建筑物火灾中的人群疏散研究

Research on Crowd Evacuation in Building Fires

建筑物火灾是我国频发的安全事故,所以应研究建筑物火灾人群安全疏散问题。由于在建筑物火灾中,人群疏散时出现拥堵,存在不安全因素,造成人员伤亡。针对在现有的研究中未考虑人员行为的影响,提出了智能体(Agent)的人群行为建模技术在建筑物火灾中的人群疏散仿真中的应用方法。仿真结果显示基于Agent的行为模型可以仿真出人员特性及决策过程对人群疏散的影响,弥补现有的人群疏散模型的不足。仿真结果证明,Agent的行为建模技术具有仿真火灾全过程中人员疏散行为的功能,适用于建筑物火灾中的人群优化疏散策略。

Building fires frequently happen in our country, therefore, analyzing the crowd evacuation discipline is of great instruction meaning for decreasing occupant casualty. Considering that human behavior factors are not analyzed adequately in existing researches, how to apply the agent - based crowd behavior modeling technologies to model crowd evacuation in building fires was studied preliminary. The results of simulation demonstrate that the agent - based behavior model can reflect the effects of individual properties, the decision making processes of humans, and the fire productions in the crowd evacuation scenarios, which can make up the limitations of the most existing crowd evacuation models. At last, through the analysis of the simulation results, it is concluded that the agent - based behavior modeling technologies are qualified for modeling the whole process of the fire in buildings and appropriate for analyzing the crowd evacuation discipline.