Efficient indoor personnel evacuation simulation and optimization method based on social force model

Abstract

The invention discloses an efficient indoor personnel evacuation simulation and optimization method based on a social force model, and belongs to the technical field of evacuation modeling and simulation. According to the method, a social force model including internal driving force, psychological repulsive force, extrusion force and attraction force is constructed, and the Newton's second law is combined to dynamically simulate the movement track and behavior change of people in an evacuation scene. The simulation platform adopts MATLAB and other tools, and supports multi-scene modeling and multi-target optimization, including shortest evacuation time, minimum congestion risk and the like. Based on the model, dynamic simulation and layout optimization of a complex geometric site can be realized, such as newly-added exit position design, aisle width adjustment and the like, so that the evacuation efficiency is remarkably improved. The method is suitable for high-density places such as schools, office buildings, subway stations and gymnasiums, an optimization scheme can be provided in disaster scenes such as fire disasters and earthquakes, and theoretical support and technical tools are provided for smart city construction and public safety design.

Description

Technical Field

[0001] The present invention belongs to the technical field of evacuation modeling and simulation, and particularly relates to an efficient indoor personnel evacuation simulation and optimization method based on the social force model. Background Art

[0002] In recent years, with the rapid advancement of urbanization and the continuous increase in population density, the design and management of large indoor public places have faced increasingly severe challenges. In the event of an emergency or an emergency evacuation scenario, the high-density distribution of people easily leads to problems such as crowding, trampling, and evacuation delays, thus posing a serious threat to the lives and safety of people. In this case, how to effectively design the site layout, optimize the evacuation route, and reduce risks has become an important research topic in the field of public safety.

[0003] Existing evacuation models are mostly based on rules or simple geometric models, such as the cellular automaton model (CA model). These methods have certain advantages in describing the evacuation trend of people in a large area, but they have obvious deficiencies in dealing with dynamic behavior changes and interactions between individuals in complex scenarios. These models often cannot accurately depict the subtle interactions between individuals in a high-density environment and their impact on the overall evacuation efficiency, nor can they fully consider the randomness and diversity of individual behaviors.

[0004] The social force model has received extensive attention in recent years as a tool based on physical dynamics. However, there are still certain limitations in existing research based on the social force model. For example, in actual scenarios, there is currently a lack of systematic tools and methods for adjusting the site layout, determining the exit location, and optimizing the passage design based on the simulation results to shorten the evacuation time and improve safety. In addition, for dynamic changes in high-density places (such as emotional fluctuations and irrational behaviors caused by emergencies), the social force model still needs to be improved in terms of parameter optimization and environmental modeling. Summary of the Invention

[0005] Object of the Invention: The object of the present invention is to provide an efficient indoor personnel evacuation simulation and optimization method based on the social force model. By improving individual dynamics modeling and dynamic simulation technology, it realizes the analysis of crowd behavior and the optimization of evacuation efficiency under different indoor layout conditions.

[0006] Technical Solution: An efficient indoor personnel evacuation simulation and optimization method based on the social force model of the present invention includes the following steps:

[0007] Step 1, model the site, input the site size, the number and location of exits, and the obstacle distribution information, and generate a two-dimensional or three-dimensional site layout diagram;

[0008] Step 2: Construct a social force model including driving force, attraction force, psychological repulsion force, and extrusion force;

[0009] Step 3: Based on the site layout map and the social force model, construct an individual dynamics model, solve the individual motion state through the dynamics equation, and dynamically simulate the evacuation process;

[0010] Step 4: Adjust the site layout based on the simulation results of the dynamic simulation evacuation process to optimize the evacuation time.

[0011] Furthermore, in Step 2, the driving force is an important factor for an individual to generate acceleration. The acceleration vector of the individual points to the target point currently selected by it. The whole process occurs according to the individual's needs and behavioral decisions. The individual expects to reach the target point in the shortest distance, at the lowest cost, or in the most comfortable way, and drives itself to accelerate forward through the driving force. The expression of the driving force is:

[0012]

[0013] where, e i (t) is the expected direction of individual i at time t, and the expression is:

[0014]

[0015] where, p i (t) represents the target position of individual i at time t, r i (t) represents the actual position of individual i at time t, ||p i (t)-r i (t)|| represents the Euclidean distance between the target position and the current position of individual i at time t;

[0016] Furthermore, in Step 2, the attraction force is that when the target individual follows the retrograde individual to move towards the exit, the retrograde individual generates an attraction effect on the target individual; the attraction effect is the attraction force of the retrograde individual on the target individual, which prompts the target individual to move towards the retrograde individual; the retrograde individual plays a guiding role to help the target individual move in the correct direction to achieve the goal of smooth evacuation. Its expression is:

[0017]

[0018] where, C j is a constant, representing the attraction force of target individual j on other individual i, r im represents the maximum effective distance of the attraction force of individual i, d ij (t) represents the distance between individual i and target individual j at time t, r ij (t) represents the distance direction vector between individual i and individual j at time t.

[0019] Furthermore, in step 2, the psychological repulsive force is an internal driving force that an individual experiences during the evacuation process, causing them to maintain a distance. As the distance between individuals decreases, the internal driving force increases, resulting in a stronger psychological sense of repulsion between individuals, which is called the inter-individual psychological repulsive force. The existence of this force causes individuals to adjust their actions, and the expression is:

[0020]

[0021] where A i is a constant representing the intensity of the psychological repulsive force generated by individual i on individual j, d ij (t) represents the distance between individual i and the target individual j at time t, r i represents the psychological radius of individual i, r j represents the psychological radius of individual j, B i is a constant representing the intensity of the psychological repulsive force generated by individual i on individual j, r ij (t) represents the distance direction vector between individual i and individual j at time t;

[0022] where the expression for d ij (t) is:

[0023] d ij (t) = ||r i (t) - r j (t)||

[0024] r ij (t) is expressed as:

[0025]

[0026] When there are obstacles, individuals maintain a distance to avoid collisions with the walls or obstacles. When an individual comes into contact with a wall or an obstacle, a force is generated, and the expression for the force is:

[0027]

[0028] where A w is a constant representing the intensity of the collision avoidance force generated by the obstacle w on individual i, r i represents the psychological radius of individual i, d iw (t) represents the distance between individual i and the obstacle w, B w is a constant representing the collision avoidance force attenuation coefficient of the obstacle w, controlling the speed at which the collision avoidance force changes with distance, n iw (t) is a unit vector representing the direction from individual i to the obstacle w.

[0029] The total psychological repulsive force is:

[0030]

[0031] Furthermore, in Step 2, the extrusion force is the total additional force of an individual in the extrusion state; when individual i is in a crowded state, the calculation of the total additional force considers the interaction force between individual j and other surrounding individuals i. Each individual exerts a force on individual i, and this force is decomposed into two parts: body contact force and sliding friction. The formula is expressed as:

[0032]

[0033] where k is a constant representing the force proportionality coefficient, related to the physical characteristics of the individual, g represents a function indicating the variation of the force with distance or relative velocity, r i represents the psychological radius of individual i, r j represents the psychological radius of individual j, d ij represents the actual distance between individual i and the target individual j, n ij is a unit vector indicating the direction from individual i to individual j, Δv ij represents the velocity change between individual i and individual j, t ij represents the contact time or interaction time between individual i and individual j.

[0034] g(x) is a piecewise function that takes effect when x exceeds 0 and has a value of 0 when it is below 0

[0035]

[0036] Furthermore, Step 3 is specifically as follows: when there are obstacles in the scene, the force between an individual and an obstacle is regarded as a comprehensive experience of the individual's psychological intention and physical interaction with the obstacle, and the physical force consists of the extrusion force generated by contact with the wall and the sliding friction force generated by relative movement after contact with the wall; after a pedestrian contacts a wall or an obstacle, the force formula between the individual and the wall or obstacle is expressed as:

[0037]

[0038] where k is a constant representing the force proportionality coefficient, related to the physical characteristics of the individual, g(x) represents a function that may indicate the variation of the force with distance or relative velocity, r i represents the psychological radius of individual i, d iw represents the distance between individual i and the obstacle w, n iw (t) is a unit vector indicating the direction from individual i to the obstacle w. Δn iw (t) represents the relative velocity change between individual i and the obstacle w.

[0039] Then the total additional force is:

[0040]

[0041] According to Newton's second law:

[0042]

[0043] The present invention also discloses an efficient indoor personnel evacuation simulation and optimization system based on the social force model, including a site modeling module, an individual behavior modeling module, a dynamic simulation module, and a layout optimization module;

[0044] The site modeling module supports users to customize and input site dimensions, the number and positions of exits, and obstacle distribution information; and automatically generates a two-dimensional or three-dimensional site layout diagram;

[0045] The individual behavior modeling module constructs an individual dynamics model based on the social force model, and supports parametric definitions including individual body size, desired speed, and psychological characteristics;

[0046] The dynamic simulation module uses the time step iteration method to simulate the individual movement trajectories, outputs time series results and overall evacuation dynamics; and supports multi-threaded parallel computing;

[0047] The layout optimization module adopts the traversal method or the genetic algorithm to search for the optimal layout plan, and the optimization objectives are the shortest evacuation time or the least congested area.

[0048] Advantageous effects: Compared with the prior art, the present invention has the following remarkable advantages:

[0049] Based on the social force model and combined with dynamics modeling and dynamic simulation technologies, the present invention proposes an efficient indoor personnel evacuation simulation and optimization method. By simulating the behavior dynamics of the crowd in an emergency, this method can accurately evaluate the evacuation time, identify congested areas, and provide an optimized layout design plan. Multiple implementation cases verify its applicability and efficiency in complex geometric scenarios. For example, measures such as optimizing the position of new exits and adjusting the width of passages significantly improve the evacuation efficiency and minimize the congestion risk and the possibility of people being trapped.

[0050] The present invention can not only meet the safety evacuation requirements of places such as schools, office buildings, subway stations, and stadiums, but also be applied to the dynamic planning of disaster emergency scenarios, providing a theoretical basis and practical tool for the formulation of emergency plans and the safety design of public places. In addition, this method can be deeply integrated with artificial intelligence, big data analysis, and Internet of Things technologies, providing a broader application prospect for crowd flow monitoring and real-time layout optimization in the construction of smart cities, and having important social significance and economic value. Description of the Drawings

[0051] Figure 1 This is a schematic diagram of collision detection for the present invention.

[0052] Figure 2 This is a schematic diagram of the MATLAB simulation process, where the unit of the abscissa and ordinate is meter. Specific implementation manners

[0053] The technical solution of the present invention will be further described below in conjunction with the accompanying drawings.

[0054] Embodiment: School emergency evacuation simulation and optimization

[0055] As a special public place, schools carry the dreams of countless students. In the case of such a large population base, how to take correct emergency measures in a timely manner in case of emergencies is a topic of great concern in this field. In addition to strengthening emergency drills for emergencies on a daily basis, the evacuation time can also be reduced by reasonably arranging the room layout, thereby obtaining a greater chance of survival. Based on the social force model, combined with dynamics modeling and dynamic simulation technology, the present invention proposes an efficient indoor personnel evacuation simulation and optimization method. By simulating the dynamic behavior of the crowd in case of emergencies, this method can accurately evaluate the evacuation time, identify congested areas, and provide an optimized layout design scheme. The specific solution process is as follows:

[0056] Step 1: Site construction

[0057] Personnel distribution: 100 students are randomly distributed in a 20m long and 20m wide indoor site. Assume that the line of sight is good at this time, and there is a door with a width of 2m located at the midpoint of one side.

[0058] Environmental conditions: Good line of sight, no obstacles, no sudden stampede phenomenon.

[0059] Mechanical assumptions: The mutual forces between individuals include driving force, psychological repulsive force, squeezing force, etc.

[0060] Personnel characteristics: The mass of each individual is 80kg, the shoulder width is 0.3m, and the maximum speed is 2m / s.

[0061] Step 2: Construction of social force model

[0062] Driving force: In the social force model, the driving force is an important factor for an individual to generate acceleration. The generation of this force stems from the need of the individual to reach the destination comfortably. The acceleration vector of the individual points to the currently selected target point. The whole process occurs according to the needs and behavior decisions of the individual. The individual hopes to reach the target point in the shortest distance, at the lowest cost or in the most comfortable way, and drives himself to accelerate forward through the driving force. The expression is

[0063]

[0064] where e i (t) is the expected direction of individual i at time t, and the expression is:

[0065]

[0066] Attraction force: When the target individual follows the retrograde individual towards the exit, the retrograde individual generates an attraction effect on the target individual. This attraction effect can be understood as the attraction force of the retrograde individual on the target individual, which prompts the target individual to move towards the retrograde individual. In this case, the retrograde individual plays a guiding role, helping the target individual move in the correct direction to achieve the goal of smooth evacuation. The expression is:

[0067]

[0068] Psychological repulsion force: Individuals will feel an internal driving force during the evacuation process, which prompts them to maintain a certain distance to avoid collisions. As the distance between individuals decreases, this internal driving force will increase, resulting in a stronger psychological repulsion between individuals. We call it the psychological repulsion force between individuals. The existence of this force enables individuals to effectively adjust their actions to maintain an appropriate distance and ensure the safe progress of the evacuation process. The expression is:

[0069]

[0070] where d ij (t) is expressed as:

[0071] d ij (t) = ||r i (t) - r j (t)||

[0072] r ij (t) is expressed as:

[0073]

[0074] When there are obstacles, individuals hope to maintain a certain distance to avoid collisions with the walls or obstacles. When an individual comes into contact with a wall or an obstacle, a force will be generated. The expression of the force is:

[0075]

[0076] The total psychological repulsion force is:

[0077]

[0078] Total additional force on an individual in a squeezing state: When individual \(i\) is in a crowded state, the calculation of the total additional force can consider the interaction forces between individual \(j\) and other surrounding individuals \(i\). Specifically, each of these individuals exerts a force on individual \(i\), and this force can be decomposed into a body contact force and a sliding friction force into two parts. The formula is expressed as:

[0079]

[0080] \(g(x)\) is a piecewise function that takes effect when \(x\) exceeds 0, and its value is 0 when it is less than 0.

[0081]

[0082] When there are obstacles in the scene, the force between an individual and an obstacle can be regarded as a comprehensive experience of the individual's psychological intention and physical interaction with the obstacle, and the physical force is composed of the squeezing force generated by contact with the wall and the sliding friction force generated by relative movement after contact with the wall. After a pedestrian contacts a wall or an obstacle, the force formula between the individual and the wall or obstacle is expressed as

[0083]

[0084] Then the total additional force is:

[0085]

[0086] According to Newton's second law:

[0087]

[0088] Step 3: Simulation and optimization algorithm

[0089] Use MATLAB to simulate the interaction forces between each individual to observe the behavior and dynamic changes of the entire group.

[0090] Perform a variable step-size simulation process for time, taking the time interval as 0.01 s to achieve the effect of simulating the movement process, and finally simulate the entire escape process. Through comprehensive analysis and referring to a large number of references, it is concluded that excluding the differences in irrelevant factors, the evacuation speed has a great relationship with the classroom layout, mainly reflected in several elements such as the number, size, and orientation of the aisles. We use MATLAB combined with the hill-climbing method to traverse the door positions in the left half area. The following figure shows the evacuation times of some traversal points. To consider the randomness of the initial positions of individuals, we set up a control group with an initial position uniformly distributed in a rectangle and a test group with random positions, and conduct simulation tests on them to obtain the differences in evacuation times between the two groups.

[0091] The waveform diagram and the final comparison results show that when the center coordinate of the second door is selected as the bottom (directly below the first door), the evacuation time is the shortest at 10:35 s, ensuring the reliability of the results.

[0092] Based on the social force model, combined with dynamics modeling and dynamic simulation technology, the present invention proposes an efficient indoor personnel evacuation simulation and optimization method. By simulating the dynamic behavior of the crowd in an emergency, this method can accurately evaluate the evacuation time, identify congested areas, and provide an optimized layout design plan. Multiple implementation cases verify its applicability and efficiency in complex geometric scenarios. For example, measures such as optimizing the location of new exits and adjusting the width of passages significantly improve the evacuation efficiency, minimizing the risk of congestion and the possibility of people being trapped.

[0093] The present invention can not only meet the safety evacuation requirements of places such as schools, office buildings, subway stations, and stadiums, but also be applied to the dynamic planning of disaster emergency scenarios, providing a theoretical basis and practical tool for the formulation of emergency plans and the safety design of public places. In addition, this method can be deeply integrated with artificial intelligence, big data analysis, and Internet of Things technologies, providing a broader application prospect for crowd flow monitoring and real-time layout optimization in the construction of smart cities, and having important social significance and economic value.

[0094] In this specification, the use of terms such as "embodiment", "specific embodiment", or "certain embodiments" is intended to indicate that the specific features, materials, structures, or characteristics described in connection with these embodiments or examples are applied in at least one embodiment of the present invention. However, the exemplary expressions of the terms do not necessarily refer to the same embodiment. In addition, the specific features, materials, structures, or characteristics described herein can be combined in any suitable manner in one or more embodiments.

Claims

1. An efficient indoor personnel evacuation simulation and optimization method based on social force model, characterized in that: The steps include: Step 1: Model the site, input site dimensions, exit quantity and location, obstacle distribution information, and generate a two-dimensional or three-dimensional site layout map; Step 2: Construct a social force model including internal drive, attraction, psychological repulsion and squeeze; Step 3: Based on the site layout diagram and the social force model, an individual dynamics model is constructed, and the individual motion state is solved by the dynamics equation to dynamically simulate the evacuation process; Step 4: Adjust the site layout based on the simulation results of the dynamic simulation evacuation process to optimize the evacuation time.

2. According to claim 1, a method for simulating and optimizing efficient indoor evacuation of personnel based on a social force model is characterized in that: In step 2, the intrinsic drive is an important factor in the acceleration of the individual. The acceleration vector of the individual points to the target point currently selected by the individual. The whole process occurs according to the individual's needs and behavioral decisions. The individual expects to reach the target point in the shortest distance, lowest cost or most comfortable way, and drives himself to accelerate forward through the intrinsic drive. The expression of the intrinsic drive is: Among them, e i (t) is the expected direction of individual i at time t, expressed as: Among them, p i (t) represents the target position of individual i at time t, r i (t) represents the actual position of individual i at time t, ||p i (t)-r i (t)|| represents the Euclidean distance between the target position and the current position of individual i at time t.

3. According to the social force model-based efficient indoor personnel evacuation simulation and optimization method of claim 1, it is characterized in that: In step 2, the attraction is that when the target individual follows the retrograde individual to move toward the exit, the retrograde individual produces an attraction effect on the target individual; The attraction effect is the attraction of the retrograde individual to the target individual, which prompts the target individual to move toward the retrograde individual; the retrograde individual plays a guiding role, helping the target individual to move in the right direction to achieve the goal of smooth evacuation, and its expression is: Among them, C j is a constant representing the attraction of target individual j to other individuals i, r im represents the maximum effective distance of attraction of individual i, d ij (t) represents the distance between individual i and target individual j at time t, r ij (t) represents the distance direction vector between individual i and individual j at time t.

4. According to the method of claim 1, the method is characterized in that: In step 2, the psychological repulsion force is an internal driving force that individuals receive during the evacuation process, which keeps them apart. As the distance between individuals decreases, the internal driving force increases, causing a stronger sense of psychological repulsion between individuals, which is called inter-individual psychological repulsion. The existence of this force causes individuals to adjust their actions, and the expression is: Among them, A i is a constant, indicating the intensity of the psychological repulsion that individual i exerts on individual j, d ij (t) represents the distance between individual i and target individual j at time t, r i represents the psychological radius of individual i, r j represents the psychological radius of individual j, B i is a constant, indicating the strength of the psychological repulsion that individual i exerts on individual j, r ij (t) represents the distance direction vector between individual i and individual j at time t; where d ij The expression of (t) is: d ij (t)=||r i (t)-r j (t)|| r ij The expression of (t) is: When there are obstacles, the individual keeps a distance to avoid collision with the wall or obstacle. When the individual contacts the wall or obstacle, a force will be generated. The expression of the force is: Among them, A w is a constant, indicating the strength of the collision avoidance force exerted by obstacle w on individual i, r i represents the psychological radius of individual i, d iw (t) represents the distance between individual i and obstacle w, B w is a constant, representing the attenuation coefficient of the collision avoidance force of obstacle w, controlling the speed at which the collision avoidance force changes with distance, n iw (t) is a unit vector, indicating the direction from individual i to obstacle w; The total psychological repulsion is:

5. The method for simulating and optimizing efficient indoor evacuation based on social force model according to claim 1 is characterized in that: In step 2, the squeezing force is the total additional force of the individual in the squeezing state; When individual i is in a crowded state, the total additional force calculation takes into account the interaction force between individual j and other individuals i around it. Each individual exerts a force on individual i. This force is decomposed into two parts: body contact force and sliding friction. The formula is expressed as: Where k is a constant representing the proportionality coefficient of force, which is related to the physical properties of the individual, g represents a function representing the change of force with distance or relative speed, and r i represents the psychological radius of individual i, r j represents the psychological radius of individual j, d ij represents the actual distance between individual i and target individual j, n ij is a unit vector, indicating the direction from individual i to individual j, Δv ij represents the speed change between individuals i and j, t ij represents the contact time or interaction time between individuals i and j; g(x) is a piecewise function. It takes effect when x exceeds 0 and becomes 0 when it is below 0.

6. The method for simulating and optimizing efficient indoor evacuation based on social force model according to claim 1, characterized in that: Step 3 is as follows: When there is an obstacle in the scene, the force between the individual and the obstacle is regarded as the comprehensive experience of the individual's psychological willingness and physical interaction with the obstacle, and the physical force is composed of the squeezing force generated by the contact with the wall and the sliding friction generated by the relative movement after the contact with the wall; after the pedestrian contacts the wall or obstacle, the force formula between the individual and the wall or obstacle is expressed as: Where k is a constant representing the proportionality coefficient of force, which is related to the physical properties of the individual, g(x) represents a function representing the change of force with distance or relative speed, and r i represents the psychological radius of individual i, d iw represents the distance between individual i and obstacle w, n iw (t) is a unit vector, indicating the direction from individual i to obstacle w, Δn iw (t) represents the relative speed change between individual i and obstacle w; The total additional force is: From Newton's second law:

7. An efficient indoor personnel evacuation simulation and optimization system based on a social force model, used to implement the method as claimed in claim 1, characterized in that: It includes site modeling module, individual behavior modeling module, dynamic simulation module and layout optimization module; The site modeling module supports user-defined input of site dimensions, exit quantity and location, and obstacle distribution information; and automatically generates a two-dimensional or three-dimensional site layout map; The individual behavior modeling module constructs an individual dynamics model based on the social force model, and supports parameterized definitions including individual body size, expected speed, and psychological characteristics; The dynamic simulation module uses the time step iteration method to simulate individual motion trajectories, outputs time series results and overall evacuation dynamics; supports multi-threaded parallel computing; The layout optimization module uses a traversal method or a genetic algorithm to search for an optimal layout solution, and the optimization goal is the shortest evacuation time or the least congested area.

8. A computer device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method of claim 1.

9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by the processor Implement the steps of the method described in claim 1.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to claim 1 are implemented.

Images