交通运输系统工程与信息 ›› 2009, Vol. 9 ›› Issue (2): 38-43 .

• 智能交通系统与信息技术 • 上一篇    下一篇

快速路入口匝道排队现象分析及其对策

齐驰;侯忠生*;卜旭辉   

  1. 北京交通大学 电子与信息工程学院先进控制系统研究所,北京 100044
  • 收稿日期:2008-04-19 修回日期:2008-08-14 出版日期:2009-04-25 发布日期:2009-04-25
  • 通讯作者: 侯忠生
  • 作者简介:齐驰(1982-),男,河南周口人,博士生.

Analysis on Excessive Queen on Ramp and Its Solution

QI Chi;HOU Zhong-sheng; BU Xu-hui   

  1. Advanced Control Systems Laboratory of the School of Electronics and Information Engineering, Beijing Jiaotong University, Beijing 100044, China
  • Received:2008-04-19 Revised:2008-08-14 Online:2009-04-25 Published:2009-04-25
  • Contact: HOU Zhong-sheng

摘要: 首先分析了快速路入口匝道排队导致地面交通拥堵的现象,然后综合考虑快速路主线交通流密度和入口匝道排队长度两个因素,提出了一种基于神经元自适应PID控制的快速路入口匝道控制策略。与传统的入口匝道调节方法ALINEA相比,新方法将快速路系统和平面道路系统有机的结合起来,仅略微地增加快速路交通流密度,且对快速路的负面影响较小,有效降低了入口匝道排队长度并消除了入口匝道回溢现象与快速路交通拥挤的反复振荡现象,对于实际应用中的随机干扰具有良好的抑制能力,提高了整个交通网络的运行效率。最后通过仿真对新方法进行了说明和验证。

关键词: 神经元自适应PID控制, 快速路, 入口匝道控制, 排队长度

Abstract: The phenomenon that excessive queue on ramp brings traffic congestion on the artery is first analyzed in this paper. Then, it proposes a ramp metering algorithm in which the neuron adaptive PID control algorithm is applied to tune controller parameters in real time, with both the on-ramp queue and the desired mainline traffic density are considered. Compared with the traditional feedback ramp metering method, the proposed method effectively reduces the on-ramp queue, creates smaller negative effect on the freeway and increases efficiency of the whole traffic network as well. The simulation analysis indicates that the presented method is effective to improve the on-ramp control.

Key words: neuron adaptive PID control, freeway, ramp metering, on-ramp queue

中图分类号: