论文部分内容阅读
针对短时交通流预测中忽视交通流所处交通状态的转变及所处的状态。首先简要介绍了相干递归图及相干定量递归分析,以1分钟为间隔的实测数据为例,通过相干递归图分别可视化常发性拥挤和偶发性拥挤中时间占有率和流量的递归特性,然后运用相干定量递归分析分别确定了常发性拥挤和偶发性拥挤交通状态的转变时刻,得到了不同交通状态的统计特征值,并做出分析。结果表明将交通流划分为四个状态更具合理性,同时各状态的统计特性分析对短时交通量预测及交通拥堵机理研究都有重要的意义。
In view of the short-term traffic flow forecast traffic flow in the neglect of the traffic flow in the state and the state. Firstly, the coherent recursive graphs and coherent quantitative recursive analysis are briefly introduced. Taking the 1-minute intervals of measured data as an example, the coherence recursive graphs are used to visualize the recursive characteristics of time share and traffic in the frequent and occasional crowds respectively. Coherent quantitative recursive analysis respectively determines the changing moments of the frequent traffic congestion and the occasional traffic congestion, and obtains the statistical eigenvalues of different traffic states and makes the analysis. The results show that it is more reasonable to divide the traffic flow into four states, and the statistic characteristics of each state are of great significance for the study of short-term traffic volume and traffic congestion mechanism.