An emergency evacuation simulation model for pedestrians at metro station hall was established by three kinds of agents, that is, turtles, patches and observer, whose simulation environment was built using the multi-agent simulation platform named NetLogo. Through programming, some walking strategies of pedestrians were realized, which include the detection, obstacle avoidance, random swing, walking to destination and so on. Then, the personnel walking path was planned and the simulation process was advanced, which helps study the influence of the number of pedestrians at the station and exit conditions on the emergency evacuation. The simulation results show that the evacuation time increases linearly with the number of pedestrians at the station hall and upsurges after such number exceeds a certain threshold. The evacuation efficiency under four outlets is significantly higher than that under two outlets, and the evacuation efficiency of a single exit is similar to each other under either situation. Meanwhile, the probability distributions about the total number of pedestrians passing through all outlets in one unite time are tested to obey the normal distribution. Moreover, the evacuation time decreases rapidly with the increase of the exit width and obtains little decease after the width reaches a certain value. But it increases linearly with the speed of outputting persons from the metro station platform and jumps after such speed exceeds a certain value. These conclusions provide some feasible suggestions for designing the metro station, emergency evacuation and emergency command for pedestrians.%基于多智能体仿真平台NetLogo搭建仿真环境,利用海龟(Turtles)、瓦片(Patches)和观察员(Observer)3类智能体建立地铁车站站厅人员紧急疏散仿真模型.通过编程实现主体的探测、躲避障碍物、随机摆动、走向目的地等行走策略,规划人员的行走路径,推进仿真进程,研究站厅人员数量、出口条件、有站台输出人员对紧急疏散的影响.仿真结果表明:疏散时间随站厅人员数量线性递增且在超过某阈值后增长率突然变大;开放4个出口的疏散效率显著高于开放2个出口的情形且每种情形单个出口的疏散效率相似;2种情形下单位时间通过出口总人数的频次概率分布符合正态分布;疏散时间随出口宽度增加而迅速减小,宽度增至某值后减小速率趋于平缓;疏散时间随站台输出人员速率线性递增,人员生成速率超过某值后疏散时间迅速增大.本研究所得结论可为地铁车站站厅设计、人员紧急疏散和应急指挥提供可行性建议.
展开▼
