基于突发事件影响力传播的双向人流疏散仿真

发布时间：2018-01-13 10:44

本文关键词：基于突发事件影响力传播的双向人流疏散仿真　出处：《计算机应用》2017年05期 　论文类型：期刊论文

【摘要】：当突发事件发生时,行人行走行为会因为突发事件本身以及突发事件在人群中的传播而改变。对于突发事件影响下的行人行走特征的研究能够提高人流疏散的效率。针对现有研究中数据获取方式的不足,对真实行人场景视频进行图像处理,提取相关数据后分析了无突发事件时行人一般行走特性。针对突发事件下的行人流,利用k-邻近算法和合力的思想描述了突发事件的影响传播和突发事件下行人流的自组织现象,并由此提出一种新的元胞自动机模型,该模型中的行人元胞会受到正常行走、突发事件、安全标识这三个因素所抽象产生的三个作用力的影响。利用模型对突发情况下的双向人流疏散进行仿真,实验结果表明,当安全标识的距离为0、10、20个元胞时,在小范围行人通道中安全标识分布的距离对人群疏散作用不明显;通过对人群间是否存在影响力的研究发现,疏散的效果主要受到附近行人对突发事件传播的影响;突发事件的影响程度太大或影响范围过小都会引发拥堵,不利于人群的疏散。仿真结果与真实世界中的双向行人流疏散情况基本吻合。
[Abstract]:When an emergency happens. Pedestrian walking behavior will be changed by the sudden events themselves and the spread of emergencies in the crowd. The study of pedestrian walking characteristics under the influence of emergencies can improve the efficiency of pedestrian evacuation. Lack of access to data. Real pedestrian scene video image processing, extraction of relevant data after the analysis of pedestrian walking characteristics without emergencies. For emergency pedestrian flow. In this paper, the influence propagation of sudden events and the self-organization of downlink flow are described by using k-neighbor algorithm and resultant force, and a new cellular automata model is proposed. The pedestrian cells in the model will be affected by the three forces abstracted by normal walking, emergency events and safety identification. The model is used to simulate the two-way passenger evacuation in the emergency situation. The experimental results show that when the distance of safety mark is 0 ~ 10 and 20 cells, the distance of safety mark distribution in a small pedestrian passageway has no obvious effect on crowd evacuation. It is found that the effect of evacuation is mainly affected by the propagation of unexpected events by nearby pedestrians through the study of whether there is influence between the crowd and the crowd. The impact of unexpected events is too large or too small will lead to congestion, which is not conducive to the evacuation of the crowd. The simulation results are basically consistent with the two-way pedestrian evacuation in the real world.
【作者单位】：华南农业大学数学与信息学院;
【基金】：2015年国家级大学生创新训练计划项目(201510564279)~~
【分类号】：TP391.9;U491.226
【正文快照】： 0引言真实情形下,行人的行走行为具有较强的复杂性、随机性以及相互作用关系。普通的物理模型很难精确地描述行人的行走行为和特征。随着数学模型不断改进,利用元胞自动机对行人复杂行为的仿真和行人行走数据采集取得了很好的进展。行人流的研究可以从宏观和微观两个角度进行

【相似文献】