中文核心期刊
CSCD来源期刊
中国科技核心期刊
RCCSE中国核心学术期刊

重庆交通大学学报(自然科学版) ›› 2016, Vol. 35 ›› Issue (3): 105-109.DOI: 10.3969/j.issn.1674-0696.2016.03.22

• 交通运输工程 • 上一篇    下一篇

基于时空特征分析的短时交通流预测模型

田保慧,郭彬   

  1. (河南交通职业技术学院 交通信息工程系,河南 郑州 450000)
  • 收稿日期:2014-12-16 修回日期:2015-03-09 出版日期:2016-06-20 发布日期:2016-06-20
  • 作者简介:第一作者:田保慧(1975—),女,河南郑州人,副教授,主要从事交通信息化方面的研究。E-mail:18341615@qq.com。
  • 基金资助:
    河南省交通运输厅科技计划项目(2014G21)

A Short-Term Traffic Flow Prediction Model Based on Spatio-Temporal Characteristics Analysis

TIAN Baohui, GUO Bin   

  1. (Department of Traffic Information Engineering, Henan Communication Vocational Technology College, Zhengzhou 450000, Henan, P.R.China)
  • Received:2014-12-16 Revised:2015-03-09 Online:2016-06-20 Published:2016-06-20

摘要: 交通流预测的实时性和准确性直接影响到交通流诱导系统的高效性,是智能交通领域研究的热点。为了进一步提高短时交通流预测的精度,提出一种基于时空特征分析的短时交通流预测模型。在分析路段时空相关性的基础上,利用云模型改进的遗传算法对支持向量机的参数进行优化,得到最优的支持向量机模型,并实现短时交通流预测。以长春市局部路网的实测数据为基础,验证了所提出模型的有效性和可行性。

关键词: 交通运输工程, 交通量预测, 时空特征分析, 云模型, 遗传算法, 支持向量机

Abstract: The real-time and accuracy of traffic flow prediction directly affect the efficiency of traffic flow guidance system, which is a hot issue of intelligent transportation system research. In order to improve the accuracy of short-term traffic flow forecasting further, a short-term traffic flow prediction model based on spatio-temporal characteristics analysis was proposed. On the basis of spatio-temporal correlativity analysis of section, the parameters of support vector machine (SVM) were optimized by using the genetic algorithm improved by cloud model. At last, the optimal SVM model was obtained, and it realized the short-term traffic flow prediction. Based on the measured data of local road network in Changchun city, the feasibility and effectiveness of the proposed model were verified.

Key words: traffic and transportation engineering, traffic flow forecasting, spatio-temporal characteristics analysis, cloud model, genetic algorithm, support vector machine

中图分类号: