引用本文:邝先验,许伦辉,黄艳国.交通信号公交优先控制策略及智能控制方法[J].控制理论与应用,2012,29(10):1284~1290.[点击复制]
KUANG Xian-yan,XU Lun-hui,HUANG Yan-guo.Traffic signal bus-priority control strategy and intelligent control method[J].Control Theory and Technology,2012,29(10):1284~1290.[点击复制]
交通信号公交优先控制策略及智能控制方法
Traffic signal bus-priority control strategy and intelligent control method
摘要点击 3028  全文点击 2142  投稿时间:2011-07-20  修订日期:2012-04-09
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/j.issn.1000-8152.2012.10.CCTA110848
  2012,29(10):1284-1290
中文关键词  交通控制  公交优先  模糊控制  变论域  粒子群优化神经网络
英文关键词  traffic control  bus-priority  fuzzy control  variable universe  particle swarm optimization neural network
基金项目  国家自然科学基金资助项目(60664001, 51268017, 61263024); 江西省自然科学基金资助项目(2010GQS0076).
作者单位E-mail
邝先验 华南理工大学 土木与交通学院
江西理工大学 电气工程与自动化学院 
xianyankuang@163.com 
许伦辉* 华南理工大学 土木与交通学院 lhxu@scut.edu.cn 
黄艳国 华南理工大学 土木与交通学院
江西理工大学 电气工程与自动化学院 
 
中文摘要
      针对公交优先交通信号控制问题, 研究了公交优先的信号控制策略, 提出了一种变论域模糊神经网络公交优先智能控制方法. 提出了基于相位优先度值的公交优先相位选择方法, 并给出了其数学描述. 建立了绿灯时间的3层模糊控制模型, 分别为红灯排队疏散时间、绿灯延长时间和论域调节因子模糊控制器, 其中红灯排队疏散时间和绿灯延长时间两个模糊控制器的输出变量均采用变论域, 论域的变化考虑混合交通、天气情况、车流转向等因素由论域调节因子模糊控制器确定. 模糊控制器采用粒子群优化神经网络实现. 仿真结果表明该方法具有较好的公交优先控制效果.
英文摘要
      We investigate the intelligent control for the traffic signal with bus-priority strategy, and propose a control method with variable-universe fuzzy neural network. This strategy decides the bus-priority phase based on the prioritydegree, of which the mathematical description is developed. A three-layer fuzzy controller for the green-light time is built, that are queuing-evacuation time fuzzy controller, green-light extension time fuzzy controller, and the fuzzy controller for the regulation factor of the universe. Variable universes are employed for the output of the queuing evacuation time fuzzy controller and the output of the green-light extension time fuzzy controller. The variation of the universe reflects the mixed traffic flow, the weather and the traffic turning. The three-layer fuzzy controller is implemented by using the particle-swarm optimization neural network. Results show that the proposed method has good bus-priority control effects.