地铁站是人员密集场所，空间结构复杂并且客流量大，一旦发生突发事件地铁站的人员应急疏散将面临严峻挑战。然而突发事件往往还引发人员恐慌，恐慌情绪对行人正常疏散行为有极大影响，若地铁站内乘客的恐慌情绪未能及时受到引导，因不熟悉周围环境和交通工具、以及恐慌情绪等因素，容易引发大规模踩踏事故，进而造成严重后果。

本文基于元胞自动机模型（简称CA）对恐慌情绪下地铁站人员应急疏散仿真进行研究，首先对恐慌情绪感染和人员应急疏散的相关理论基础进行讨论，包括对应急疏散过程中人员的行为特征及心理特征进行总结阐述，分析影响人群疏散效率的重要因素，包括疏散者的运动速度、生理因素、心理动态及场景因素。

其次改进元胞自动机模型，通过对行人路径选择、心理及行为特性进行研究，增加SIS算法考虑恐慌情绪感染对地铁站行人疏散的影响，并采用A*算法完善在含有障碍物情境下的最优避障路径选择。设定元胞自动机模型运行规则，分别编写A*寻路程序、距离计算程序、地图模型绘制程序、地图显示程序和计算主程序，进而建立恐慌情绪元胞自动机疏散模型，基于MATLAB仿真软件，绘制30*30的元胞模型进行预试验来验证模型的有效性和可靠性。

在实例使用阶段经过实地调研，充分考虑西安市14号线某寨地铁站的周边环境、人流情况、设施种类和位置等因素后，确定模拟仿真参数通过CAD平面图建立基于MATLAB仿真软件所对应的环境地图，将现有实际情况应用于之前构建出的模型。对比得出地铁站人员有无恐慌情绪、情绪衰减率、行人密度、恐慌情绪阈值对疏散效率的影响。结果发现：有无恐慌情绪地铁站站台层和站厅层的疏散时间具有显著差异，行人有恐慌情绪时的疏散时间几乎是正常疏散下的2倍；情绪阈值升高时疏散时间会逐渐减少，当到达一定水平时，疏散时间与恐慌情绪阈值无关；情绪衰减速率小于0.06时，恐慌值会一直维持较高水平，情绪衰减速率越大，恐慌值降低得越快；行人密度也是成为影响疏散的关键，行人密度越大恐慌情绪在房间内反复传播进而不断加强，恐慌值增加导致疏散时间更长。最后通过地铁站模拟结果，经过全面分析从站台层、站厅层和其他方面（个人角度、管理者角度），对车站的管理和行人提出了相应的建议，以确保在应急情况下能够科学、高效、稳定、及时的对地铁站乘客进行疏散管理。

Subway station is a crowded place with complex spatial structure and large passenger flow. Once an emergency occurs, subway station personnel emergency evacuation will face severe challenges. However emergencies often also trigger a panic, the panic has great influence on normal pedestrian evacuation behavior, if the subway passengers panic failed to boot, because not familiar with the surrounding environment and traffic tools, and panic, easy to cause massive stampede, resulting in serious consequences.

Based on cellular automata model (CA) for panic under subway station personnel emergency evacuation simulation study, the first to panic infection and the related theoretical basis of personnel emergency evacuation are discussed, including the personnel in the process of emergency evacuation behavior characteristics and psychological characteristics were summarized in this paper, the analysis of the important factors that affect the crowd evacuation efficiency, Including evacuee's movement speed, physiological factors, psychological dynamics and scene factors.

Secondly, the cellular automata model was improved to study the pedestrian path choice, psychological and behavioral characteristics. SIS algorithm was added to consider the impact of panic emotion infection on subway station pedestrian evacuation, and A* algorithm was adopted to improve the optimal obstacle avoidance path selection in the context of obstacles. Set operation rules of cellular automata model respectively. Write A * pathfinding program, distance calculation procedures, map model map program, display program and calculation of the main program, and then establish the panic evacuation cellular automaton model, based on the MATLAB simulation software, draw 30 * 30 cellular automata model to carry on the preliminary experiments to verify the validity and reliability of the model.

Instance using phase after field investigation, in full consideration of xi 'an on line 14 a village subway station of the surrounding environment, the traffic situation, the type of facility and location factors, such as simulation parameters through CAD floor plan based on the MATLAB simulation software corresponding environment map, will be applied to the current actual situation before building the model. The effects of panic, emotion attenuation rate, pedestrian density and panic threshold on evacuation efficiency are compared. The results show that there is a significant difference in the evacuation time between the platform floor and the station floor with or without panic, and the evacuation time of pedestrians with panic is almost twice that of normal evacuation. When the emotional threshold increased, the evacuation time gradually decreased, and when it reached a certain level, the evacuation time had nothing to do with the panic threshold. When the emotion attenuation rate is less than 0.06, the panic value will always maintain a high level. The higher the emotion attenuation rate is, the faster the panic value decreases. Pedestrian density is also a key factor affecting evacuation. The higher the pedestrian density is, the panic will spread repeatedly in the room and strengthen continuously, and the increase of panic value will lead to the longer evacuation time. Finally, through the simulation results of subway station, through a comprehensive analysis from the platform layer, station hall layer and other aspects (personal perspective, manager perspective), the station management and pedestrians put forward the corresponding suggestions, to ensure that in emergency situations can be scientific, efficient, stable and timely evacuation management of subway station passengers.