一种考虑斜向运动参量的行人疏散模型研究

A simulated pedestrians' evacuation model with a diagonal movement parameter

目前,元胞自动机疏散模型采用Von Neumann型和Moore型两种形式的邻域进行演化处理。这两种邻域形式描述行人疏散均有其不足。Von Neumann型邻域忽略了行人斜向运动的可能性,而Moore型邻域虽然考虑了斜向运动,却不能有效地区分斜向运动与水平竖直运动的区别。基于此,在选择Moore型邻域的基础上,提出了斜向运动参量的概念,重新阐述了疏散演化规则,修正了原Moore型邻域在描述疏散中的不足,更精确地反映了行人的疏散动力学特征。进一步利用新模型模拟仿真了疏散空间的行人疏散过程,结果表明:斜向运动参量越小,行人选择水平竖直方向运动的可能性越大;疏散时间会随斜向运动参量f值减小而增加。斜向运动参量取0.7左右时,疏散效果最佳;而当斜向运动参量取0时,Moore型邻域的疏散结果演化为Von Neumann型邻域的疏散结果。

This evacuation model is based on the cellular automata including two kinds of cellular neighborhoods that are the Von Neumann neighborhood and Moore neighborhood. However, it is easy to find the defects of the two neighborhoods when they were used to describe the pedestrian evacuation. In Von Neumann neighborhood, the diagonal movement can be ignored, for the diagonal movement can be taken as that of the Moore neighborhood, but it cannot be distinguished between the horizontal and vertical movement in the diagonal movement. Based on the defects of the two neighborhoods, the definition of diagonal-movement parameters can be introduced on the basis of Moore neighborhood. Then the novel evacuation method has been proposed, with the novel evolution rules for the pedestrians illustrated, which can modify the Moore neighborhood and describe the dynamic characteristics of the pedestrian evacuation more accurately. Moreover, the pedestrian evacuation and the dynamic feature of the pedestrian escapers can be simulated by using the novel model. Comparing the diagonal-movement parameters before and after the simulation is done, it would be necessary to illustrate the influence and significance of the diagonal-movement parameters. For, the results of our simulation show that, in differentiating the initial pedestrian density, the smaller the diagonal-movement parameter is, the greater the possibility of the horizontal and vertical movement will be, and the longer the evacuation time will it take. For example, when the initial pedestrian density e is 0.095 3 per m 2 , the time difference between the longest and the shortest steps is 16 times. By comparing the pedestrian evacuation in different values of the parameter f , the optimal evacuation pattern can be found, which shows that the pedestrian evacuation can be described by using the evacuation model with about 0.7 parameter value of f . More simulation results show that when the diagonal-movement parameter f is zero, the evacuation model of Moore neighborhood should be mainly the same as the evacuation model of Von Neumann neighborhood, which means that the Moore neighborhood and Von Neumann neighborhood can be regarded as the special cases for f =1.0 and 0.0, respectively.