Building evacuation simulation method considering influence of path multi-space parameters on evacuation behavior

Abstract

The invention discloses a building evacuation simulation method considering influence of path multi-space parameters on evacuation behaviors, and belongs to the technical field of personnel evacuation simulation. The method comprises the following steps: determining a typical plane; performing typical plane traffic space composition analysis; analyzing typical plane traffic space parameters; building a building model; creating a moving mode and an interaction system of the virtual reality platform; an evacuation path selection experiment is designed; building an evacuation path selection model; development of an evacuation simulation process; and improving the evacuation model flow. According to the method, the influence of evacuation path multi-space parameters on evacuation path selection is integrated into a building evacuation model, and an improved evacuation model is constructed, so that an evacuation simulation result better conforms to a real situation. Evacuation simulation result evaluation based on the method provides a framework for knowledge-intensive building safe evacuation design, and the evacuation time can be shortened by 31.71% at most by optimizing a building traffic space plane based on the method.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of personnel evacuation simulation, and particularly relates to a building evacuation simulation method considering the influence of path multi-space parameters on evacuation behavior. BACKGROUND

[0002] The increasingly complex layout of large public buildings increases the difficulty of finding suitable evacuation paths for people in emergency situations, posing a new threat to building emergency management and safety design. In view of this situation, it is of great practical significance to study individual evacuation behavior and group evacuation rules using virtual reality technology and simulation tools. On the one hand, virtual reality technology can create evacuation scenarios close to real situations, which is conducive to adjusting research variables in a virtual environment. It provides a theoretical basis and mathematical model for modeling individual evacuation behavior in the process of emergency evacuation. On the other hand, evacuation models are important tools for studying evacuation rules and can provide a basis for risk assessment and performance design in safety management. The original evacuation model (OEM) only considers the influence of path distance on evacuation path selection in terms of spatial parameters, ignoring the influence of path width and the left-right position of the path at indoor intersections. This leads to OEM being unable to accurately reflect the individual's evacuation path selection behavior in real situations and unable to restore reality. SUMMARY

[0003] The present application provides a building evacuation simulation method considering the influence of path multi-space parameters on evacuation behavior to solve the problem that current evacuation models ignore the potential guiding effect of building space on evacuation behavior, resulting in unrealistic simulation results. This method integrates the influence of path multi-space parameters (path width and left-right position relative to the intersection) on evacuation path selection into the building evacuation model to construct a refined evacuation model (REM), making the simulation results more realistic.

[0004] To achieve the above-mentioned purpose, the technical solutions adopted by the present application are as follows:

[0005] A building evacuation simulation method considering the influence of path multi-space parameters on evacuation behavior, the method comprising the following steps:

[0006] S1: Determine the building plane for the evacuation path selection experiment;

[0007] S2: Typical plane traffic space composition analysis: The area where the evacuation path selection experiment is carried out is limited to a traffic space, that is, the combination of corridors and open spaces, in order to obtain a mathematical model between traffic space parameters and path selection results, a plurality of traffic space planes need to be established as the scene of the evacuation path selection experiment, and sufficient experimental results need to be ensured to establish a mathematical model between traffic space parameters and path selection results;

[0008] S3: Typical plane traffic space parameter analysis: The combination of corridors and open spaces is analyzed into five space parameters, which are: (1) the length (L) of the corridor and the open space, C (2) the corridor width (W O ), (3) the open space width (W C ), (4) the vertical position (V) of the open space, that is, the position of the open space in the y-axis direction, and (5) the horizontal position (H) of the open space, that is, the position of the open space in the x-axis direction, the five traffic space parameters together constitute any traffic space plane;

[0009] S4: Building model construction: The building model of the scene is established in Sketchup and exported in fbx format; the fbx file is imported into Unity; the material ball is created using the Material component in Unity, and the material ball is used to give the building model a material;

[0010] S5: Mobile mode and interactive system creation of virtual reality platform: add the created building model to the Unity scene, and develop an autonomous roaming system in the scene using the XR Origin component of Unity;

[0011] S6: Evacuation path selection experiment design: the position of the tester is an intersection, and there are two paths (corridors and open spaces) on the left and right of the intersection leading to the evacuation door, the lengths of the two paths are the same; the tester selects one of the corridors or open spaces as the path leading to the evacuation door;

[0012] S7: Evacuation path selection model construction: count the number of people who select corridors and open spaces in the evacuation path selection experiment; the path selection result index is the ratio of the number of people who select corridors and open spaces (C / O), C is the number of people who select corridors, and O is the number of people who select open spaces. According to the path selection result (C / O), take L, W C , W O , V, and H as independent variables, and C / O as dependent variable, and use linear regression method to establish path selection model; stepwise regression method is used in the model establishment process to eliminate insignificant variables in the model;

[0013] S8: Evacuation simulation process development: Establish the building model for evacuation simulation in the social force model simulation software (Anylogic); the PedSource module is used to generate pedestrians in the specified area, the pedestrian generation area is defined by the node component, the number of pedestrians is controlled by the Parameter component provided by anylogic, the Target module represents the target of different evacuation paths, the PedSink module is used to eliminate pedestrians who have reached the target, and the PedSelectOutput module is used to distribute pedestrians on different paths;

[0014] S9: Refined Evacuation Model (REM): In the S8 original evacuation model simulation process, the expressions obtained in S7 are used to further calculate the proportion of pedestrians in the corridor (PPC) and open space (PPO), and the calculated PPC and PPO are input into the PedSelectOutput component to realize pedestrian distribution based on the influence of multi-space parameters on evacuation behavior, and to develop an improved evacuation model.

[0015] Further, in step S1, the aspect ratio of the typical plane is between 1 and 3, the single-layer area is between 2800m 2 ~3500m 2 , and a atrium is provided, and each room or functional area is organized around the atrium, in addition, the evacuation stairs are usually located at the four corners of the plane.

[0016] Further, in step S3, 25 evacuation path selection experimental scenarios are determined by orthogonal experiment, and the designed scenarios have balanced dispersion and uniform comparability.

[0017] Further, in step S5, in the virtual reality experiment, the tester's moving speed is 2.5m / s, and the rotation picture frame rate is set to 60 frames.

[0018] Further, in step S6, before the formal experiment starts, all participants have a rest time to ensure that their physiological conditions remain stable and avoid various factors before the experiment affecting the experimental results.

[0019] Further, in step S9, the calculation method of PPC and PPO is:

[0020]

[0021]

[0022]

[0023]

[0024] The initial speed of the pedestrian is set to 1.2 m / s-1.7 m / s, and the expected speed is set between 2 m / s to 3 m / s; in addition, the cross section of the pedestrian is set to a circle to simplify the model, and the diameter is set between 0.4-0.5 m; the simulation ends when all pedestrians reach the target.

[0025] Further, the function component is used to calculate the non-evacuated pedestrians in real time. Finally, the TimePlot component and the Densitymap component are used to output the evacuation time and the pedestrian density distribution results during the evacuation process.

[0026] The beneficial effects of the present application relative to the prior art are:

[0027] 1. The existing evacuation model rarely takes into account the influence of spatial parameters other than distance on evacuation behavior. The present method integrates spatial factors such as width and position into the evacuation model in addition to distance. The method reveals the different effects of traffic behavior and spatial preferences on evacuation route selection. The modification rate of the existing evacuation model is increased by up to 46.43%.

[0028] 2. The evacuation simulation result evaluation based on the present method provides a framework for knowledge-intensive building safety evacuation design, and provides architects with evidence-based library space parameter optimization scheme reference and strategy basis. Based on the optimization of the building traffic space plane according to the present method, the evacuation time can be shortened by up to 31.71%. BRIEF DESCRIPTION OF DRAWINGS

[0029] Figure 1 The present application is a technical flowchart;

[0030] Figure 2 It is a typical plane and spatial parameter description diagram of a university library for a virtual reality route selection experiment;

[0031] Figure 3 It is a traffic space plan of 25 university libraries;

[0032] Figure 4 It is a virtual reality experiment site photo and equipment diagram;

[0033] Figure 5 It is a virtual reality experiment scene diagram;

[0034] Figure 6 It is a flowchart of the original evacuation model simulation;

[0035] Figure 7 It is a flowchart of the improved evacuation model simulation;

[0036] Figure 8 It is an evacuation time statistical diagram;

[0037] Figure 9Density distribution graphs of the simulation of the influence of multiple space parameters on the distribution of the density peak of the personnel during evacuation for Case 1 to Case 5;

[0038] Figure 10 Density distribution graphs of the simulation of the influence of multiple space parameters on the distribution of the density peak of the personnel during evacuation for Case 6 to Case 10. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solutions and advantages of the present application clearer and more obvious, the implementation and application of the present evacuation simulation method will be further described in detail below in combination with embodiments and drawings. The suitable implementation manner of the present application and the description thereof are only used to explain the present application and do not limit the present application.

[0040] The present embodiment discloses a building evacuation simulation method considering the influence of multiple space parameters on evacuation behavior. In order to study the influence of multiple space parameters on evacuation path selection, a virtual reality technology is used to carry out an evacuation path selection experiment. As shown in Figure 1 the method comprises the following steps:

[0041] S1: Typical plane determination: determine the building plane of the virtual reality experiment of evacuation path selection. In order to make the application of the present method universal, a typical plane needs to be selected as the building plane of the virtual reality experiment of evacuation path selection. A university library is a typical representative of large public buildings. The layout of the university library is investigated. The reason for selecting the university library as an example is that it is functionally composite (learning, research, cultural transmission) and has a large space scale, which embodies the public and service nature and is a typical representative of large public buildings. In addition, a large number of young students are gathered in the university library, which has the necessity to carry out the research on evacuation path selection. After investigation, the university library has multiple planes, and the length-width ratio of the plane is between 1 and 3. The single-layer area is between 2800m 2 and 3500m 2 . In order to meet the needs of ventilation, lighting, etc., a central court is usually provided. Each room or reading area is organized around the central court. In addition, the evacuation staircase is usually located at the four corners of the plane. Based on these research data, the typical plane established by the present method is shown in Figure 2 a.

[0042] S2: Analysis of the composition of the typical plane traffic space: The bearing area of the evacuation behavior is the traffic space. In this method, the area where the evacuation path selection experiment is carried out is limited to the traffic space. Through the investigation of the university library, the typical plane traffic space is composed of corridors and open spaces. Open spaces are wider than corridors, and have other functions besides transportation, such as leisure and social functions, which are common types of traffic space in large public buildings. The traffic space referred to in this invention is the combination of corridors and open spaces, which is a common combination form of traffic space in large public buildings taking the university library as an example. In order to obtain the mathematical model between the traffic space parameters and the path selection results, multiple traffic space planes need to be established as the scene of the evacuation path selection experiment, ensuring that there are enough experimental results to establish the mathematical model between the traffic space parameters and the path selection results.

[0043] S3: Analysis of the parameters of the typical plane traffic space: The combination of corridors and open spaces is analyzed into five space parameters, which are: (1) the length of the corridor and the open space (L), (2) the width of the corridor (W C ), (3) the width of the open space (W O ), (4) the vertical position of the open space (V), i.e. the position of the open space in the y-axis direction, (5) the horizontal position of the open space (H), i.e. the position of the open space in the x-axis direction, Figure 2 b shows a schematic diagram of parameters V and H. These five traffic space parameters together constitute any one traffic space plane. This method uses orthogonal experiment to determine the number of evacuation path selection experiment scenes. Orthogonal experiment method can efficiently analyze the influence of traffic space parameters on path results with 25 evacuation path selection experiment times through scientific arrangement of multi-factor and multi-level experiments, which significantly saves time and cost. Its balanced dispersion and uniform comparability ensure the reliability of the evacuation path selection experiment data. The orthogonal design scheme of the experiment scene is generated in IBM SPSS 27.0.1, and 25 traffic space parameter combinations of the experiment scene are generated. The traffic space layout from scene 1 (S1) to scene 25 (S25) is shown in Figure 3 . The traffic space parameters from S1 to S25 are shown in Table 1.

[0044] Table 1 Traffic space parameters of 25 scenes

[0045]

[0046] S4: Architectural Model Construction: Build architectural models for 25 scenes in Sketchup and export them in .fbx format. Import these .fbx files into Unity. Use the Material component in Unity to create shaders, which are then used to assign materials to the architectural models. The goal of assigning materials is to make the VR environment more similar to the real world. There's no need to specify specific materials, as long as they resemble the real world.

[0047] S5: Creation of a Movement Mode and Interaction System for the VR Platform: Add the created building model to a Unity scene and use Unity's XR Origin component to develop an autonomous roaming system within the scene. This allows participants to achieve evacuation behaviors that are more realistic within the virtual environment. In a virtual environment, higher movement speeds can cause dizziness, while lower movement speeds tend to reduce engagement. Setting the movement speed of participants in VR experiments to 2.5 m / s is a common and appropriate value. To ensure that the human eye's continuous perception of dynamic images is close to that of reality and enhances realism, the frame rate of the rotating images in VR experiments is set to 60 frames.

[0048] S6: Experimental design of evacuation route selection: experimental photos Figure 4 The virtual scene in the experiment is Figure 5 Generally speaking, the evacuation route selection of large public buildings usually occurs at intersections. According to the survey, intersections consisting of corridors and open spaces are more common in large public buildings. This method is used to study this common situation. Participants were told that they were in an emergency evacuation state and their location was an intersection. There were two paths (corridor and open space) on the left and right of the intersection leading to the evacuation door ( Figure 5 d and Figure 5 f) The two paths were of equal length; their task was to select a path in the corridor or open space that led to the evacuation door. The evacuation path selection experiment included 25 scenarios. The traffic space parameters for all scenarios are shown in Table 1. Before the actual experiment began, all participants had a rest period to ensure their physiological condition remained stable and to prevent various pre-experimental factors from influencing the experimental results.

[0049] S7: Evacuation route selection model construction: Count the number of people who choose corridors and open spaces in the evacuation route selection experiment. The route selection result indicator is the ratio of the number of people who choose corridors to the number of people who choose open spaces (C / O). C is the number of people who choose corridors, and O is the number of people who choose open spaces. Based on the route selection result (C / O), L, W C , W O, V, H are independent variables, C / O is dependent variable, the route choice model is established by using linear regression method. In the process of model establishment, stepwise regression method is used to eliminate insignificant variables in the model; the final model results are shown in Table 2.

[0050] Table 2 Route choice model

[0051]

[0052] The expression is obtained as follows:

[0053]

[0054] Wherein, C / O is the ratio of the number of people choosing the corridor and the open space; W O is the width of the open space, and H is the horizontal position of the open space;

[0055] S8: Development of evacuation simulation process: the building model for evacuation simulation is established in the social force model simulation software (Anylogic); the PedSource module is used to generate pedestrians in the specified area, the pedestrian generation area is defined by the node component, the number of pedestrians is controlled by the Parameter component provided by anylogic, the Target module represents the target of different evacuation paths, the PedSink module is used to eliminate pedestrians reaching the target, and the PedSelectOutput module is used to distribute pedestrians of different paths;

[0056] The overall process of the original evacuation model is shown in Figure 6 a. Figure 6 The positions of the above modules in the simulation plane are shown in Fig. b. Pedestrians are generated in the blue dashed line area (node component control), and the pedestrians move towards the targets of the corridor and the open space. TargetC is the target of the corridor, and TargetO is the target of the open space. PedsinkC is used to eliminate pedestrians reaching the corridor target, and PedsinkO is used to eliminate pedestrians reaching the open space target.

[0057] S9: Development of refined evacuation model (REM) Figure 7 ): In the simulation process of the original evacuation model in S8, the expression obtained in S7 is used to further calculate the proportion of pedestrians in the corridor (PPC) and the open space (PPO). The calculated PPC and PPO are input into the PedSelectOutput component, so as to realize the pedestrian distribution based on the influence of multi-space parameters on evacuation behavior, and to develop the refined evacuation model. The calculation methods of PPC and PPO are as follows:

[0058]

[0059]

[0060]

[0061]

[0062] The initial speed of the pedestrian is set to 1.2 m / s - 1.7 m / s, and the expected speed is set between 2 m / s to 3 m / s. In addition, the cross section of the pedestrian is set to a circle to simplify the model, and the diameter is set between 0.4 - 0.5 m. The simulation ends when all pedestrians reach the target.

[0063] The simulation ends once all pedestrians reach the target location. Function components are used to calculate the unswept pedestrians in real time. Finally, the TimePlot component and Densitymap component are used to output the evacuation time and the pedestrian density distribution results during the evacuation process.

[0064] Example 1:

[0065] S1: Take the university library as an example, and build a building model for evacuation simulation in the social force model simulation software (Anylogic). According to the path selection model, Wo and H are significant influencing factors, and L, W C and V are non-significant influencing factors. The traffic space parameters Wo and H used for simulation are variables. L, W C and V are fixed values. These fixed values are taken from the commonly used data of building design. L is set to 16 m, W C is set to 3 m, and V is set to 3. There are a total of ten case layouts, and the traffic space parameters of the 10 cases are shown in Table 3.

[0066] Table 3 Traffic space parameters of 10 cases

[0067]

[0068] S2: Develop the improved evacuation model (REM) simulation process (see Figure 7 ). The PedSource module is used to generate pedestrians in a specified area, and the number of pedestrians is controlled by the Parameter component provided by Anylogic. Based on the typical layout area of the library and the per capita use area, the maximum number of evacuees in a single fire compartment is calculated to be 150. The number of evacuees is set to 30, 60, 90, 120 and 150. The simulation ends once all pedestrians reach the target. In order to ensure the accuracy of the simulation results, each case is simulated 10 times, and the average value is taken as the simulation result. Finally, the TimePlot component and Densitymap component are used to output the evacuation time, see Figure 8 . In addition, the pedestrian density distribution results during the evacuation process are also counted, see Figure 9 andFigure 10 .

[0069] S3: Table 4 shows the evacuation time of all cases simulated by the improved evacuation model. The results show that the width and location of the open space will cause differences in evacuation time when the distance is constant. Based on the optimization of building traffic space plane by the simulation method, the evacuation time can be shortened by 31.71% at most (case 6 with 150 people compared with case 1).

[0070] Table 4 Evacuation time (seconds) of all cases

[0071]

[0072] In addition, the correction ratio of the improved evacuation model (REM) to the evacuation time simulated by the original evacuation model (OEM) is analyzed, as shown in Table 5. The improved evacuation model reveals the different effects of traffic behavior and space preference on evacuation route selection. The accuracy of the existing evacuation model is improved by 46.43%.

[0073] Table 5 Correction percentage (%) of REM to OEM evacuation time

[0074]

[0075] The technical features of the above-described embodiments can be combined in any manner. In order to make the description concise, not all possible combinations of the technical features in the above-described embodiments are described, however, as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present disclosure.

Claims

1. A building evacuation simulation method considering the impact of multiple spatial parameters of a path on evacuation behavior, characterized by: The method comprises the following steps: S1: Typical plane determination: Determine the building plane for the evacuation route selection experiment; S2: Analysis of the composition of typical plane traffic spaces: The area where the evacuation route selection experiment was conducted was limited to traffic spaces, which are a combination of corridors and open spaces. In order to obtain a mathematical model between traffic space parameters and route selection results, it is necessary to establish multiple traffic space planes as scenarios for the evacuation route selection experiment to ensure that there are sufficient experimental results to establish a mathematical model between traffic space parameters and route selection results. S3: Analysis of typical plane traffic space parameters: The combination of corridors and open spaces is analyzed into five spatial parameters, namely: (1) the length of the corridor and open space (L), (2) the width of the corridor (W C ), (3) Open space width (W O ), (4) the vertical position of the open space (V), that is, the position of the open space in the y-axis direction, (5) the horizontal position of the open space (H), that is, the position of the open space in the x-axis direction. These five traffic space parameters together constitute any traffic space plane; S4: Building Model: Created the architectural model of the scene in Sketchup and exported it in .fbx format; imported the .fbx file into Unity; used the Material component in Unity to create a material ball, and used the material ball to assign materials to the architectural model; S5: Creation of movement mode and interaction system for virtual reality platform: Add the created building model to the Unity scene and use Unity's XR Origin component to develop the autonomous roaming system in the scene; S6: Experimental design for evacuation route selection: The subject is located at an intersection with two paths (corridor and open space) leading to the evacuation door on either side of the intersection. The two paths are of the same length. The subject chooses one path from the corridor or the open space as the path leading to the evacuation door. S7: Evacuation route selection model construction: Count the number of people who choose corridors and open spaces in the evacuation route selection experiment; the route selection result indicator is the ratio of the number of people who choose corridors to the number of people who choose open spaces (C / O), where C is the number of people who choose corridors and O is the number of people who choose open spaces; according to the route selection result (C / O), L, W C , W O , V, H are independent variables, C / O is the dependent variable, and the path selection model is established using the linear regression method; in the process of model establishment, the stepwise regression method is used to eliminate the insignificant variables in the model; S8: Evacuation simulation process development: Establish a building model for evacuation simulation in the social force model simulation software; the PedSource module is used to generate pedestrians in a specified area. The pedestrian generation area is defined by the node component, and the number of pedestrians is controlled by the Parameter component provided by anylogic. The Target module represents the target of different evacuation paths. The PedSink module is used to eliminate pedestrians who reach the target. The PedSelectOutput module is used to allocate pedestrians to different paths. S9: Improved evacuation model: In the original evacuation model simulation process of S8, the expression obtained in S7 is used to further calculate the pedestrian proportions in corridors (PPC) and open spaces (PPO). The calculated PPC and PPO are input into the PedSelectOutput component respectively to realize pedestrian allocation based on the influence of multiple spatial parameters on evacuation behavior, so as to develop an improved evacuation model.

2. A building evacuation simulation method considering the influence of multiple spatial parameters of a path on evacuation behavior according to claim 1, characterized in that: In step S1, the typical plane has a plane aspect ratio between 1 and 3, and a single-layer area of ​​2800 m 2 ~3500m 2 There is an atrium between the buildings, and the rooms or functional areas are organized around the atrium. In addition, the evacuation stairs are usually located at the four corners of the plan.

3. The building evacuation simulation method considering the influence of multiple spatial parameters of a path on evacuation behavior according to claim 1 is characterized by: In step S3, an orthogonal experiment is used to determine 25 experimental scenarios for evacuation route selection, and the designed scenarios are balanced, dispersed, and neatly comparable.

4. The building evacuation simulation method considering the influence of multiple spatial parameters of a path on evacuation behavior according to claim 1 is characterized by: In step S5, in the virtual reality experiment, the tester's moving speed is 2.5 m / s, and the frame number of the rotating screen is set to 60 frames.

5. The building evacuation simulation method considering the influence of multiple spatial parameters of a path on evacuation behavior according to claim 1 is characterized by: In step S6, before the formal experiment begins, all participants have a rest period to ensure that their physiological conditions remain stable and to prevent various factors before the experiment from affecting the experimental results.

6. The building evacuation simulation method considering the influence of multiple spatial parameters of a path on evacuation behavior according to claim 1 is characterized by: In step S9, PPC and PPO are calculated as follows: The initial speed of the pedestrians is set to 1.2m / s-1.7m / s, and the expected speed is set between 2m / s and 3m / s. In addition, the pedestrian cross-section is set to a circle to simplify the model, and the diameter is set between 0.4-0.5m. The simulation ends when all pedestrians reach the target.

7. The building evacuation simulation method according to claim 6, wherein: The function component is used to calculate the number of pedestrians who have not been evacuated in real time; finally, the TimePlot component and Densitymap component are used to output the evacuation time and pedestrian density distribution results during the evacuation process.

Images