| 基于优化SVM的城市快速路网交通流状态判别 |
| |
| 引用本文： | 董春娇,邵春福,熊志华.基于优化SVM的城市快速路网交通流状态判别[J].北方交通大学学报,2011(6):13-16,22. |
| |
| 作者姓名： | 董春娇 邵春福 熊志华 |
| |
| 作者单位： | [1]田纳西大学交通研究中心,田纳西37996 [2]北京交通大学城市交通：复杂系统理论与技术教育部重点实验室,北京100044 |
| |
| 基金项目： | 国家自然科学基金资助项目（51178032）;中国发展研究基金会2009年度“通用汽车中国发展研究青年奖学金”项目资助;北京交通大学优秀博士生科技创新基金资助项目（141082522） |
| |
| 摘 要： | 以交通流率、速度和占有率为输入参数，采用交叉验证法优化模型惩罚参数C和核函数参数γ，建立以径向基为核函数的支持向量机模型，判断道路断面交通流状态；结合设计的道路网综合状态指数，依据自由流、拥挤流和阻塞流状态下占有率划分区间，构建城市快速路网交通流状态判别方法；最后以某一区域路网为例，进行了实证性研究．结果表明：该方法对道路断面交通流状态判别精度可达92．22％；同时能够实现道路网范围内对自由流、拥挤流和阻塞流状态的判别，判别精度可达86．67％．
|
| 关 键 词： | 交通流状态判别 支持向量机模型 道路网综合状态指数 交叉验证法 |
Identification of traffic states with optimized SVM method on urban expressway network |
| |
| DONG Chunjiao,SHAO Chunfu,XIONG Zhihua.Identification of traffic states with optimized SVM method on urban expressway network[J].Journal of Northern Jiaotong University,2011(6):13-16,22. |
| |
| Authors: | DONG Chunjiao SHAO Chunfu XIONG Zhihua |
| |
| Affiliation: | 1. Center for Transportation Research, the University of Tennessee, Tennessee 37996, USA; 2. MOE Key Laboratory for Urban Transportation Complex Systems Theory and Technology l Beijing Jiaotong University, Beijing 100044, China) |
| |
| Abstract: | A support vector machine model based on Radial Basis Function is developed to achieve the end of traffic states identification. The proposed model used traffic flow, speed and occupancy for the input data, and adapted the K-CV method to optimize the parameters c and γ in the model. The method combined with the road network integrated state index,which is designed according to occu- pancy split interval of free traffic, oongested traffic and jam traffic,for identifying traffic states on urban expressway network. The empirical research shows that the proposed method can not only realize the traffic states identification with accuracy 92.22% to the road section, but also achieve the identification to free traffic,oon- gested traffic and jam traffic for the range of road network with the accuracy 86.67%. |
| |
| Keywords: | traffic states identification support vector machine model road network integrated stateindex K-fold cross validation（K-CV） method |
| 本文献已被 维普 等数据库收录！ |
