交通流实测数据的时间序列分析

Time Series Analysis for the Measured Data of Traffic Flow

提出了一个建立车速、车头间距等交通流参数大样本时间序列的方法,依据实际交通录像获得总数据量超过8万条的4个时间序列样本,对它们进行分形指标的计算和分析.研究发现:Hurst指数和平均循环周期的计算结果较为客观,受数据采集过程中的人为因素影响较小;交通流时间序列存在趋势的正相关特征,并随着路段拥挤程度提高而加强;构建的时间序列适用于短期或较长时间的交通流预报和诱导.

A method is proposed to establish the time series based on the large samples of traffic flow parameters such as velocity, headway, etc. Four time series amounted to over 80 000 data in total are established from the actual traffic video recordings, and the fractal indexes of them are calculated and analyzed. It is found： the Hurst exponents and the average cycle times are more objective, affected less by the human factors in the data collection procedure; the traffic flow demonstrates positively correlated trend, and this trend is increasing with the crowdedness of the road; the established time series can be applied for the short-term or long-term traffic flow prediction and management.