主管：中华人民共和国应急管理部
主办：应急管理部天津消防研究所
ISSN 1009-0029  CN 12-1311/TU
消防科学与技术

消防科学与技术 ›› 2023, Vol. 42 ›› Issue (10): 1362-1369.

• • 上一篇    下一篇

基于改进蚁群算法的展览馆火灾疏散路径研究

黄莺, 郑雪琳, 赵鑫   

  1. （西安建筑科技大学 土木工程学院，陕西 西安 710000）
  • 出版日期:2023-10-15 发布日期:2023-10-15
  • 作者简介:黄 莺（1979- ），女，西安建筑科技大学副教授，博士，主要从事公共建筑消防管理、结构安全监测与检测、施工安全风险实施监控与评价、土木工程施工新技术、BIM云协同平台在施工管理中的应用等方面的研究工作，西安建筑科技大学1号信箱，710000。
  • 基金资助:
    基金项目：国家自然科学基金项目（52208204）

Research on fire evacuation path of exhibition hall based on improved ant colony algorithm

Huang Ying, Zheng Xuelin, Zhao Xin   

  1. (College of Civil Engineering, Xi'an University of Architecture and Technology, Shaanxi Xi'an 710000, China)
  • Online:2023-10-15 Published:2023-10-15

摘要： 为最大限度避免展览场馆发生火灾时造成的伤亡和损失，本文基于改进蚁群算法对展览类建筑的火灾疏散路径进行研究。首先，利用量化的火灾产物、角度和距离因素共同改进启发函数，建立符合人群疏散过程的信息素更新模型，对传统的蚁群算法进行改进。其次，建立人员动态疏散模型，以疏散时间和距离最短为目标函数，以人员密度、出口宽度等外界因素为约束条件，建立二维平面工作环境，进而上升到三维火灾疏散环境模型。以西安市某展馆为例进行实证研究，在MATLAB中建立30×25×3的三维栅格环境，对参数进行最优组合配置，通过改进后的蚁群算法对不同阶段的火灾场景进行仿真模拟，分析其收敛情况。结果表明：改进后的蚁群算法适用于三维空间的人员疏散工作，算法具有较好的收敛速度和全局搜索能力；将算法用于人员密集、安全隐患较多的大型展览场馆中，可快速找到最佳人员疏散路径，有助于提高展览场馆的人员疏散效率。

关键词: 改进蚁群算法, 三维疏散环境模型, 火灾动态模拟, 人员疏散效率

Abstract: In order to minimize the casualties and losses caused by fire in the exhibition hall. This paper puts forward research on fire evacuation path of exhibition hall based on improved ant colony algorithm. Firstly, quantified fire products, angle and distance factors are used to improve the heuristic function. The pheromone update model conforming to the process of crowd evacuation was established. Thereby, the traditional ant clony algorithm was improved. Secondly, the dynamic evacuation model of personnel was established, with the shortest evacuation time and distance as the target function; the two-dimensional evacuation environment and the grid map are built, and then rises to the 3D environment model. Finally, taking an exhibition hall in Xi'an as an example to conduct research, the 3D grid environment of 30×25×3 was established in MATLAB, the parameters were optimally combined and configured. The improved ant colony algorithm was used to simulate fire scenes at different stages and the convergence was analyzed. Results show that the improved ant colony algorithm is suitable for 3D evacuation models. The algorithm has excellent convergence speed and search capabilities. Using this improved algorithm in large exhibition hall, the best evacuation route can be quickly found. It helps to improve the efficiency of personnel evacuation in exhibition hall.

Key words: improved ant colony algorithm, 3D evacuation environment model, dynamic simulation of the fire, personnel evacuation efficiency