Particle Swarm Optimization (PSO). Taking Xuefu Road in Kunming City as an example, the paper classifies different traffic performance and proposes corresponding traffic control strategies based on the massive data of traffic flow and three-phase signal timing theory. The traffic simulation results on three adjacent intersections on Xuefu Road show an average 21.0% reduction in intersection delay and a 12.4% decrease in queue length under the same flow condition before and after signal timing improvement. Under the traffic congestion, the reduction in average delay is 32.0% and in average queue length is 24.9%. The results demonstrate the efficiency of the model in improving signal timing.

Keywords：traffic control; intersections; signal timing; particle swarm; optimization models; control strategies; simulation evaluation