基于投影寻踪的快速路交织区交通状态判别方法  

A Method to Identify Traffic States of Weaving Segments on Urban Freeways Using Projection Pursuit

作　　者：苏俊杰[1] 柴干[1] 季文韬 SU Junjie;CHAI Gan;JI Wentao(Intelligent Transportation System Research Center,Southeast University,Nanjing 211189,China)

机构地区：[1]东南大学智能运输系统研究中心,南京211189

出　　处：《交通信息与安全》2019年第2期114-119,134,共7页Journal of Transport Information and Safety

基　　金：国家自然科学基金项目(61573106)资助

摘　　要：为了准确判别城市快速路交织区的交通状态,实现交通控制策略的优化决策,基于投影寻踪模型与k-means聚类算法,研究了一种新的交通状态判别方法。以交通状态的量化分析为目标,考虑投影寻踪模型的特性,定义了交通状态系数;根据类内聚集度与异类间散度的分析,建立了聚类效果评价系数表达式;应用推导的改进式遗传算法,结合k-means聚类算法,计算获得最优投影方向与聚类中心;应用最优投影方向将新观测的交通流数据转化为交通状态系数,判定欧式距离最小的聚类中心,获得相对应的交通流状态。新方法克服了传统方法对专家经验的依赖性,解决了熵权法对小概率事件信息熵的过量估计问题,并改进了投影寻踪模型的聚类效果评价系数。仿真实验结果表明,新方法状态判别准确率为96.63%,较神经元网络和决策树算法分别提高了5.58%和7.01%,能够准确判别交织区交通流状态。In order to accurately identify traffic states of urban expressways for optimizing traffic control decision,a new identification method for traffic states is proposed based on projection pursuit model and k means algorithm.The new method regards quantitative analysis of traffic states as the goal.Considering characteristics of projection pursuit model,coefficients of traffic states are defined.An equation for clustering effect evaluation coefficient is developed based on an analysis of concentration degree and dispersion degree.An improved genetic algorithm and k means clustering algorithm are applied to calculate the optimal projection direction and cluster center.Traffic data is converted into traffic state coefficient by optimal projection direction,to identify the minimal Euclidean distance,and obtain the corresponding traffic flow states.The new method less relies on expert experiences than traditional methods;and can solve the problem of overestimation on small probability events when using entropy method;and improves clustering effect evaluation coefficient from projection pursuit.Simulation results show that accuracy of identification of the new methodology is 96.63%,which is 5.58%and 7.01%higher than Neural Network and Decision Tree,respectively.The results verify that the new method can accurately identify traffic states of weaving areas.

关 键 词：交通工程 快速路交织区 投影寻踪 交通状态 改进式遗传算法 

分 类 号：U491[交通运输工程—交通运输规划与管理]