基于遗传算法的交通信号控制多目标优化  

Traffic signal control multi-object optimization based on genetic algorithm

作　　者：李瑞敏[1] 陆化普[1] 

机构地区：[1]清华大学交通研究所,北京100084

出　　处：《长安大学学报（自然科学版）》2009年第3期85-88,共4页Journal of Chang’an University(Natural Science Edition)

基　　金："十五"国家重点攻关项目(2002BA404A20B);北京市科技计划项目(gyykw05070006)

摘　　要：针对城市交通信号控制面临的多目标优化问题,提出了利用遗传算法实现多目标优化的方法。分析了城市交通信号控制多目标优化的本质及特点,采用基于遗传算法的多目标优化方法,提出了信号控制多层模糊控制模型,以平均延误和停车次数作为优化目标,采用遗传算法中的随机权重方法来进行该模型的多目标综合优化,给出了各模型参数的计算方法和优化步骤,最后进行了仿真试验。仿真结果表明,给出的多目标优化方法优于传统的多目标固定权重方法,可以获得更好的交通信号控制效果。Aimed at the problem of multi-object optimization solution on the problem by using genetic algorithm is presented. The essential and characteristics of multi-object optimization in urban traffic signal control are analyzed. Then, the genetic algorithm （GA） method is applied to multi-object optimization and a multi-layer fuzzy traffic signal control model is proposed. Regarded average delay and stop times as optimal object, random-weight approach of GA is used to optimize the multi-object of the model, the optimization steps and calculation method of some parameters and the determination method of random weight is presented. At the end, the performance of this optimization method based on GA is compared to that with traditional fixed weight at a simulated four-approach intersection. The result indicates that the optimization method presented in this paper is effective in reducing average delay and stop times at an intersection. 2 figs, 10 refs.

关 键 词：交通工程 交通信号控制 多目标优化 遗传算法 

分 类 号：U491[交通运输工程—交通运输规划与管理]