基于混沌理论的短时公交到站时间预测

(1.西藏大学藏文信息技术研究中心,2.西藏大学工学院,西藏 拉萨 850000)

交通工程; 公交运营; 到站时间; 混沌特性; Lyapunov指数

Short-term Bus Arrival Time Prediction Based on Chaos Theory
AN Baokun1*,ZHOU Huanhuan2

(1.Research Center of Tibet Information Technology,Tibet University,2.College of Engineering,Tibet University,Lhasa 850000,China)

traffic engineering; bus operation; arrival time; chaos characteristics; Lyapunov exponent

DOI: 10.6043/j.issn.0438-0479.201710001

备注

为分析公交到站数据的混沌特征及可预测性,选用运行在非公交专用道上连续6日的单线公交到站实测数据,利用能够应用于工程实践的混沌判别手段和预测方法对数据进行分析.结果表明:公交单日到站数据均具有混沌特性,且在工作日和休息日均具有该特性,同时公交多日到站数据同样具有一定的混沌特性证实了混沌特征在公交到站数据中的存在.可见基于混沌理论的预测方法能够对公交到站延误和站点停靠时间的多日数据进行有效预测.

To analyze chaotic characteristics and the predictability of bus arriving data at a single station,we choose measured data of six consecutive days and use the chaotic discriminant analysis and forecasting method that can be applied in engineering practice.Results show that the single day's data of these vehicles exhibit chaotic characteristics.Furthermore,data on both weekdays and weekends as well as on multi-days all exhibit certain chaotic behaviors.The prediction method based on chaos theory can effectively predict the multi-day data of bus arriving delay and dwell times.These results confirm the existence of chaotic characteristics in the bus arrival data,and the chaos theory can be used to predict bus arrival data with better results.