CN106096168A

CN106096168A - A kind of space based on Auditory Perception crowd evacuation analogy method

Info

Publication number: CN106096168A
Application number: CN201610452004.2A
Authority: CN
Inventors: 武悦; 康健; 张姗姗; 朱丽玮; 王超; 薛明辉; 付本臣
Original assignee: Harbin Institute of Technology Shenzhen
Current assignee: Harbin Institute of Technology Shenzhen
Priority date: 2016-06-21
Filing date: 2016-06-21
Publication date: 2016-11-09
Anticipated expiration: 2036-06-21
Also published as: CN106096168B

Abstract

The invention relates to a simulation method for crowd evacuation in space based on auditory perception, and the invention relates to a simulation method for crowd evacuation in space based on auditory perception. The present invention aims to solve the problem that the existing evacuation simulation method does not consider the influence of auditory elements such as the number and location of sirens. The steps of the present invention are as follows: Step 1, dividing the building plane into grids, the divided grids are divided into two situations, one is occupied by people or obstacles, and the other is empty; Step 2, setting the grid Personnel moving speed and direction; step 3, get the probability that person i moves to grid j in the next time step according to the auditory perception cellular automata model; step 4, calculate the movement probability of all personnel in the grid, The entire evacuation process from the current time position to the exit is simulated. The invention is applied in the field of building safety evaluation.

Description

A spatial crowd evacuation simulation method based on auditory perception

技术领域technical field

本发明涉及基于听觉感知的空间人群疏散模拟方法。The invention relates to a simulation method for spatial crowd evacuation based on auditory perception.

背景技术Background technique

目前，人员疏散模型的建模方法大体上可分为两种：一种是宏观的方法，即把行人视为连续流动介质，因为现代人员疏散研究是从交通流的研究中分化出来的，因而也就很自然地继承了流体研究中己经完善和成熟的方法。最早的宏观模型是由Henderson提出的，他认为行人的运动行为类似于气体或液体的流动，行人行为的气态动力学方程与Boltzmann方程相似，不过它考虑了行人之间的相互影响和行人的目的。Hughes采用连续介质理论(continuum theory)研究大型人群的运动特征，并根据Navier-Stokes方程进一步推导出大型人群流动的控制方程和行人避免向高密度人群运动的方程，该模型成功地解释了麦加朝圣中的人群的运动状况。然而人群是不遵守动量和能量守恒的，宏观模型的缺点在于没有将人员的相互作用考虑在内，因此不适于紧急情况下对行人流进行研究。宏观模型忽略了个体之间的差异，于是研究者又提出一种新的模型，即微观模型。At present, the modeling methods of the personnel evacuation model can be roughly divided into two types: one is the macroscopic method, that is, the pedestrian is regarded as a continuous flow medium, because the modern personnel evacuation research is differentiated from the traffic flow research, so It also naturally inherits the perfect and mature methods in fluid research. The earliest macroscopic model was proposed by Henderson. He believed that the motion behavior of pedestrians was similar to the flow of gas or liquid. The gas dynamic equation of pedestrian behavior was similar to the Boltzmann equation, but it considered the interaction between pedestrians and the purpose of pedestrians. . Hughes used continuum theory to study the movement characteristics of large crowds, and further deduced the control equation for large crowd flow and the equation for pedestrians to avoid moving to high-density crowds according to the Navier-Stokes equation. This model successfully explained Mecca The movement status of the crowd during the pilgrimage. However, the crowd does not obey the conservation of momentum and energy. The shortcoming of the macro model is that it does not take the interaction of people into account, so it is not suitable for the study of pedestrian flow in emergency situations. The macro model ignores the differences between individuals, so the researchers proposed a new model, the micro model.

另一种微观的方法，它把行人视为相互作用的粒子，其中最为著名的就是Helbing的社会力模型。微观模型可以对行人流的具体行为进行描述，近年来引起了极大的关注。该模型的研究方法主要有连续型和离散型两种。其中具有代表性的模型有社会力模型、磁场力模型、元胞自动机模型等和格子气(Lattice Gas)模型。Another microscopic approach, which treats pedestrians as interacting particles, the most famous of which is Helbing's social force model. Microscopic models can describe the specific behavior of pedestrian flows and have attracted great attention in recent years. There are mainly two types of research methods for this model: continuous and discrete. The representative models include social force model, magnetic field force model, cellular automata model, etc. and Lattice Gas model.

发明内容Contents of the invention

本发明是为了解决现有疏散方法中没有考虑听觉要素如警报器的数量和位置的影响的问题，而提出的一种基于听觉感知的空间人群疏散模拟方法。The present invention proposes a spatial crowd evacuation simulation method based on auditory perception to solve the problem that the influence of auditory elements such as the number and position of sirens is not considered in the existing evacuation methods.

一种基于听觉感知的空间人群疏散模拟方法按以下步骤实现：A spatial crowd evacuation simulation method based on auditory perception is implemented in the following steps:

步骤一：将建筑物平面进行网格划分，划分后的网格分为两种情况，一种为被人或障碍物占据，一种为空；Step 1: Divide the building plane into grids. The divided grids are divided into two situations, one is occupied by people or obstacles, and the other is empty;

步骤二：设置网格中人员移动速度和方向；Step 2: Set the moving speed and direction of people in the grid;

步骤三：依据听觉感知元胞自动机模型得到人员i在下一时间步移动到网格j的概率为：Step 3: According to the auditory perception cellular automata model, the probability that person i moves to grid j in the next time step is obtained as:

${P P}_{i i,, j j} = = \frac{exp exp {{{k k}_{S S} \cdot &Center Dot; {S S}_{i i,, j j} + + {k k}_{D D.} \cdot &Center Dot; {D D.}_{i i,, j j}}} \cdot &Center Dot; (({n no}_{max max} - - {n no}_{j j}))}{{Σ Σ}_{j j &Element; &Element; M m} [[exp exp {{{k k}_{S S} \cdot &Center Dot; {S S}_{i i,, j j} + + {k k}_{D D.} \cdot &Center Dot; {D D.}_{i i,, j j}}} \cdot &Center Dot; (({n no}_{max max} - - {n no}_{j j}))]]}$

n_max和n_j为网格j最大容纳人数和当前已容纳人数，n_max＝n_j时表示网格j被障碍物占据，k_s为静态信息吸引力系数，k_D为动态信息吸引力系数；n _max and n _j are the maximum capacity of grid j and the number of people currently accommodated. When n _max = n _j , it means that grid j is occupied by obstacles. k _s is the coefficient of attraction of static information, and k _D is the coefficient of attraction of dynamic information ;

S_i,j为网格j对人员i的静态信息吸引力，D_i,j为网格j对人员i的动态信息吸引力，M为建筑物平面分割的网格数；S _i,j is the static information attraction of grid j to person i, D _i,j is the dynamic information attraction of grid j to person i, and M is the number of grids divided by the building plane;

步骤四、对网格中所有人员的移动概率进行计算，对人员在当前时间位置至到达出口的整个疏散过程进行模拟。Step 4: Calculating the movement probability of all personnel in the grid, and simulating the entire evacuation process of the personnel from the current time position to the exit.

发明效果：Invention effect:

本发明提出了基于听觉感知的疏散模型，模型原型为元胞自动机模型。本发明考虑了听觉要素如警报器的数量和位置的影响，并通过声场测试和疏散实验来验证确定模型的变量。模型的精确度通过体育场内观测实验来比对得到确定。模型不仅能用来预计疏散时间，而且还能帮助分析和改进建筑的扬声器布局对疏散的影响，评估建筑在紧急情况发生时人员疏散的安全性，利用模型给出相应的疏散设计和预警方案。本发明算法与测量值的误差小于10％。The invention proposes an evacuation model based on auditory perception, and the prototype of the model is a cellular automata model. The invention considers the influence of auditory elements such as the number and position of sirens, and verifies the variables of the determination model through sound field tests and evacuation experiments. The accuracy of the model was confirmed by comparison with in-stadium observation experiments. The model can not only be used to predict the evacuation time, but also help to analyze and improve the impact of the loudspeaker layout of the building on the evacuation, evaluate the safety of the evacuation of the building when an emergency occurs, and use the model to give the corresponding evacuation design and early warning scheme. The error between the algorithm of the invention and the measured value is less than 10%.

附图说明Description of drawings

图1为听觉感知元胞自动机模型移动概率分布示意图；Figure 1 is a schematic diagram of the movement probability distribution of the auditory perception cellular automata model;

图2为疏散观测实验出口位置示意图；Figure 2 is a schematic diagram of the exit position of the evacuation observation experiment;

图3为各出口疏散人数方差比对图；Figure 3 is a comparison chart of the variance of the number of evacuated people at each exit;

图4为疏散初始人员位置示意图；Figure 4 is a schematic diagram of the position of the initial evacuation personnel;

图5为疏散进行至20s时的人员位置示意图；Figure 5 is a schematic diagram of the personnel position when the evacuation is carried out to 20s;

图6为疏散进行至60s时的人员位置示意图；Fig. 6 is a schematic diagram of personnel positions when evacuation is carried out to 60s;

图7为疏散进行至100s时的人员位置示意图；Figure 7 is a schematic diagram of the personnel position when the evacuation is carried out to 100s;

图8为疏散进行至140s时的人员位置示意图；Figure 8 is a schematic diagram of the personnel position when the evacuation is carried out to 140s;

图9为疏散进行至180s时的人员位置示意图；Fig. 9 is a schematic diagram of personnel positions when evacuation is carried out to 180s;

图10为计算机模拟实验中通过各个出口的人员数量的方差对比图；Fig. 10 is a variance comparison chart of the number of personnel passing through each exit in the computer simulation experiment;

图11为单位时间内通过各个出口的疏散人数平均值图。Figure 11 is a graph of the average number of evacuated people passing through each exit per unit time.

具体实施方式detailed description

具体实施方式一：一种基于听觉感知的空间人群疏散模拟方法包括以下步骤：Specific embodiment one: a kind of spatial crowd evacuation simulation method based on auditory perception comprises the following steps:

模型基于同元胞自动机基本模型，对建筑物的平面进行均匀的网格划分，每个网格被障碍物占据、或被人员占据、或为空。元胞自动机模型中每个网格对应空间为边长为0.5m的正方形，但应用在大空间建筑内时，网格大小可根据建筑物尺寸做出相应调整，每个网格也可以根据人员密度不再限定为一人，但需要注意的是为了简化计算处在同一个网格(元胞)中的人员彼此不受影响即可。人员每次移动最多只能行进一格。The model is based on the basic model of cellular automata, and the plane of the building is uniformly divided into grids, and each grid is occupied by obstacles, occupied by people, or empty. In the cellular automata model, the space corresponding to each grid is a square with a side length of 0.5m, but when it is applied in a large space building, the size of the grid can be adjusted according to the size of the building, and each grid can also be adjusted according to the size of the building. The density of personnel is no longer limited to one person, but it should be noted that personnel in the same grid (cell) are not affected by each other in order to simplify the calculation. Personnel can only move up to one space at a time.

${P P}_{i i,, j j} = = \frac{exp exp {{{k k}_{S S} \cdot &Center Dot; {S S}_{i i,, j j} + + {k k}_{D D.} \cdot &Center Dot; {D D.}_{i i,, j j}}} \cdot &Center Dot; (({n no}_{max max} - - {n no}_{j j}))}{{Σ Σ}_{j j &Element; &Element; M m} [[exp exp {{{k k}_{S S} \cdot \cdot {S S}_{i i,, j j} + + {k k}_{D D.} \cdot \cdot {D D.}_{i i,, j j}}} \cdot \cdot (({n no}_{max max} - - {n no}_{j j}))]]}$

式中P_i，j表示人员i在下一时间步移动到网格j的概率，在人员所处邻域内有 In the formula, P _i,j represents the probability that person i moves to grid j in the next time step, and there is

n_max和n_j为网格j最大容纳人数和当前已容纳人数，n_max＝n_j时表示网格j被墙壁等障碍物占据，k_s为静态信息吸引力系数，k_D为动态信息吸引力系数；n _max and n _j are the maximum capacity of grid j and the number of people currently accommodated. When n _max = n _j , it means that grid j is occupied by obstacles such as walls, k _s is the static information attraction coefficient, and k _D is the dynamic information attraction force coefficient;

S_i,j为网格j对人员i的静态信息吸引力，D_i,j为网格j对人员i的动态信息吸引力，M为建筑物平面分割的网格数，根据建筑物平面尺寸确定。S _i,j is the static information attraction of grid j to person i, D _i,j is the dynamic information attraction of grid j to person i, M is the number of grids divided by the building plane, according to the building plane size Sure.

具体实施方式二：本实施方式与具体实施方式一不同的是：所述网格为0.5m的正方形，每个网格中大于等于1人小于等于5人。其它步骤及参数与具体实施方式一相同。Embodiment 2: This embodiment differs from Embodiment 1 in that: the grid is a 0.5m square, and each grid has 1 person or more and 5 people or less. Other steps and parameters are the same as those in Embodiment 1.

其它步骤及参数与具体实施方式一相同。Other steps and parameters are the same as those in Embodiment 1.

具体实施方式三：本实施方式与具体实施方式一或二不同的是：所述步骤二中网格中人员移动速度和方向具体为：Specific implementation mode three: the difference between this implementation mode and specific implementation mode one or two is: the moving speed and direction of the personnel in the grid in the step two are specifically:

在模型中采用Von Neumann邻域，即人员能向自身周围前后左右四个方向运动，如图1所示。The Von Neumann neighborhood is used in the model, that is, people can move in four directions around themselves, front, back, left, and right, as shown in Figure 1.

设置人员移动速度为1m/s～2m/s，人员在所在位置向和四个位置移动。Set the moving speed of personnel to 1m/s~2m/s, location to and Four positions move.

人正常行走的速度约为1m/s，但在紧急情况下快速行走速度可达1.5m/s，而小跑的速度可以达到2m/s。另外个人的身体素质差异较大，年轻人的行动速度高于老人，男性高于女性；行李和随行的老人孩子都会使人员行动能力减慢。为了体现这些差别，模型里人员的最大速度可以设置两档：2m/s和1m/s。分别代表一个时间步移动一次和两个时间步移动一次。The normal walking speed of a person is about 1m/s, but in an emergency, the fast walking speed can reach 1.5m/s, and the trotting speed can reach 2m/s. In addition, there is a big difference in the physical fitness of individuals. The action speed of young people is higher than that of old people, and that of men is higher than that of women; luggage and accompanying old people and children will slow down the mobility of personnel. In order to reflect these differences, the maximum speed of the person in the model can be set in two gears: 2m/s and 1m/s. Respectively represent one time step move once and two time step move once.

其它步骤及参数与具体实施方式一或二相同。Other steps and parameters are the same as those in Embodiment 1 or Embodiment 2.

具体实施方式四：本实施方式与具体实施方式一至三之一不同的是：所述步骤三中k_D具体为：Embodiment 4: The difference between this embodiment and one of Embodiments 1 to 3 is that k _D in the step 3 is specifically:

高能见度无声音信息指令情况下k_D＝0；高能见度有声音信息指令情况下k_D＝0.84；低能见度无声音信息指令情况下k_D＝1.43；低能见度有声音信息指令情况下k_D＝1.2。高能见度为大于等于5米，低能见度为小于5米。k _D ＝0 in the case of high visibility without voice information command; k _D ＝0.84 in the case of high visibility with voice information command; _k _D ＝1.43 in the case of low visibility without voice information command; 1.2. High visibility is greater than or equal to 5 meters, and low visibility is less than 5 meters.

其它步骤及参数与具体实施方式一至三之一相同。Other steps and parameters are the same as those in Embodiments 1 to 3.

具体实施方式五：本实施方式与具体实施方式一至四之一不同的是：所述步骤三中k_s具体为：Specific implementation mode five: the difference between this implementation mode and one of the specific implementation modes one to four is that k _s in the step three is specifically:

当网格j是出口时，k_s＝1表示该因素对疏散个体有吸引力；当网格j是墙壁等阻碍物时，k_s＝-1表示该因素对疏散个体有排斥力；k_s＝0表示不考虑该因素。When grid j is an exit, k _s =1 means that this factor is attractive to evacuated individuals; when grid j is an obstacle such as a wall, k _s =-1 means that this factor is repulsive to evacuated individuals; k _s = 0 indicates that this factor is not considered.

其它步骤及参数与具体实施方式一至四之一相同。Other steps and parameters are the same as in one of the specific embodiments 1 to 4.

具体实施方式六：本实施方式与具体实施方式一至五之一不同的是：所述步骤三中S_i,j的表达式为：Specific embodiment six: the difference between this embodiment and one of the specific embodiments one to five is: the expression of S _i,j in the step three is:

${S S}_{i i,, j j} = = {a a}_{i i,, {e e}_{k k}} . . {r r}_{i i,, j j}$

为阻碍系数,，表示人员i到出口e_k这段距离中墙和楼梯对人员的阻碍程度； is the obstruction coefficient, which indicates the obstruction degree of the wall and stairs to personnel in the distance from personnel i to exit e _k ;

${a a}_{i i,, {e e}_{k k}} = = α α \frac{{r r}_{f f}}{{r r}_{i i,, j j}} + + β β \frac{{r r}_{s the s}}{{r r}_{i i,, j j}}$

α和β为阻碍系数，r_f是网格j到出口e_k这段距离中的平地长度,r_s是网格j到出口e_k这段距离中的楼梯长度，元胞所在位置距离出口越近，对疏散人员的吸引力就越大，r_i,j是网格j到出口e_k的距离，采用坐标差值法通过网格j到出口e_k的距离来确定；α and β are resistance coefficients, r _f is the length of flat ground in the distance from grid j to exit e _k , r _s is the length of stairs in the distance from grid j to exit e _k , the closer the cell is located to the exit The closer, the greater the attraction to the evacuated personnel, r _i,j is the distance from grid j to exit e _k , which is determined by the distance from grid j to exit e _k by the coordinate difference method;

r_i,j表示为：r _i,j is expressed as:

${r r}_{i i,, j j} = = {max max}_{((i i,, j j))} {{{min min}_{{e e}_{k k}} [[| | {x x}_{j j} - - {x x}_{{e e}_{k k}} | | + + | | {y the y}_{j j} - - {y the y}_{{e e}_{k k}} | |]]}} - - {min min}_{{e e}_{k k}} [[| | {x x}_{j j} - - {x x}_{{e e}_{k k}} | | + + | | {y the y}_{j j} - - {y the y}_{{e e}_{k k}} | |]]$

设建筑物中共有k个出口，式中表示出口e_k的位置坐标；(x_j,y_j)表示网格j的位置坐标；Suppose there are k exits in the building, where Indicates the position coordinates of exit e _k ; (x _j , y _j ) indicates the position coordinates of grid j;

式中等式右侧第一项表示所有格点距各出口最远距离，右侧第二项表示元胞j到各出口的最小值。这样能够保证距离最近出口最远处的元胞对人员的吸引力为0。The first term on the right side of the equation represents the furthest distance from all grid points to each exit, and the second term on the right represents the minimum value from cell j to each exit. This can ensure that the attractiveness of the cell farthest from the nearest exit to personnel is 0.

其它步骤及参数与具体实施方式一至五之一相同。Other steps and parameters are the same as one of the specific embodiments 1 to 5.

具体实施方式七：本实施方式与具体实施方式一至六之一不同的是：所述步骤三中D_i,j的表达式为：Specific embodiment seven: the difference between this embodiment and one of the specific embodiments one to six is: the expression of D _{i, j} in the step three is:

${D D.}_{i i,, j j} = = {b b}_{i i . . j j} \cdot &Center Dot; {f f}_{i i,, {w w}_{k k}} - - - - - - ((t t))$

b_i,j体现影响人员对报警信号做出的反应，其取值采用Weibull分布；表示人员i在网格j内所受到的噪音干扰，表示为：b _i,j reflect the response of the affected personnel to the alarm signal, and its value adopts Weibull distribution; Indicates the noise interference experienced by person i in grid j, expressed as:

${f f}_{i i,, {w w}_{k k}} ((t t)) = = {g g}_{i i,, j j} ((t t)) \cdot &Center Dot; [[{θ θ}_{i i,, j j} \cdot &Center Dot; {l l}_{i i,, {w w}_{k k}} ((t t))]]$

g_i,j(t)为能见度系数，0＜g_i,j(t)＜1，使用Gaussian分布：g _i,j (t) is the visibility coefficient, 0<g _i,j (t)<1, using Gaussian distribution:

${g g}_{i i,, j j} ((t t)) = = \frac{11}{\sqrt{22 π π} σ σ} {e e}^{\frac{{((x x - - μ μ))}^{22}}{22 {σ σ}^{22}}} ((σ σ > > 00))$

为人员i在当前时刻能够接收到的警报声信息的权重，警报声信息权重取决于当前位置的声压级L_p，警报器为点声源，声功率级为L_w，L_p为当前位置的声压级，当前位置声压级随测点距声源距离的变化为： is the weight of the alarm sound information that person i can receive at the current moment, the weight of the alarm sound information depends on the sound pressure level L _p of the current position, the siren is a point sound source, the sound power level is L _w , and L _p is the current position The sound pressure level of the current position varies with the distance from the measuring point to the sound source as follows:

${L L}_{p p} = = {L L}_{w w} + + 1010 lg lg ((\frac{Q Q}{44 {πr πr}^{22}} + + \frac{44}{R R}))$

其中，Q是指向性因子，r是声源距测点距离，R是房间常数,其中s是室内总表面积，室内平均吸声系数；因此表示为：Among them, Q is the directivity factor, r is the distance from the sound source to the measuring point, R is the room constant, where s is the total surface area of the interior, The average sound absorption coefficient of the room; therefore Expressed as:

${l l}_{i i,, {w w}_{k k}} ((t t)) = = l l n no [[{L L}_{w w} + + 1010 lg lg ((\frac{Q Q}{44 {πr πr}^{22}} + + \frac{44}{R R}))]]$

其它步骤及参数与具体实施方式一至六之一相同。Other steps and parameters are the same as one of the specific embodiments 1 to 6.

实施例一：Embodiment one:

实验在哈尔滨工业大学体育馆中进行。体育馆一层东西两侧共有A、B、C、D四个出口用来疏散，如图2所示。本次实验共有1127名实验对象参加，均匀分布在一层体育场大厅内。根据经验疏散开始后的预判时间设置为10s，表示在接收到疏散指令到开始疏散的时间是10s。The experiment was carried out in the gymnasium of Harbin Institute of Technology. There are four exits A, B, C, and D on the east and west sides of the first floor of the gymnasium for evacuation, as shown in Figure 2. A total of 1127 subjects participated in this experiment, which were evenly distributed in the stadium hall on the first floor. According to experience, the pre-judgment time after the evacuation starts is set to 10s, which means that the time from receiving the evacuation instruction to the start of evacuation is 10s.

通过各出口疏散的人数以及各出口疏散时间被记录下来，疏散结果表明，总疏散用时是3分52秒，每个出口成功疏散人数非常接近。A出口疏散270人，疏散时间3分40秒；B出口疏散284人，疏散时间3分10秒；C出口疏散330人，疏散时间4分；D出口疏散274人，疏散时间3分30秒。出口B时间略短于其他出口，出口D的波动性显示高于其他出口。图3显示了各个出口通行人员数量的方差。虽然每个出口的疏散过程是相似的，但其疏散能力是不均衡的。疏散过程分为四个阶段。第一阶段是疏散开始至20s，通过出口的人数较少，但随着时间逐渐增加。随着越来越多的人往出口方向聚集以及出口通行能力的限制，疏散人数增加的趋势变得缓慢，达到3人/秒后不再增加。第三阶段是100s以后，疏散人员数量的方差开始保持不变，接近1.5人/秒。180s后进入第四阶段，疏散人员数量开始下降直至疏散过程结束。The number of people evacuated through each exit and the evacuation time of each exit were recorded. The evacuation results showed that the total evacuation time was 3 minutes and 52 seconds, and the number of people successfully evacuated at each exit was very close. 270 people were evacuated from Exit A, and the evacuation time was 3 minutes and 40 seconds; 284 people were evacuated from Exit B, and the evacuation time was 3 minutes and 10 seconds; 330 people were evacuated from Exit C, and the evacuation time was 4 minutes; 274 people were evacuated from Exit D, and the evacuation time was 3 minutes and 30 seconds. The time of exit B is slightly shorter than other exits, and the volatility of exit D is higher than other exits. Figure 3 shows the variance of the number of people passing through each exit. Although the evacuation process of each exit is similar, its evacuation capacity is uneven. The evacuation process is divided into four stages. The first stage is from the beginning of evacuation to 20s, the number of people passing through the exit is small, but gradually increases with time. As more and more people gather toward the exit and the exit capacity is limited, the increase in the number of evacuated people slows down and stops after reaching 3 people/second. The third stage is after 100s, and the variance of the number of evacuated personnel begins to remain constant, close to 1.5 persons/second. After 180s, it enters the fourth stage, and the number of evacuated personnel begins to decrease until the evacuation process ends.

计算机模拟的实验数据录入与疏散实验相同，首先用CAD画出体育馆平面图，接下来录入人员位置，人数1127人平均分布在体育馆一层，预判时间设置为10s，人员移动速度1m/s。图4-图9显示了利用听觉感知元胞自动机模型模拟不同时间段的疏散过程，分别是疏散过程刚开始和进行到20s、60s、100s、140s和180s时的人员位置。The experimental data entry of the computer simulation is the same as that of the evacuation experiment. First, CAD is used to draw the floor plan of the gymnasium, and then the personnel positions are entered. The number of people is 1127, which are evenly distributed on the first floor of the gymnasium. The prediction time is set to 10s, and the movement speed of personnel is 1m/s. Figures 4-9 show the evacuation process simulated by the auditory perceptual cellular automata model at different time periods, which are the positions of personnel at the beginning of the evacuation process and at 20s, 60s, 100s, 140s and 180s, respectively.

从过程图中可以看出，各个出口疏散的人数基本平均。疏散总用时为3分19秒。模拟结果同时还得出了每10秒内各个出口通行的人数已经各出口的总通行人数，详见表1。It can be seen from the process diagram that the number of people evacuated from each exit is basically average. The total evacuation time was 3 minutes and 19 seconds. The simulation results also show the number of people passing each exit and the total number of people passing each exit every 10 seconds, see Table 1 for details.

表1 计算机模拟实验每10s内各出口通行人数Table 1 The number of people passing each exit in every 10s of the computer simulation experiment

图10显示了模拟实验中通过各个出口的人员数量的方差，这说明了各个出口疏散的人员数量越接近平均值，疏散过程所用的时间越短。疏散过程是在疏散的起始时间疏散的人数很低，经过一个快速的上升期后持续一段时间，再平稳的下降。Figure 10 shows the variance of the number of people passing through each exit in the simulation experiment, which shows that the closer the number of people evacuated by each exit is to the average value, the shorter the time taken for the evacuation process. The evacuation process is that the number of evacuated people is very low at the beginning of the evacuation, after a period of rapid rise, it lasts for a period of time, and then declines steadily.

观测实验所得的数据，每十秒疏散人数的方差被计算用以与听觉感知元胞自动机模型模拟所得数据进行对比。数值小于1的占22％，数值大于1小于3的占21％，数值大于3小于5的占27％，数值大于5小于10的占26％，而数值大于10的为3％。最大值是15，最小值是0(意味着等于平均值)。方差值大于5的说明在该单位时间内疏散的人数明显的不同于平均值。观测实验与模拟实验结果的最大差值用以与平均疏散人数进行比较，其所占百分比越小表示精确度越高。通过出口A的人数最大差值出现在190s，差值人数为9人，占平均疏散人数的9.4％。通过出口B的人数中最大差值出现在80s，差值人数为7人，占平均疏散人数的15.2％。通过出口C的人数中最大差值出现在120s，差值人数为10人，占平均疏散人数的11.1％。通过出口D的人数中，最大差值出现在40s，差值人数为15人，占平均疏散人数的11.1％。The data obtained from the observation experiment, and the variance of the number of evacuated people every ten seconds were calculated to compare with the simulated data from the auditory perception cellular automata model. 22% had values less than 1, 21% had values greater than 1 and less than 3, 27% had values greater than 3 and less than 5, 26% had values greater than 5 and less than 10, and 3% had values greater than 10. The maximum value is 15 and the minimum value is 0 (meaning equal to the average). A variance value greater than 5 indicates that the number of people evacuated in this unit time is significantly different from the average value. The maximum difference between the observation experiment and the simulation experiment results is used to compare with the average number of evacuated people, and the smaller the percentage, the higher the accuracy. The largest difference in the number of people passing through Exit A occurred in the 190s, and the difference was 9 people, accounting for 9.4% of the average number of evacuated people. The largest difference in the number of people passing through Exit B appeared in the 80s, with a difference of 7 people, accounting for 15.2% of the average number of evacuated people. The largest difference in the number of people passing through Exit C appears at 120s, and the difference is 10 people, accounting for 11.1% of the average number of evacuated people. Among the number of people passing through Exit D, the largest difference occurred in the 40s, and the difference was 15 people, accounting for 11.1% of the average number of evacuated people.

图11所示为观测实验和模拟实验中单位时间内通过各个出口的疏散人数平均值比较，最大差值出现在100s，为25人，所占平均疏散人数的10.8％。其他值较为接近，差值低于所占平均疏散人数的10％的比例为80％。说明利用听觉感知元胞自动机模型模拟的各个出口的模拟结果接近平均值。听觉感知元胞自动机模拟的结果中，模拟显示的疏散过程在20s以前和160s以后与观测实验的疏散过程是极为相近的。其区别在于元胞自动机没能很好的模拟在出口处出现拥堵的情况，出现较大差距的地方主要原因是对于瓶颈现象缺乏考虑造成的。然而，这些差距是在一个可接受的范围内。因此利用听觉感知元胞自动机模型的模拟结果是可信的。Figure 11 shows the comparison of the average number of evacuated people passing through each exit per unit time in the observation experiment and the simulated experiment. The maximum difference appears in 100s, which is 25 people, accounting for 10.8% of the average number of evacuated people. The other values are closer, with 80% of the difference being less than 10% of the average evacuation population. It shows that the simulation results of each outlet simulated by the auditory perception cellular automata model are close to the average value. In the results of the auditory perception cellular automaton simulation, the evacuation process shown by the simulation is very similar to the evacuation process of the observation experiment before 20s and after 160s. The difference is that the cellular automata cannot simulate the congestion at the exit well, and the main reason for the large gap is the lack of consideration of the bottleneck phenomenon. However, these gaps are within an acceptable range. Therefore, the simulation results using the cellular automata model of auditory perception are credible.

Claims

1. space based on an Auditory Perception crowd evacuation analogy method, it is characterised in that described sky based on Auditory Perception Between crowd evacuation method comprise the following steps:

Step one: building plane is carried out stress and strain model, the grid after division is divided into two kinds of situations, and one is by people or obstacle Thing occupies, and one is empty；

Step 2: personnel's translational speed and direction in grid are set；

Step 3: obtain the personnel i probability in future time moved further to grid j according to Auditory Perception cellular Automation Model For:

n_maxAnd n_jFor grid j maximum galleryful and current galleryful, n_max=n_jTime represent that grid j is occupied by barrier, k_SFor static information captivation coefficient, k_DFor multidate information captivation coefficient；

S_i,jFor the grid j static information captivation to personnel i, D_i,jFor the grid j multidate information captivation to personnel i, M is for building Build the grid number of object plane segmentation；

Step 4, the movement probability of all personnel in grid is calculated, to personnel at current time location to arriving outlet Whole evacuation process simulation.

A kind of space based on Auditory Perception the most according to claim 1 crowd evacuation analogy method, it is characterised in that institute State the square that grid is 0.5m, more than or equal to 1 people less than or equal to 5 people in each grid.

A kind of space based on Auditory Perception the most according to claim 2 crowd evacuation analogy method, it is characterised in that institute State in step 2 personnel's translational speed and direction in grid particularly as follows:

Designer's translational speed is 1m/s～2m/s, Ren YuanPosition toWithFour positions Mobile.

A kind of space based on Auditory Perception the most according to claim 3 crowd evacuation analogy method, it is characterised in that institute State k in step 3_DParticularly as follows:

K in the case of high-visibility voiceless sound information command_D=0；High-visibility has k in the case of acoustic information instruction_D=0.84；Low K in the case of visibility voiceless sound information command_D=1.43；Low visibility has k in the case of acoustic information instruction_D=1.2.

A kind of space based on Auditory Perception the most according to claim 4 crowd evacuation analogy method, it is characterised in that institute State k in step 3_SParticularly as follows:

When grid j is outlet, k_s=1 represents individual attractive to evacuating；When grid j be obstructed thing occupy time, k_s=-1 Represent that evacuation individuality is had repulsive force；k_s=0 expression does not consider static information captivation.

A kind of space based on Auditory Perception the most according to claim 5 crowd evacuation analogy method, it is characterised in that S in described step 3_i,jExpression formula be:

For resistivity, represent that personnel i is to exporting e_kDistance in wall and the stair obstruction degree to personnel；

α and β is resistivity, r_fIt is grid j to outlet e_kLevel land length, r_sIt is grid j to outlet e_kStair length, r_i,j It is grid j to outlet e_kDistance, use coordinate differential technique by grid j to export e_kDistance determine；

r_i,jIt is expressed as:

If total k outlet in building, in formulaRepresent outlet e_kPosition coordinates；(x_j,y_j) represent grid j position Put coordinate.

A kind of space based on Auditory Perception the most according to claim 6 crowd evacuation analogy method, it is characterised in that institute State D in step 3_i,jExpression formula be:

b_i,jThe reaction that alarm signal is made by expression personnel, its value uses Weibull distribution；Expression personnel i is at net Noise jamming suffered in lattice j, is expressed as:

g_i,jT () is visibility factor, 0 ＜ g_i,jT () ＜ 1, uses Gaussian to be distributed:

For the weight of the alarm song information that personnel i is able to receive that at current time, it is expressed as:

L_wFor acoustic power level, Q is directed to sex factor, r be sound source away from measuring point distance, R is room constant,Wherein s is room Interior total surface area,Indoor average sound absorption coefficient.