近年来随着国家城镇化建设的推进，居民出行距离及出行频率逐年增加，地面交通压力持续增大，地铁已成为是城市中重要的交通方式。地铁由于其特殊的运行环境，不占用任何地面交通资源，能有效规避地面出行所带来的交通拥堵。但由于换乘枢纽站内结构复杂、疏散路径曲折多变，大多过往乘车人员对车站环境不熟悉，一旦发生火灾等突发情况容易造成群体恐慌，在应急疏散时形成拥堵引发事故，进而造成人员的伤亡和财产损失。因此在地铁换乘枢纽站发生突发事件时如何更加迅速高效的实现人员的安全疏散，在地铁安全的研究中是十分迫切和必要的。

本文首先以查阅文献和现场问卷调研的方式收集整理了大量的乘车人员特征、行为特征、应激反应等基础数据，并进一步对地铁站在应急事件下的安全疏散隐患、人员行为表现，群体疏散特性等做出了详细分析，通过微观个体行为特性和宏观群体运动轨迹两方面对地铁应急疏散过程的影响因素进行了深入讨论。以西安地铁换乘枢纽站为背景，分析不同类型换乘枢纽站的空间结构下人员的换乘路线、日常管理和应急管理现状中存在的问题，辨识不同情况下客流分布特征对西安地铁换乘枢纽应急疏散的脆弱性因素，提出当前换乘枢纽站应急疏散管理所存在的问题。以这些特征和应急疏散脆弱性因素为基础，通过调整人员所处位置受全局最佳路径和最优出口的影响、恐慌情绪下对个体反应时间和期望速度的影响，建立基于改进PSO（粒子群优化算法—Particle Swarm Optimization）算法的西安地铁换乘枢纽应急疏散模型。以模型为基础，运用Anylogic软件进行仿真，分析地铁换乘枢纽站在常规运营情况下楼梯组、闸机口以及关键节点处等位置的疏散压力，提出影响应急疏散的瓶颈问题并加以分析研究。通过对完整疏散过程的研究发现，疏散过程很容易受到拥挤现象的影响，人员往往会向最宽的出口或最近的出口方向逃跑，如果行人有序的完成了疏散，此时恐慌心理也会降低，拥挤现象减少。连接站台层和站厅层的楼梯组，在自动扶梯总量减少并楼梯宽度增加1.0 m的情况下，总疏散长度不变时应急疏散时间减少54.7 s；站台屏蔽门调整至1.2 m时，应急疏散用时最短；受最佳出口的影响，不同站台层的楼梯、闸机等宽度的限制也会对疏散时间造成影响。本文研究成果为西安地铁应急管理部门提供了一些针对性优化建议，可供地铁换乘枢纽站的应急疏散设计以及日常安全管理提供参考.

In recent years, with the promotion of national urbanization construction, the travel distance and travel frequency of residents are increasing year by year, and the ground traffic pressure continues to increase. The subway has become an important mode of transportation in cities.Due to its special operation environment, the subway does not occupy any ground traffic resources, and can effectively avoid the traffic congestion caused by ground travel.However, due to the complex structure of the transfer hub station and the tortuous and changeable evacuation path, most of the passing passengers are not familiar with the station environment. Once fire and other emergencies occur, it is easy to cause group panic, causing congestion and accidents during emergency evacuation, and thus causing casualties and property losses.Therefore, it is very urgent and necessary to realize the safe evacuation of personnel more quickly and efficiently in the subway transfer hub station in the study of subway safety. This article first to consult literature and field questionnaire survey collected a large number of passenger characteristics, behavior characteristics, stress response and other basic data, and further to the subway station in emergency safety evacuation, personnel behavior performance, group evacuation characteristics made a detailed analysis, through the micro individual behavior characteristics and macro group trajectory factors discussed the subway emergency evacuation process.Xi 'an subway transfer hub station as the background, the analysis of different types of transfer hub station under the spatial structure of personnel transfer route, daily management and emergency management problems in the present situation, identify the passenger flow distribution characteristics of xi' an subway transfer hub emergency evacuation vulnerability factors, put forward the problems existing in the emergency evacuation management current transfer hub station.Based on these characteristics and emergency evacuation vulnerability factors, by adjusting the influence of the global optimal path and optimal exit, panic on individual response time and expected speed, establish xi'an metro transfer hub emergency evacuation model based on improved PSO algorithm.Based on the model, Anylogic software is used to analyze the evacuation pressure of the stair group, the gate port and the key nodes under the routine operation, put forward the bottleneck problems affecting the emergency evacuation and study them.Through the study of the complete evacuation process, it is found that the evacuation process is easily affected by the crowding phenomenon. People will often escape to the widest exit or the nearest exit direction. If the pedestrians complete the evacuation orderly, the panic will be reduced and the crowding phenomenon will be reduced.For the stair group connecting the platform floor and the station floor, the emergency evacuation time decreases by 54.7s when the total evacuation length increases by 1.0 m. When the platform shield door is adjusted to 1.2 m, the width of stairs and gates of different platform floors will also affect the evacuation time.The research results of this paper provide some targeted optimization suggestions for xi'an Metro emergency management department, which can provide reference for the emergency evacuation design and daily safety management of the subway transfer hub station.

[1]王晟华．地铁车站火灾疏散瓶颈问题及疏散策略探讨[J]．现代职业安全，2020（8）：68-71．

[2]房萍萍．西安大雁塔地铁站集散能力建模仿真分析[D]．西安：西安科技大学，2020．

[3]雷嘉烨．西安地铁换乘枢纽大客流应急疏散研究[D]．西安：西安理工大学，2020．

[4]Pandit S K. Numerical Simulations of Natural Convection and Entropy Generation in a Square Cavity with an Adiabatic Body[M]// Applied Mathematics. Springer India, 2015.

[5]Abhishek Saha, Swetaprovo Chaudhuri, Chung K. Law. Flame surface statistics of constant-pressure turbulent expanding premixed flames[J]. Physics of Fluids, 2014, 26(4): 61-66.

[6]GB 50157-2003，地铁设计规范[S]．

[7]刘晨，王立晓，孙小慧．影响地铁应急疏散行为的乘客空间认知分析[J]．交通科技与经济，2021，23（5）：7-13．

[8]Joris Wagenaar, Leo Kroon, loannis Fragkos. Rolling stock rescheduling in passenger railwaytransportation using dead-heading trips and adjusted passenger demand[J]. Transportation Research Pat B, 2017, 101.

[9]Saidi-Mehrabad M, Dehnavi-Arani S, Evazabadian F, et al. AnAntColony Algorithm (ACA) for solving the new integrated model of job shop scheduling and conflict-free routing of AGVs[J]. Computers & Industrial Engineering, 2015, 86(C): 2-13.

[10]Kefayat M, Ara A, Niaki S A. A hybrid of ant colony optimization and artificial bee colony algorithm for probabilistic optimal placement and sizing of distributed energy resources[J]. Energy Conversion & Management, 2015, 92(3): 149-161.

[11]Zhang X, Chen w, Wang B, et al. Intelligent fault diagnosis of rotating machinery using support vector machine with ant colony algorithm for synchronous feature selection and parameter optimization[M]. Elsevier Science Publishers B. V. 2015, (7): 1221-1228

[12]Salama K, Freitas A. Ant colony algorithms for constructing Bayesian multi-net classifiers[J]. Intelligent Data Analysis, 2015, 19(2): 233-257.

[13]Chen Zhao. Subway evacuation simulation based on Pathfinder[J]. International Core Journal of Engineering, 2021, 7(12).

[14]Zhen-yu Li, Minbo Tang, Dong Liang, Zhe Zhao. Numerical Simulation of Evacuation in a Subway Station[J]. Procedia Engineering, 2016, 135.

[15]Jing-hong Wang, Wen-yu Yan, You-ran Zhi, Jun-cheng Jiang. Investigation of the Panic Psychology and Behaviors of Evacuation Crowds in Subway Emergencies[J]. Procedia Engineering, 2016, 135.

[16] Evacusled Inc. Patent Issued for Subway Evacuation Mattress and Method of Manufacture[J]. Bioterrorism Week, 2014.

[17]Evacusled Inc. Patent Issued for Subway Evacuation Mattress and Method of Manufacture[J]. Bioterrorism Week, 2014.

[18]Toris Wagenaar, Leo Kroon, Ioannis Fragkos. Rolling stock rescheduling in passenger railwaytran sportation using dead-heading trips and adjusted passenger demand[J]. Transportation Research Part B, 2017, 101.

[19]Cheng-Jie Jin, Rui Jiang, Jun-Lin Yin, Li-Yun Dong, Da-wei Li. Simulating bi-directional pedestrian flow in a cellular automaton model considering the body-turning behavior[J]. Physica A: Statistical Mechanics and its Applications, 2017, 482.

[20]谢知航．列车越站对侧岛混式地铁站台火灾烟气蔓延的影响[D]．西安：西安科技大学，2020．

[21]陈子健．某地铁侧式车站火灾烟气运移规律及人员疏散模拟研究[D]．安徽：安徽建筑大学，2020．

[22]王起全，李登尧，杨鑫刚．地铁事故应急疏散模拟及优化研究[J]．中国安全生产科学技术，2019，15（11）：170-178．

[23]张强，王峰，魏利伟．地铁车站换乘分析——以西安地铁小寨站为例[J]．住宅与房地产，2019（33）：248．

[24]薄杰，徐福顺，袁昕，等．元胞自动机与pathfinder疏散模拟对比研究[C]．公共安全科学技术学会（China Association for Public Safety）、全国高校安全科学与工程学术年会委员会：中国科学技术大学火灾科学国家重点实验室，2019：167．

[25]徐丹．地铁环境内人员应急疏散行为研究[D]．首都经济贸易大学，2019．

[26]鲁玲艳．公共交通枢纽火灾条件下人员疏散仿真研究[D]．北京：首都经济贸易大学，2019．

[27]许梦，任委峰，袁静茹．基于乘客需求的轨道交通换乘车站导向系统优化研究—以西安地铁北大街站为例[J]．科技资讯，2019，17（11）：178-180．

[28]朱庆松．地铁区间隧道火灾烟气流动特性及人员疏散研究[D]．安徽：安徽工业大学，2019．

[29]孙亚杰．地铁站闸机设置方式对人员紧急疏散的影响研究[J]．科学技术创新，2019（10）：126-127．

[30]鲁丽莉．地铁车站火灾应急疏散建模与仿真研究[D]．兰州：兰州交通大学，2019．

[31]张建平，庞瑾．西安地铁3号线客流特征及运营方案优化研究[J]．城市轨道交通研究，2018，21（7）：30-33．

[32]Chen Zhao. Subway evacuation simulation based on Pathfinder[J]. International Core Journal of Engineering, 2021, 7(12).

[33]Zhen-yu Li, Minbo Tang, Dong Liang, Zhe Zhao. Numerical Simulation of Evacuation in a Subway Station[J]. Procedia Engineering, 2016, 135.

[34]Jing-hong Wang, Wen-yu Yan, You-ran Zhi, Jun-cheng Jiang. Investigation of the Panic Psychology and Behaviors of Evacuation Crowds in Subway Emergencies[J]. Procedia Engineering, 2016, 135.

[35] Evacusled Inc. Patent Issued for Subway Evacuation Mattress and Method of Manufacture[J]. Bioterrorism Week, 2014.

[36]Evacusled Inc. Patent Issued for Subway Evacuation Mattress and Method of Manufacture[J]. Bioterrorism Week, 2014.

[37]Sibel Gokce, Ozhan Kayacan. Study on bi-directional pedestrian movement using ant algorithms[J]. Chinese Physics B, 2016, 25(1): 529-538.

[38]Claudio Feliciani, Katsuhiro Nishinari. An improved Cellular Automata model to simulate the behavior of high density crowd and validation by experimental data[J]. Physica A: Statistical Mechanics and its Applications, 2016, 451.

[39]Svistunova A S. Using the AnyLogic software product in modeling the passenger traffic of a railway station[J]. Computing, Telecommunication and Control, 2021, 13(67).

[40]Oh Chang Hyup, Kim Min Hee, Kim Byung In, Ko Young Myoung. An Efficient Building Evacuation Algorithm in Congested Networks[J]. IEEE Access, 2019, 7.

[41]Antonova V. M. Grechishkina N. A. Kuznetsov N. A. Sukhorukova N. A. Traffic simulation of queueing systems in the ANYLOGIC application with an example of passenger flow of the subway station[J]. Journal of Radio Electronics, 2018(3).

[42]Wu Ni, Wenbo Li, Hailei Wang, Chengxin Xiong, Dan Guo. Capacity-Constrained Contraflow Adaption for Lane Reconfiguration in Evacuation Planning[J]. Journal of Advanced Transportation, 2018.

[43]Marina Gravit, Ivan Dmitriev, Kirill Kuzenkov. Phased evacuation algorithm for high-rise buildings[J]. MATEC Web of Conferences, 2018, 245.

[44]孙春伟．地铁车站火灾情景演化研究[D]．北京：北京交通大学，2018．

[45]刘隽雅．基于换乘优化设计的西安地铁小寨站内部空间流线组织研究[D]．西安：长安大学，2018．

[46]Minhyuck Lee, Hyunwoo Nam, Chulmin Jun. Multiple exits evacuation algorithm for real-time evacuation guidance[J]. Spatial Information Research，2017，25(2).

[47]孙锐．突发事件下建筑房间安全出口疏散仿真研究[D]．黑龙江：东北石油大学，2018．

[48]王付宇，王骏．突发事件情景下地铁站人员应急疏散问题综述[J]．计算机应用研究，2018，35（10）：2888-2893．

[49]王蕊．西安地铁北大街站紧急疏散下的客流仿真与优化措施[D]．西安：长安大学，2018．

[50]刘佼，郑建振，王雨晨．西安地铁二号线客流预测后评估[J]．黑龙江：黑龙江交通科技，2017，40（5）：163-164．

[51]吴鹏飞．地铁站安全疏散优化设计研究[D]．西安：西安建筑科技大学，2017．

[52]刘梦洁．基于FDS和Pathfinder的地铁车站火灾疏散研究[D]．湖北：华中科技大学，2016．

[53]杜长宝，朱国庆，李俊毅．疏散模拟软件STEPS与Pathfinder对比研究[J]．消防科学与技术，2015，34（4）：456-460．

[54]成琳娜．基于Pathfinder的地铁站火灾应急疏散仿真研究[D]．兰州：兰州交通大学，2014．

[55]周继彪．综合交通换乘枢纽行人交通特性及安全疏散研究[D]．西安：长安大学，2014．

[56]丁波，彭瑾蓉，周传钰，等．基于AnyLogic的轨道交通车站拥堵疏解方法分析及方案设计[J]．交通企业管理，2021，36（6）：67-69．

[57]唐健，胡玉玲，王亚慧．基于Anylogic的大空间建筑引导员疏散引导研究[J]．消防科学与技术，2021，40（11）：1594-1598．

[58]郭桃明，曹宇齐．地铁消防疏散问题思考与研究[J]．现代城市轨道交通，2021（9）：58-62．

[59]李博伦．地铁车站建筑防火及安全疏散设计[J]．居舍，2021（26）：83-84+102．

[60]卢颖，吕希凡，仇乐，等．地铁乘客社会脆弱性影响因素调查与统计研究[J/OL]．安全与环境学报：1-8[2022-03-14]．

[61]王琳，张国奥，徐杭．地铁车站疏散设施布局对乘客疏散时间的影响[J]．城市轨道交通研究，2021，24（9）：51-54．

[62]陈泓旭，李焕．基于Anylogic的轨道交通换乘站高峰客流分析[J]．中国航务周刊，2021（35）：48-49．

[63]谭家磊，刘海力，陈云飞，等．基于Anylogic的某高铁客运站人群疏散设施优化研究[J]．安全，2021，42（8）：13-19．

[64]张润畦．地铁站台火灾应急疏散研究[J]．华北科技学院学报，2021，18（4）：67-73．DOI：10．19956/j．cnki．ncist．2021．4．11．

[65]宋守信，陈明利，陈川南．大客流下的地铁应急疏散管理[J]．劳动保护，2021（7）：13-15．

[66]周方，张丽，柏萍，等．基于BIM的地铁车站火灾安全疏散研究[J]．中国高新科技，2021（12）：26-27．

[67]周美琦，杨晓霞，张纪会，等．基于改进元胞自动机模型的地铁车厢乘客疏散模拟[J]．复杂系统与复杂性科学，2021，18（3）：35-44．

[68]吴佳梦，朱顺兵，郝雪彤，等．地铁换乘站高峰期火灾及人员疏散模拟研究[J]．消防科学与技术，2021，40（6）：875-879．

[69]付红飞． 地铁换乘站应急疏散方案研究[D]．徐州：中国矿业大学，2021．

[70]郝水清． 基于改进社会力模型的地铁车站火灾应急疏散仿真研究[D]．北京：北京交通大学，2021．

[71]郝闵熙． 基于行为决策模型的地铁应急疏散仿真研究[D]．乌鲁木齐：新疆大学，2021．

[72]丁波，彭瑾蓉，周传钰，等．基于AnyLogic的轨道交通车站拥堵疏解方法分析及方案设计[J]．交通企业管理，2021，36（6）：67-69．

[73]唐健，胡玉玲，王亚慧．基于Anylogic的大空间建筑引导员疏散引导研究[J]．消防科学与技术，2021，40（11）：1594-1598．

[74]陈泓，李焕．基于Anylogic的轨道交通换乘站高峰客流分析[J]．中国航务周刊，2021（35）：48-49.