基于动态视觉算法的人员疏散预动作时间预测模型

Predictive model for pre-movement time of individuals based on dynamic vision algorithms

在元胞自动机的基础上,提出一种考虑视觉遮挡的动态视觉算法,结合视觉算法建立了预动作时间模型。从跟随效应的角度出发,在作用机理的层面上解释预动作时间在不同布局场景内产生分布差异的原因。通过试验验证了模型的合理性并探究了场景中不同因素对人群预动作时间的影响,试验结果表明:(1)本模型能够体现不同类型疏散场景下人群预动作时间变化的趋势,且各场景的仿真试验结果与真实试验结果数值相近,本模型具备一定合理性;(2)在相同场景内,提高室内空间的视野通透性可以将人群的预动作时间均值缩短7.05%;(3)疏散过程中当引导员在人群中的占比从0达到20%时,人群的预动作时间均值缩短23.17 s;(4)在单房间场景中,人群密度从0.1人/m^(2)提升至0.9人/m^(2),人群预动作时间均值缩短9.66 s。提出的模型为疏散预动作时间的生成提供了一种新理论,为建筑疏散优化提供了一种新方法。

Based on cellular automata,a dynamic visual algorithm considering visual obstruction was proposed,and a pre-movement time model was established in combination with the dynamic visual algorithm.Different from explaining the distribution of pre-movement time from an inductive perspective of data,the present paper intended to approach the subject from the perspective of the following effect,the reason for the distribution difference of pre-movement time in different layout scenarios was explained at the level of the action mechanism.The rationality of the model was verified by simulating evacuations in different types of building scenarios and comparing the results with real experimental data.Based on the proposed model,the influence of visual transparency,evacuation guiders,and crowd density on the pre-movement time was investigated by conducting controlled variable tests,the variables were whether or not obstructions obstructed the vision,the proportion of evacuation guides,and the density of the crowd,and the layout of the scenes in each group of test remained consistent.The test results show that:(1)The model can reflect the trend of pre-movement time changes of the crowd under different types of evacuation scenarios,and the simulation results of each scenario were close to the real test results,indicating that the model had a certain rationality.(2)The effect of visual transparency was not obvious:increasing the visual transparency reduced the average pre-movement time from 67.53 s to 62.77 s,reducing by 7.05%.(3)The effect of guiders was significant:when the proportion of guiders increases from 0 to 20%,the average pre-movement time was reduced from 53.34 s to 30.17 s,reducing by 43.44%.(4)Increasing the crowd density can help to reduce pre-movement time:in a single room scenario,when the crowd density increases from 0.1 person/m^(2) to 0.9 person/m^(2),the average pre-movement time of the crowd is shortened by 9.66 s,reducing by 17.45%.