论文部分内容阅读
选取交叉口进口饱和度和路段平均行程速度作为路网状态检测的基本参数,采用小波包变换的时频高分辨率特性,以能量分析方法识别进口饱和度和路段平均行程速度的突变与异常状况,并定义了交通状态系数来定量描述交通状态变化,设计了基于小波分析的交通状态检测算法,并采用贝叶斯算法对交通状态进行预测。仿真分析结果表明:小波包变换可有效识别节点能量分布的突变区间,据此可准确判别交通状态发生变化的时段;当采样数据的模极大值点为200～243时,此段节点能量变化比较剧烈,信号在此出现突变,由较平稳向非平稳状态变化,对应的路段交通状态系数大于0.300h.km-1,为拥挤状态。该方法原理简单,检测响应时间短,检测结果可靠。
Select the intersection of import saturation and the average travel speed of the road section as the basic parameters of the road network status detection, using wavelet packet transform time-frequency high-resolution characteristics of energy analysis method to identify import saturation and section of the average travel speed of the mutation and abnormal conditions , And defines the traffic status coefficients to quantitatively describe the traffic status changes. The traffic status detection algorithm based on wavelet analysis is designed, and the traffic status is predicted by Bayesian algorithm. The simulation results show that the wavelet packet transform can effectively identify the sudden change of the energy distribution of the nodes, and thus can accurately determine the period when the traffic state changes. When the maximum of the sampled data is 200-243, More intense, the signal in this mutation, from a more stable to non-stationary state changes, the corresponding road traffic factor greater than 0.300h.km-1, for the crowded state. The method has the advantages of simple principle, short detection response time and reliable detection result.