基于社会力模型的人群疏散仿真研究

发布时间：2018-07-10 18:37

本文选题：群体行为 + 人群疏散　；参考：《大连理工大学》2016年硕士论文

【摘要】：近年来,国内外大城市中的突发事件频繁发生,而如何在紧急情况下进行有效的人群疏散已经引起各国的广泛关注。而我国正处于社会转型期,城市化进程加快使得城市内人口和建筑物密度急剧增加,公共建筑物内人群聚集问题更加突出,发生突发事件后造成的财产、人员损失也更加严重。因此,研究公共场所内的群体行为现象和人群疏散特征以及在突发事件下如何组织有效合理的人群疏散从而减少人员伤亡和财产损失具有重要的现实意义。本文对现有的人群疏散研究进行综述后发现：现有的研究大多集中在研究行人流特性方面,利用现有的人群疏散模型进行改进和创新从而对人群疏散提供理论支持。但在现实情况下,公共场所内环境复杂、出口众多,如何针对具体情境提出对应的仿真模型和疏散策略显得十分必要。本研究基于Helbing提出的社会力模型,提出了一个同时考虑了公共场所内的行人特性以及人为干预这两种情况作用下的人群疏散模型。以一个具体的大型购物广场一层为例开展考虑不同因素作用下的仿真实验,提出对公共场所内人群疏散的有关建议,为减免具体场景下突发事件导致的人员伤亡做出一定贡献。本文的主要研究工作如下：(1)构建符合公共场所内群体特征的人群疏散模型。首先总结前人针对群体行为的研究成果,揭示出突发事件下群体的行为特征：如恐慌心理蔓延、行为簇等。然后将恐慌因子、小团体、引导行为加入到社会力模型中构建适用于公共场所的人群疏散模型。(2)将粒子群算法改进后用来刻画人群疏散过程中的团体逃生行为。对具体的公共场所进行场景描述和参数抽取后,针对疏散时出现的两种小团体行为,前文中用社会力模型的参数修改来实现联系紧密的紧密型小团体,而疏散时形成的人数多、内部不紧密的逃生型小团体则采用粒子群算法来模拟行人的疏散行为。(3)对突发事件下、小团体、商场的人群疏散进行仿真模拟。利用Anylogic仿真平台针对加入恐慌因子引导行为、粒子群等6个情景多次调整参数进行重复试验,根据实验结果数据分析各个因子或参数对疏散效率的影响,也验证了本文中的人群疏散模型的有效性。(4)针对大型购物广场内突发事件下的人群疏散提出相应对策。在对实验模拟结果数据进行分析讨论的基础上,针对紧急事件下如何引导和平时的应急预案设置提出相关建议。
[Abstract]:In recent years, sudden events occur frequently in big cities at home and abroad, and how to evacuate effective crowd in emergency has aroused wide attention of all countries. However, China is in a period of social transformation. The acceleration of the process of urbanization makes the population and building density increase sharply in the city, and the problem of crowd gathering in public buildings is more abrupt. It is of great practical significance to study the phenomenon of group behavior and crowd evacuation in public places and how to organize effective and reasonable evacuation in public places to reduce casualties and property losses in public places. After a summary of the research, it is found that most of the existing studies focus on the study of pedestrian flow characteristics, and use the existing crowd evacuation model to improve and innovate in order to provide theoretical support for the evacuation of the crowd. The evacuation strategy is very necessary. Based on the social force model proposed by Helbing, this paper puts forward a crowd evacuation model which considers the pedestrian characteristics and human intervention in public places at the same time, taking a specific shopping plaza as an example to take into account the simulation experiments under the action of different factors, which is taken as an example. This paper puts forward some suggestions on evacuation in public places to make some contributions to reduce the casualties caused by unexpected incidents in specific scenes. The main research work of this paper is as follows: (1) building a crowd evacuation model conforming to the community characteristics in public places. First, it summarizes the results of previous studies on group behavior and reveals the unexpected events. The behavior characteristics of the group: such as panic spread, behavior cluster and so on. Then, panic factor, small group, guide behavior are added to the social force model to construct the crowd evacuation model suitable for public places. (2) the particle swarm optimization is improved to describe the group escape behavior in the process of crowd evacuation. After the scene description and parameter extraction, for the two small group behaviors in evacuation, the previous article uses the parameters of the social force model to realize close close small groups, and the number of people formed in the evacuation is more, and the small group which is not close in the interior uses particle swarm optimization to simulate the evacuation behavior of pedestrians. (3) the burst of emergency. In the event, the crowd evacuation of the small group and the shopping mall is simulated. The Anylogic simulation platform is used to repeat the parameters of 6 scenarios, such as panics factor guidance behavior, particle swarm optimization, and the effect of each factor or parameter on the evacuation efficiency according to the experimental data, and the crowd evacuation model in this paper is verified. The validity of the model. (4) the corresponding countermeasures are put forward for the evacuation of the crowd in the large shopping plaza. On the basis of the analysis and discussion of the experimental data, some suggestions are put forward to guide the emergency plan setting under the emergency.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：D63

【相似文献】