针对现有限流设施与策略智能化程度不高,灵活性较差的问题,提出一种基于光流特征描述子的站点限流设施优化方法.首先,根据枢纽内场景特点,设置感兴趣区域(region of interest,ROI),从而降低后续操作的计算量,提高算法的执行效率;然后,...针对现有限流设施与策略智能化程度不高,灵活性较差的问题,提出一种基于光流特征描述子的站点限流设施优化方法.首先,根据枢纽内场景特点,设置感兴趣区域(region of interest,ROI),从而降低后续操作的计算量,提高算法的执行效率;然后,在建立光流特征描述子的基础上,对图片序列进行特征分析;最后,基于人群聚集特征,对经典单分类支持向量机进行调整,并实现超负荷状态的检测.实验结果表明,提出的方法能够对站台人群状态进行准确检测,有效增强限流设施的自动化水平,为轨道交通站点客流组织与管理提供数据支撑和理论依据.展开更多
Metro passenger flow control problem is studied under given total inbound demand in this work,which considers passenger demand control and train capacity supply.Relevant connotations are analyzed and a mathematical mo...Metro passenger flow control problem is studied under given total inbound demand in this work,which considers passenger demand control and train capacity supply.Relevant connotations are analyzed and a mathematical model is developed.The decision variables are boarding limiting and stop-skipping strategies and the objective is the maximal passenger profit.And a passenger original station choice model based on utility theory is built to modify the inbound passenger distribution among stations.Algorithm of metro passenger flow control scheme is designed,where two key technologies of stopping-station choice and headway adjustment are given and boarding limiting and train stopping-station scheme are optimized.Finally,a real case of Beijing metro is taken for example to verify validity.The results show that in the three scenarios with different ratios of normal trains to stop-skipping trains,the total limited passenger volume is the smallest and the systematic profit is the largest in scenario 3.展开更多
文摘针对现有限流设施与策略智能化程度不高,灵活性较差的问题,提出一种基于光流特征描述子的站点限流设施优化方法.首先,根据枢纽内场景特点,设置感兴趣区域(region of interest,ROI),从而降低后续操作的计算量,提高算法的执行效率;然后,在建立光流特征描述子的基础上,对图片序列进行特征分析;最后,基于人群聚集特征,对经典单分类支持向量机进行调整,并实现超负荷状态的检测.实验结果表明,提出的方法能够对站台人群状态进行准确检测,有效增强限流设施的自动化水平,为轨道交通站点客流组织与管理提供数据支撑和理论依据.
基金Projects(RCS2015ZZ002,RCS2014ZT25)supported by State Key Laboratory of Rail Traffic Control&Safety,ChinaProject(2015RC058)supported by Beijing Jiaotong University,China
文摘Metro passenger flow control problem is studied under given total inbound demand in this work,which considers passenger demand control and train capacity supply.Relevant connotations are analyzed and a mathematical model is developed.The decision variables are boarding limiting and stop-skipping strategies and the objective is the maximal passenger profit.And a passenger original station choice model based on utility theory is built to modify the inbound passenger distribution among stations.Algorithm of metro passenger flow control scheme is designed,where two key technologies of stopping-station choice and headway adjustment are given and boarding limiting and train stopping-station scheme are optimized.Finally,a real case of Beijing metro is taken for example to verify validity.The results show that in the three scenarios with different ratios of normal trains to stop-skipping trains,the total limited passenger volume is the smallest and the systematic profit is the largest in scenario 3.
