计算机工程与应用 ›› 2016, Vol. 52 ›› Issue (6): 31-34.

• 理论与研发 • 上一篇    下一篇

基于MF-DFA的交通流量多重分形研究

王俊丽,张辰彦,何红弟,林国龙   

  1. 上海海事大学 物流研究中心,上海 201306
  • 出版日期:2016-03-15 发布日期:2016-03-17

Multifractal analysis of traffic flow based on MF-DFA

WANG Junli, ZHANG Chenyan, HE Hongdi, LIN Guolong   

  1. Logistics Research Center, Shanghai Maritime University, Shanghai 201306, China
  • Online:2016-03-15 Published:2016-03-17

摘要: 针对交通流时间序列的波动性变化特征,采用多重分形消除趋势波动分析法(Multi-Fractal Detrended Fluctuation Analysis,MF-DFA),对上海内环高架吴中路上匝道和下匝道的交通流量时间序列进行研究。结果表明,上匝道和下匝道的交通流量都具有长程相关性和多重分形特性。通过对比上匝道和下匝道的多重分形特性,发现下匝道交通流量的多重分形特征明显强于上匝道,表明下匝道的交通流量具有更显著的波动特性。这些研究成果为城市交通流时间序列的非线性研究提供了重要的理论依据。

null

关键词: 交通流, 多重分形消除趋势波动分析法, 多重分形谱

Abstract: Due to complexity and fluctuation in traffic flow time series, Multi-Fractal Detrended Fluctuation Analysis(MF-DFA) method is applied to investigate the fractal behavior of traffic volume in urban expressway. The traffic volume data on on-ramp and off-ramp of Wuzhong road in Shanghai inner ring viaduct are selected respectively and corresponding results are compared with each other. The results show that multifractality and the long-range anti-persistent behavior exist on both on-ramp and off-ramp situations. Additionally, the degree of multifractality on off-ramp is found to be higher than that on on-ramp, implying that long-range correlation behaviors are more obvious on off-ramp situation. These findings from real observations are advantage for studying the fluctuation characteristics of traffic flow time series.