考虑火灾产物影响下的人员疏散路径规划研究  

Research on Personnel Evacuation Path Planning under the Influence of Fire Products

作　　者：刘纪坤 袁雪颖 王佳茹 王翠霞 LIU Jikun;YUAN Xueying;WANG Jiaru;WANG Cuixia(College of Safety Science and Engineering,Xi'an University of Science and Technology,Xi'an 710054,China;Shaanxi Road and Bridge Group Co.,Ltd.,Xi'an 710075,China)

机构地区：[1]西安科技大学安全科学与工程学院,西安710054 [2]陕西路桥集团有限公司,西安710075

出　　处：《计算机测量与控制》2025年第4期155-161,共7页Computer Measurement &Control

基　　金：国家自然科学基金(52204235)。

摘　　要：为解决现代建筑火灾应急疏散过程中人员难以快速、安全疏散的问题,提高人员面对火灾的应急能力,提出一种考虑火灾产物影响条件下的改进蚁群算法,综合考虑可见度、CO浓度和烟气温度3种火灾产物对人员疏散的影响,对蚁群算法的启发信息、信息素更新方法进行改进;通过Dijkstra算法规划初始疏散路径,利用改进蚁群算法对初始疏散路径进行优化;以西安某地铁站台实际地形为例,通过PyroSim软件进行火灾模拟,获取地铁站台火灾产物的实时数据,采用Matlab软件分别进行改进蚁群算法和传统蚁群算法对比仿真实验、不同火灾时期下的改进蚁群算法仿真实验,结果表明:改进后的蚁群算法提高了疏散效率和收敛速度,缩短了疏散时间和路径长度,可实现火灾疏散路线的动态规划。In order to solve the problem of the difficulty in quickly and safely evacuating personnel during the emergency evacuation of modern building fires,and to improve personnel’s emergency response capabilities in the face of fires,an improved ant colony algorithm considering the effects of fire products is proposed.Comprehensively consider the effects of three fire products,namely visibility,CO concentration and smoke temperature,on personnel evacuation,and improve the heuristic information and pheromone updating methods of the ant colony algorithm.The initial evacuation path is planned by Dijkstra's algorithm,and it is optimized using the improved ant colony algorithm.By taking a subway station in Xi'an as an example,the fire simulation is carried out by PyroSim software,achieving the real-time data of the fire products of the metro platform.Matlab software is used to simulate comparative experiments between the improved ant colony algorithm and ant colony algorithm,and simulate the improved ant colony algorithm under different fire periods,respectively.The research results show that the improved ant colony algorithm improves the evacuation efficiency and convergence speed,shortens the evacuation time and path length,and realizes the dynamic planning of fire evacuation routes.

关 键 词：地铁车站 疏散路径规划 蚁群算法 DIJKSTRA算法 火灾产物 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]