~地铁车站发生火灾易造成严重的直接经济损失和人员伤亡，如今越来越多的新建地铁车站，因受到文物保护、建筑限制、交通保障及现有路线换乘关系等诸多因素限制，需要因地制宜，以特殊的车站形式来满足结构和功能的需求，这就使人员安全疏散工作变得更加复杂，因此研究火灾烟气运动情况及人员最短疏散路径尤为重要。鉴于此，本文以西安神州二路地铁车站为依托，采用数值模拟和蚁群算法相结合的研究方法，对交叠型车站火灾烟气运动规律及人员最短疏散路径进行研究，为类似车站开展消防救援提供理论支撑。主要工作及研究结论如下：
（1）通过与BIM技术结合，克服火灾模拟软件存在建模过程繁琐，模型精度差的问题，实现三维建筑模型与火灾模型的联动，提高数值模拟结果的准确性。
（2）利用Pyrosim软件，在站台层设置火源功率为3.0MW的起火源，分析三种常见火灾场景（站台两侧及中央）。通过对三种场景下站内火灾产物进行分析，得出结论：相较于站台最西侧和站台中部发生火灾，火源位于站台最东侧时，火灾产物对人员疏散影响最为严重，更不利于站内人员进行疏散；当站台最西侧发生火灾时，火灾产物在起火后得到最有效控制。
（3）以被困人员紧急疏散路径研究和智能算法在疏散模型中的应用研究为重点，利用Matlab建立由障碍物顶点构成的栅格地图，定义疏散空间内节点及疏散通道的静态属性和动态属性，调整信息素分布情况。提出引入人员活动性指数、路径通行难易系数、计算疏散路径的当量长度来改进算法中的启发函数。改进后有效克服算法早熟和收敛速度慢问题。
（4）确定火场环境对人员的影响因素，量化不同路况、火场温度、CO气体浓度、能见度对人员疏散速度的影响，结合多次仿真实验，配置最优的参数组合：α为1；β为5；ρ为0.6；Q为10；m为1.2。
（5）针对三个火灾危险点进行疏散路线的仿真模拟，结果表明：最短疏散路径随火情的发展而产生变化，随着时间的进行，火源功率逐渐提升，烟气（等燃烧产物）对人员疏散速度的影响越来越大，从而需要根据烟气流动规律对疏散路径进行动态调整，最终输出最短疏散路径。
本文通过引入火灾动态参数对蚁群算法进行改进，能够根据火场内环境状况适时调整疏散路线，减少逃生所用时间，提高人员疏散效率。

~More and more new subway stations today are restricted by many factors such as heritage protection, building restrictions, traffic security and existing route interchange relationships, etc., and need to meet the structural and functional requirements with special station forms, which makes the safe evacuation of people more complicated, so it is very important to study the fire smoke movement. Therefore, it is very important to study the fire smoke movement and the shortest evacuation path. In view of this paper adopts a combination of numerical simulation and ant colony algorithm to study the fire smoke motion law and the shortest evacuation path of personnel in the overlapping stations, and provides theoretical support for fire rescue in similar stations. The main work and research findings are as following.
(1) by combining with BIM technology, we overcome the problems of cumbersome modeling process and poor model accuracy of fire simulation software, and realize the linkage between the 3D building model and fire simulation model. The Deviation of simulation results caused by inaccurate models is effectively avoided.
(2) Use Pyrosim software to set a fire source with a power of 3.0MW on the platform layer, and analyze three common fire scenarios (on both sides of the platform and in the center). Through the analysis of the fire products in the station under the three scenarios, it is concluded that compared with the fire in the westernmost and central part of the platform, when the fire source is located in the easternmost part of the platform, the fire products have the most serious impact on the evacuation of the people in the station, and are not conducive to the evacuation of the people in the station. When a fire breaks out on the westernmost side of the platform, fire products are most effectively contained after the fire.
(3) Focusing on the study of the emergency evacuation path of trapped people and the application of intelligent algorithm in the evacuation model, Matlab is used to establish a grid map composed of obstacle vertices, define the static and dynamic attributes of nodes and evacuation channels in the evacuation space, and adjust the pheromone distribution. The heuristic function of the algorithm is enhanced by introducing personnel activity index, path difficulty coefficient and calculating equivalent length of evacuation path. The improved algorithm can effectively solve the problems of prematurity and slow convergence.
(4) Determine the influencing factors of fire environment on personnel, quantify the influence of different road conditions, fire temperature, CO gas concentration and visibility on personnel evacuation speed, and configure the optimal combination of parameters by combining multiple simulation experiments: α is 1; β is 5; ρ is 0.6; Q is 10; m is 1.2.
(5) Simulation of evacuation routes for the above three scenarios, the results show that the shortest evacuation path produces dynamic changes with the development of the fire, where the reason is that the fire develops over time, the power of the fire source gradually increases, and its combustion products have an increasing impact on the evacuation speed of the personnel, thus requiring constant dynamic adjustment of the evacuation path according to the law of smoke flow, and finally outputting the shortest evacuation path.
This paper improves the ant colony algorithm by introducing dynamic parameters of fire, which can adjust the evacuation route in time according to the environmental conditions in the fire, reduce the time employed to escape, and improve the efficiency of personnel evacuation.

[1]朱奥妮.2000—2019年国内外地铁火灾事故统计分析[J].城市轨道交通研究，2020,23(08):148-150.

[2]Art Bendelius. Memorial Tunnel Fire Ventilation Test Programme [A].One Day Seminar of Smoke and Critical Velocity in Tunnels[C]. London, 1996.

[3]Holborn P G, Nolan P F, Golt J. An analysis of fire sizes, fire growth rates and times between events using data from fire investigations [J]. Fire Safety Journal, 2004, 39(6):481-524.

[4]Ingason H, Lönnermark A. Heat release rates from heavy goods vehicle trailer fires in tunnels [J]. Fire Safety Journal, 2005,40 (7): 646-668

[5]Lonnermark A, Ingason H. Gas temperatures in heavy goods vehicle fires in tunnels [J]. Fire Safety Journal, 2005, 40(6): 506-527.

[6]Lonnermark A, Ingason H. Fire Spread and Flame Length in Large-Scale Tunnel Fires[J]. Fire Technology, 2006, 42(4): 283-302.

[7]Qiu Peiyun,Long Zeng,Liu Chang,Yang Yuxuan,Tian Xiangliang,Zhong Maohua. Full-scale experiment on smoke propagation in metro station tunnel fire under opening the platform screen doors[J]. Tunnelling and Underground Space Technology,2021,107.

[8]龙增，刘畅，杨宇轩等.含阶梯式站厅地铁岛式车站火灾全尺寸实验研究[J].清华大学学报（自然科学版）,2020,60(09):787-794.

[9]田向亮，钟茂华，陈俊沣等.地铁十字换乘车站全尺寸实验研究：Ⅰ.站厅火灾[J].中国安全生产科学技术，2019,15(03):11-18.

[10]钟茂华，刘畅，史聪灵.地铁火灾全尺寸实验研究进展综述[J].中国安全科学学报，2019,29(10):51-63.

[11]田向亮，钟茂华，刘畅等.细水雾抑制隧道甲醇火灾的全尺寸实验研究[J].火灾科学，2019,28(01):35-41.

[12]102Rie D H, Hwang M W, Kim S J, et al. A study of optimal vent mode for the smoke control of subway station fire [J]. Tunnelling & Underground Space Technology Incorporating Trenchless Technology Research, 2019, 21(3):300-301.

[13]Kim J Y, Kim K Y. Experimental and numerical analyses of train-induced unsteady tunnel flow in subway [J]. Tunnelling & Underground Space Technology Incorporating Trenchless Technology Research, 2018, 22(2):166-172.

[14]Matsushita T, Klote J H. Smoke movement in a corridor-hybrid model, simple model, and comparison with experiments [J]. International Review of the Red Cross, 1992, 86(856):815-835.

[15]Li Y Z, Lei B, Ingason H. The maximum temperature of buoyancy-driven smoke flow beneath the ceiling in tunnel fires [J]. Fire Safety Journal, 2011, 46(4):204-210.

[16]刘畅，钟茂华，田向亮等.地铁多线换乘车站火灾模型实验研究—（1）模型装置设计[J].中国安全生产科学技术，2020,16(02):11-17.

[17]陈俊沣，钟茂华，程辉航等.地铁多线换乘车站火灾模型实验研究—（2）十字换乘车站火灾[J].中国安全生产科学技术，2020,16(05):5-11.

[18]龙增，钟茂华，杨宇轩等.地铁多线换乘车站火灾模型实验研究－（3）平行换乘车站火灾[J].中国安全生产科学技术，2020,16(10):12-19.

[19]钟委，霍然，彭伟等.站厅两侧楼梯对火灾烟气流动影响的实验研究[J].中国科学技术大学学报，2008,38(12):1455-1459.

[20]赵红莉，徐志胜，李洪等.坡度对隧道火灾烟气温度分布的影响[J].中南大学学报（自然科学版）,2013,44(10):4257-4263.

[21]Chow W K,Lam K C,Fong N K. Numerical simulations for a typical train fire in china [J]. Modeling and Simulation in engineering,2018,1:50-60

[22]Park W H,Kim,D H Chang H C.Numerical predictions of smoke movement in a subway station under ventilation [J] Tunneling&Underground Space Technology Incorporating Trenchless Technology Research,2019,21(3):304-304

[23]Li Y F,Zhang Y X,Lin X X Study on Ventilation and Smoke Control System of Subway Transfer Stations[C].2017 7th International Conference on Intelligent Computation Technology and Automation(ICICTA).IEEE,2014

[24]Li Y F,Sun X,Feng X.Study on Evacuation in Subway Transfer Station Fire by STEPS[J].procedia Engineering,2018,45(2):735-740

[25]Li Y F,Lin X X,Xiao F.Life Safety Evacuation for Cross Interchange Subway Station fire [J].Procedia Engineering,2012,45(2);741-747

[26]Tsuksharn M. Koshiba Y , Ohtanit H. Effectiveness of downward evacuation in a large.scale

[27]subway fire using Fire Dynamics Simulator[J]. Tuelling and Underground Spuce Technology. 2011.26(4):573.581.

[28]Abur Zaid S A. Bendelius A G, Santoianni D A.et al, Using Computational Fluid Dynamics to design an emergency ventilation system for a transit subway station[C] .International Mechawiral Eninering Congress and Er position. US, American Society of Mechanical Engineers. 1995;75.87.

[29]Elias s R, Raw MJ .Bostwick P, et al. Numerical simulation of subway station fires and ventilation[M].Proveedings of the 1996 ASME Flaids Engineering Division Summer Meeting. New York, NY : The Anerican Society of Mechanical Engineers. 1996.238(3).557.562.

[30]Simcox s, Wilkes N. S,Jones I. Computer simulation of the flows of hot gases from the fire at King's Cross underground station [D], Fire Safety Journal. 1992(18).49.73.

[31]保彦晴.某地铁站列车火灾烟气运动规律研究[J].消防科学与技术，2014,33(06):619-621.

[32]赵哲，林基深，曾高彬等.相连商业街火灾对地铁车站影响的数值模拟分析[J].科技通报，2017,33(05):228-231.

[33]钟茂华，张磊，肖衍等.地铁“T”形换乘车站通道火灾通风模式数值模拟研究[J].中国安全生产科学技术，2017,13(09):13-19.

[34]陈和燕，池秀文，嵇舒翰等.地铁站聚氨酯与聚氯乙烯火灾特性及人员安全疏散研究[J].城市轨道交通研究，2020,23(12):103-107.

[35]陈绍宽，狄月，史荣丹等.地铁车站站台火灾影响分析与人员疏散研究[J].交通运输系统工程与信息，2017,17(01):241-248.

[36]田鑫，苏燕辰，李冬等.地铁车站火灾疏散仿真分析[J].科学技术与工程，2017,17(16):333-337.

[37]喻敏，兰志光.基于BIM与Pathfinder的地铁车站客流疏散相关研究[J].隧道建设（中英文）,2020,40(S1):179-186.

[38]邓朗妮，雷丽贞，廖羚等.基于BIM的地铁施工阶段火灾监测预警可视化研究[J].消防科学与技术，2020,39(08):1097-1100.

[39]洪洋. 某双层岛式地铁车站火灾烟气蔓延规律研究[D].安徽建筑大学，2018.

[40]张茜，曹建华，王文倩.典型地铁火灾场景中烟气蔓延与控制研究[J].中国安全生产科学技术，2011,7(07):20-25.

[41]王海涛，李彬，曹荟娟等.室外空气对地铁隧道烟气返流长度的影响[J].消防科学与技术，2020,39(06):768-771.

[42]毛军，郗艳红，胡家鹏等.分离式深埋地铁车站的火灾烟气运动及防控[J].中国安全科学学报，2019,29(06):76-82.

[43]Pauls J. Personal perspective on research, consulting and codes/standards development in fire related human behavior, 1969. 1999, with an emphasis on space and time factors[J]. Fire and Materials, 1999 ,23(6) :265.272.

[44]Roh. J. S, Ryou. H. S, Park. W. H.et al. CFD Simulation and Assessment of Life Safety in a SubwayTrain Fire [J]. Tunneling and Underground Space Technology, 2018, 24(4): 447-453.

[45]Karl Fridolf & Hakan Frantzich. Evacuation in Underground Rail Transportation Systems: ASummary of the Findings of the Metro Project [C]. Sixth International Symposium on TunnelSafety and Security, Marseille, France, March 12-14, 2018.

[46]Gulanur A, Tian C, Xie Q, Shawkat HM, Liu D. Simulation of Emergency Evacuation in a Metro-rail Transit Station. In IOP Conference Series: Earth and Environmental Science[C], 2019, 304(3):p. 032016). IOP Publishing.

[47]马浩博，季建华，何冰.大规模人流多出口应急疏散预案的优化与研究[J].系统管理学报，2011,20(02):238-243.

[48]刘栋栋，张小丽，李磊等.北京站、北京西站行人特征的调查与分析[J].建筑科学，2010,26(07):26-30.

[49]于恒. 基于火灾动力学与人群疏散模拟的地铁车站火灾安全疏散问题研究[D].华南理工大学，2020.

[50]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.

[51]唐飞，何清，朱孔金，徐童，石琴，李恒.火灾情况下某城市地铁换乘站内大规模人群疏散特征研究[J].安全与环境学报，2018,18(04):1419-1426.

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

[53]张立茂，吴贤国，李博文等.基于火灾模拟器和Pathfinder的地铁车站人员疏散[J].科学技术与工程，2018,18(04):203-209.

[54]罗振敏，郝强强，程方明，王涛.地铁隧道火灾模拟及人员疏散研究[J].消防科学与技术，2019,38(03):363-367.

[55]王海芹. 地铁突发火灾时人员安全疏散影响因素研究[D].西安建筑科技大学，2010

[56]史玉晓. 地铁站火灾烟气扩散及控制研究[D].西安建筑科技大学，2017

[57]James A. Milke. Evaluating the early development of smoke hazard from fires in large spaces[J]. ASHRAE Transactions,2000,106(1).