计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (10): 13-16.

• 博士论坛 • 上一篇    下一篇

城市过饱和路网的偏好多目标相容优化控制

陈 娟1,袁长亮2   

  1. 1.上海大学 悉尼工商学院,上海 201800
    2.交通部公路科学研究院,北京 100088
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-04-01 发布日期:2011-04-01

Urban oversaturated traffic network control based on preference multi-objective compatible optimization control

CHEN Juan1,YUAN Changliang2   

  1. 1.Syndey Institute of Language and Commerce,Shanghai University,Shanghai 201800,China
    2.Research Institute of Highway Ministry Transport,Beijing 100088,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-04-01 Published:2011-04-01

摘要:

针对城市过饱和路网的交通信号控制问题,提出将总延误分为主线路段延误和次线路段延误的划分方法,将控制问题描述为冲突的多目标控制。针对冲突多目标控制问题中最优解不唯一,而传统优化方法中一次运行只能得到一个最优解的问题,提出了一种基于偏好的相容优化控制算法:利用偏好信息动态指引寻优的方向,在偏好区域内获得更多有价值的解;提出基于偏好的目标选择函数,保证控制解在偏好区域内的稳定性。在仿真环境中对一个11个交叉口的城市过饱和路网进行实时控制,结果表明,提出的偏好相容控制算法的效果优于定时控制方案。

关键词: 相容优化控制, 多目标优化, 偏好信息, 过饱和路网, 交通控制

Abstract: To solve the traffic control problem on oversaturated linear network,the total delay is divided into two parts:the feeding delay and the non-feeding delay,and the control problem is formulated as a conflicted multi-objective control problem.According to the character that the optimal point is not single in the conflict multi-objective control problem and optimal solutions cannot be simultaneously obtained by traditional optimization methods in a single run,a new compatible control algorithm based on preference is proposed,which incorporates user’s preference information into optimization process for obtaining dense Pareto solutions in preference region,and a new selection function is defined to make control objectives stabilized in this region.The proposed compatible optimization control algorithm is used to solve the oversaturated traffic linear network control problem in a core area of seven junctions under the simulation environment.It is proved that the proposed algorithm can handle the oversaturated linear network control problem effectively than the fixed-time control method.

Key words: compatible optimization control, multi-objective optimization, preference information, oversaturation network, traffic control