基于多智能体的高层建筑分阶段火灾疏散仿真及策略研究  

Simulation and Strategy for Phased Fire Evacuation of High-Rise Building Based on Multi-Agent

作　　者：黄梓浩 景奕钦 朱后壮 徐晶[1] 马国丰[1] 

机构地区：[1]同济大学经济与管理学院,上海

出　　处：《建模与仿真》2024年第2期1592-1603,共12页Modeling and Simulation

摘　　要：随着城市化进程的深入,高层建筑的数量不断增加。高层建筑在面对火灾时的高风险使得疏散过程也尤为重要。本文将高度作为分类标准,选取不同楼层作为基准层分别进行分阶段疏散,同时考虑了火灾对不同高度的影响不同而制定了新的疏散策略。在本研究中,我们通过改变疏散顺序、延迟比例和延迟时间,基于智能体模型建立了1种正常疏散场景及30种分阶段疏散场景31种不同疏散场景。案例研究结果表明,这种分阶段疏散策略对于率先疏散中、高楼层上部位人员离开着火楼层的疏散效率起到促进作用,且分阶段疏散能够在不同程度上改变人员的疏散速度和建筑内的密度。本研究为人们在面对高层建筑火灾时采用何种疏散方式提供了建议。With the deepening of urbanization, the number of high-rise buildings is increasing rapidly. The high risk of fire in high-rise buildings makes the evacuation process particularly important. We take height as the classification standard, select different floors as datum floors for phased evacuation and formulate a new evacuation strategy. In this study, 1 total evacuation scenario and 30 phased evacuation scenarios based on the agent model are established by changing evacuation sequence, delay ratio, and delay time. The results of the case study show that the phased evacuation strategy improves the evacuation efficiency of people in the middle and upper floors, and it can change the evacuation speed of people and the density within the building to varying degrees. This study provides suggestions on how to evacuate people in fire accident.

关 键 词：高层建筑 分阶段疏散 疏散策略 智能体模型 

分 类 号：TU9[建筑科学]