论文部分内容阅读
为了对城市轨道交通枢纽通道内的集聚型异常事件进行合理的疏导和客流组织,保障城市轨道交通枢纽的安全、高效运行,本文提出了一种通道内行人集聚型异常事件的自动识别算法。该算法首先通过对通道客流基础数据平稳性和突变性的分析,创建了一种兼具平稳性和突变性特征的新数据类型,然后基于双截面客流数据设计了自动识别算法的关键参数—偏移空间差值。最后通过对关键参数变化特征的分析,建立了通道行人集聚型异常事件自动识别算法。仿真试验结果显示:该算法的检测精度为100%,反应时间均值为65 s,表明该算法对通道行人集聚事件有极强的自动检测能力和较短的反应时间。
In order to rationally divert agglomeration anomalies and passenger flow in urban rail transit hubs and ensure the safe and efficient operation of urban rail transit hubs, this paper presents an automatic recognition algorithm for pedestrian agglomeration anomalies in the tunnel. The algorithm first creates a new dataset with both stationary and abrupt characteristics by analyzing the stability and mutation of the baseline data of the passenger flow. Then, the key parameters of the automatic recognition algorithm are designed based on the double-section passenger flow data - partial Move space difference. Finally, through the analysis of the changing characteristics of the key parameters, an automatic recognition algorithm of pedestrian agglomeration anomalies is established. Simulation results show that the proposed algorithm has a detection accuracy of 100% and a mean response time of 65 s, which shows that the proposed algorithm has strong automatic detection ability and short response time to the pedestrian agglomeration event.