A public safety risk assessment method for ramp areas based on crowd simulation

Abstract

The present invention relates to the technical field of public safety risk assessment, and specifically discloses a ramp area public safety risk assessment method based on crowd simulation. By real-time analysis of the distribution of pedestrians on the ramp, the interaction characteristics of pedestrians, and environmental factors, a ramp evacuation model based on speed-density rules, pedestrian speed heterogeneity rules, pedestrian ramp interaction rules, and a ramp risk assessment model are constructed. The model is used to evaluate the risk coefficient in the ramp, thereby helping relevant personnel to evaluate the risk tolerance of the slope. The present invention introduces a speed-density-based method based on a mesoscopic model into a microscopic model. The present invention makes the speed of pedestrians subject to the constraints of crowd density, which is more in line with the actual situation. The method not only retains the high precision of the microscopic model, but also improves the calculation efficiency. Through reasonable algorithm design and calculation methods, the present invention can quickly complete real-time simulation in a high-density crowd environment, reducing the calculation burden caused by grid refinement.

Description

Technical Field

[0001] The present invention relates to the technical field of public safety risk assessment, and in particular to a ramp area public safety risk assessment method based on crowd simulation. Background Art

[0002] In recent years, with the acceleration of urbanization, the flow of people and the density of crowds in public places have increased significantly. Especially at specific times such as large-scale events or holidays, public places such as shopping malls, subway stations, and stadiums are more likely to form high-density crowds. This high-density flow and gathering of people has greatly increased the risk of crowd congestion and trampling in complex terrain (such as ramps, steps, and narrow passages). Especially in ramp areas, due to the effect of gravity and changes in speed, the movement behavior and safety risks of people on such inclined terrain appear more complex and unpredictable. The existence of ramps will change the dynamic characteristics of pedestrian flow and increase the uncertainty in the evacuation process.

[0003] In the prior art, researchers have developed a variety of models to simulate and analyze the movement characteristics of crowds. Among them, the Cellular Automaton (CA) model is a common method. The CA model divides the space into grids, defines each grid unit as a "cell", and regularly updates the state of pedestrians in each cell to simulate the flow pattern of the crowd. The advantage of this model is that it has high computational efficiency and can adapt to a variety of scenarios. Therefore, it has been widely used in dense crowd movement and evacuation scenarios. However, the traditional cellular automaton model is mainly designed for planar scenes and has certain limitations when dealing with slope scenes with slopes. Traditional models usually assume that the movement speed and behavior of the crowd in all areas are relatively uniform, and fail to fully consider the dynamic impact of the slope on the speed and density of the crowd. For example, when going downhill, the gravity will increase the speed of the crowd, while when going uphill, the speed of the crowd will decrease due to gravity resistance. This speed difference will further affect the crowd density, flow, and possible congestion. Therefore, in special terrains such as slopes, the failure to reflect the actual impact of slope on crowd movement becomes the main limitation of the traditional cellular automaton model.

[0004] In addition, existing models generally lack effective means of real-time monitoring and analysis of local risks. Traditional simulation methods are usually based on static parameter settings and overall flow trends, and cannot conduct real-time assessments of local congestion, density fluctuations, and flow changes in the ramp area. This limitation of lacking real-time monitoring and early warning mechanisms makes it difficult for the model to identify and assess local risks in a timely manner in emergencies, thereby weakening the model's prediction and response capabilities in a dynamic environment.

[0005] The prior art patent application announcement number CN117688792B describes an evacuation simulation method based on an adaptive grid-based takeover evacuation model. The patent uses the geometric dimensions of the building evacuation scene, the width of the evacuation exit, the number of pedestrians and the initial position distribution to establish a mesoscopic evacuation model based on an adaptive grid, the dynamic flow rules of pedestrian density and the flow completion rules of pedestrian density. The evacuation simulation calculation is performed based on the dynamic flow rules of pedestrian density and the flow completion rules of pedestrian density in the mesoscopic evacuation model based on an adaptive grid to obtain the evacuation simulation results. The core of this method is to study the mesoscopic characteristics of pedestrian evacuation, accurately obtain the dynamic changes in pedestrian density in different grids, so as to efficiently simulate the flow of people in reality and determine the high-density crowd areas during the evacuation process.

[0006] However, the application of mesoscopic models in complex inclined scenes such as slopes has obvious limitations. The model simulates the evacuation scene by dividing it into grids. However, in complex terrain, especially in inclined environments such as ramps, the effect of slope on pedestrian movement is difficult to accurately reflect. Ramps not only affect the movement speed of pedestrians, but may also promote pedestrians to produce nonlinear movement behaviors such as acceleration and deceleration, and even form complex dynamic characteristics such as "lane phenomenon". Since the mesoscopic model cannot capture these terrain characteristics in detail, the simulation results in inclined scenes such as slopes are greatly deviated from the actual situation. In addition, the mesoscopic model is also insufficient in observing the microscopic behavioral characteristics of pedestrians during evacuation. The model mainly relies on the change of grid density to simulate pedestrian flow, but ignores key factors such as the interaction between pedestrians, individual movement characteristics and speed changes. Especially in high-density crowds, the interaction between pedestrians, local density changes and individual speed changes on ramps have an important impact on evacuation efficiency and safety. These microscopic characteristics are difficult to capture by mesoscopic models. Moreover, the mesoscopic model also has certain problems in computational efficiency. Although this method optimizes the calculation accuracy through adaptive grids, when dealing with high-density crowds and complex environments, the refined grid will increase the calculation burden and significantly affect the simulation speed. In a densely populated environment, the amount of calculation of the model increases significantly with the improvement of grid accuracy, making the simulation process more time-consuming. Therefore, in application scenarios that require efficient real-time calculations, the computational efficiency of the mesoscopic model is low, which limits its application in actual evacuation guidance.

[0007] With the continuous increase in crowd density in urban public places, especially in complex inclined scenes such as ramps, traditional evacuation simulation methods are difficult to accurately reflect the impact of ramps on pedestrian evacuation behavior, especially in simulating the acceleration and deceleration of pedestrians due to changes in ramp slope and the complex behavioral characteristics such as micro-interactions between pedestrians. Summary of the invention

[0008] In order to solve the problems existing in the prior art, the present invention provides a method for assessing public safety risk in a ramp area based on crowd simulation. By analyzing the distribution of pedestrians on the ramp, the interactive characteristics of pedestrians and environmental factors in real time, the risk coefficient in the ramp is assessed, thereby helping relevant personnel to assess the risk tolerance of the slope and formulate an evacuation guidance plan in advance. In addition, the present invention optimizes the calculation efficiency in the simulation process under the premise of ensuring the simulation accuracy. By introducing a speed-density method based on a mesoscopic model in a microscopic model, the present invention makes the speed of pedestrians subject to the restriction of crowd density, which is more in line with the actual situation. The method not only retains the high precision of the microscopic model, but also improves the calculation efficiency. Through reasonable algorithm design and calculation method, the present invention can quickly complete real-time simulation in a high-density crowd environment, reducing the calculation burden caused by grid refinement. This optimization not only provides efficient data support for the safety assessment of complex scenes such as ramps, but also provides timely decision-making basis for risk management and emergency response in emergency situations, thereby providing important technical support for the safety management of public places and solving the problems mentioned in the above background technology.

[0009] To achieve the above object, the present invention provides the following technical solution: a ramp area public safety risk assessment method based on crowd simulation, comprising the following steps:

[0010] S1. Obtain the spatial geometric data of the ramp evacuation scene and the initial data of pedestrians;

[0011] S2, extracting the geometric parameters of the ramp, the slope of the ramp and the evacuation exit settings from the spatial geometric data of the ramp evacuation scene, and extracting the movement pattern of the pedestrians, the number of pedestrians and the initial position coordinates of the pedestrians from the initial data of the pedestrians;

[0012] Step S3, using the geometric parameters of the ramp, the slope of the ramp, the evacuation exit settings, the movement pattern of pedestrians, the number of pedestrians and the initial position coordinates of pedestrians extracted in step S2, construct a ramp evacuation model based on speed-density rules, pedestrian speed heterogeneity rules, pedestrian ramp interaction rules and ramp risk assessment model;

[0013] Step S4, performing a pedestrian evacuation simulation calculation for a ramp scene according to the ramp evacuation model based on the speed-density rule, the pedestrian speed heterogeneity rule, and the pedestrian ramp interaction rule established in step S3, and recording all information at each time step;

[0014] Step S5, input the information obtained in step S4 into the ramp risk assessment model, and calculate the risk value of each grid unit at each time step, that is, the real-time local risk value;

[0015] Step S6: After the pedestrian evacuation simulation is completed, the simulation results of the entire evacuation process and the ramp risk assessment value are output.

[0016] Preferably, in step S3, the following steps are specifically included:

[0017] Step S31: Initialize the ramp scene into a discrete uniform grid according to the geometric parameters of the ramp and the evacuation exit settings, and display the ramp scene in a top view; during the initialization process, the grid pattern does not directly reflect the slope characteristics of the ramp, and the size of the gridded ramp area is set according to the actual slope size;

[0018] Step S32: Initialize pedestrians, place pedestrians in the constructed ramp scene according to their initial position coordinates, and assign them initial speeds according to the pedestrian speed heterogeneity rule, thereby constructing a ramp evacuation model based on the speed-density rule;

[0019] Step S33: Based on the ramp evacuation model based on the speed-density rule constructed in step S32, the pedestrians are divided into an uphill group and a downhill group according to their movement patterns, and basic pedestrian ramp interaction rules and corresponding movement rules are set for each group of pedestrians;

[0020] Step S34: construct a ramp risk assessment model.

[0021] Preferably, in step S32, the following steps are specifically included:

[0022] S321. Define the static field strength SFF rule. The SFF value represents the distance from the grid to the evacuation exit. The grid close to the evacuation exit is more attractive to pedestrians, reflecting the pedestrians' willingness to find the exit. That is, the static field strength defines the basic movement rule for pedestrians to evacuate to the exit.

[0023] (1) The SFF value of the grid occupied by walls and obstacles is set to ∞, that is, pedestrians cannot move into the walls and obstacles;

[0024] (2) The SFF value of the grid where the evacuation exit is located is set to 1. Starting from the exit grid, traverse all non-exit, non-wall and non-obstacle grids until all grids are assigned corresponding SFF values. The calculation rule of the SFF value is: the SFF value of the adjacent non-diagonal grid is the SFF value of the current grid + 1, and the SFF value of the adjacent diagonal grid is the SFF value of the current grid + 1.4;

[0025] (3) For multiple exit grids, traverse each exit grid and its adjacent grids in turn until all non-wall and obstacle grids are traversed. For each non-wall and obstacle grid, calculate the SFF value from all exit grids to the grid, and select the minimum value of the SFF values ​​from different exit grids to the grid as the SFF value of the grid. The specific calculation method is:

[0026] ;

[0027] Where, en is the nth exit grid, SFF en Indicates the SFF value from the nth export grid to this grid;

[0028] S322. Define the speed change rules for pedestrians and introduce speed heterogeneity rules to simulate the speed differences of individuals with different purposes in a group, including the following:

[0029] (1) The free speed of pedestrians is divided into three types: fast, normal and slow;

[0030] (2) According to the speed heterogeneity rule, pedestrians with fast and slow speeds are mixed with pedestrians with normal speeds, thus forming three groups of people with different travel purposes;

[0031] (3) The speed of pedestrians follows the speed-density rule. The speed change rule of pedestrians based on the speed-density rule is expressed by the following formula:

[0032] ;

[0033] in, represents the free speed of pedestrians; Indicates the pedestrian’s travel purpose; is the crowd density within the pedestrian moving target grid, is the maximum crowd density in the grid, , indicating that a grid can accommodate multiple pedestrians until the crowd density in the grid reaches the maximum value, thus prohibiting pedestrians from continuing to enter; when and The smaller the difference between them, the slower the pedestrian’s speed is, reflecting the restricted mobility in a high-density environment.

[0034] Preferably, in step S33, the following steps are specifically included:

[0035] S331. According to the ramp geometry parameters and the evacuation exit settings, the evacuation exit settings are subdivided into uphill evacuation exits and downhill evacuation exits. A grid has two SFF values ​​at the same time. When calculating the static field strength, only the SFF value of the represented motion mode is referenced. Pedestrians going uphill and downhill follow the corresponding static field strength when moving:

[0036] ;

[0037] in, and are the static field strengths of pedestrians going uphill and downhill, respectively. For grid (i, j), only the 8 adjacent grids are traversed; and Respectively represent the maximum distances of uphill and downhill pedestrians from the exit in the current grid; and The SFF value of one of the eight adjacent grids of the grid (i, j) representing the uphill and downhill directions, respectively. The closer the grid is to the exit, the greater the static field strength is;

[0038] S332. According to different movement modes, i.e., uphill and downhill, corresponding speed change rules are assigned to pedestrians: when pedestrians are moving downhill, their speed gradually increases with the increase of downhill distance until a stable higher speed is reached; while when pedestrians are moving uphill, their speed gradually decreases with the increase of uphill distance until a stable lower speed is reached. That is, the speed change rules for pedestrians on the ramp are:

[0039] ;

[0040] in, represents the speed of the uphill pedestrian moving to the target grid, Represents the speed at which the downhill pedestrian moves to the target grid; parameter and Represents the slope factor, which is used to control the changing trend of pedestrian speed under different slope conditions. , ; is the movement distance of pedestrians on the ramp;

[0041] S333. Define the basic ramp interaction rules in the ramp. For pedestrians in the same group, that is, they have the same movement pattern, there is attraction between them; for pedestrians in different groups, that is, they have different movement patterns, there is repulsion between them; therefore, when pedestrians with different movement patterns flow together, a self-organized "lane" phenomenon will be formed; the dynamic field strength that reflects the pedestrian interaction rules in the ramp is defined as:

[0042] ;

[0043] in Indicates the perception range of pedestrians, which is expressed as a circular area with a radius of 1m, indicating that the movement of pedestrians will be affected by other pedestrians within their perception range; and They represent pedestrians with different motion modes. If the main movement is an uphill pedestrian, then represents an uphill pedestrian, then represents pedestrians walking downhill, and vice versa; and The calculation of the dynamic field strength only involves pedestrians within the sensing range; and They represent the number of pedestrians with the same motion pattern within the statistical perception range and the number of pedestrians with different motion patterns within the perception range respectively; and They represent the vector velocities of pedestrians with the same motion mode and pedestrians with different motion modes within the sensing range respectively; It represents the sum of the vector velocities of pedestrians with the same motion mode within the perception range, and is used to characterize the overall motion trend of pedestrians with the same motion mode, reflecting the attraction between pedestrians in the same group. It reflects the repulsive force between different groups of pedestrians; and The sensitivity parameters representing the attraction force between pedestrians in the same group and the repulsion force between pedestrians in different groups are used to adjust the weights of the two forces;

[0044] Pedestrians going uphill or downhill tend to follow the same group of pedestrians while moving, and avoid intersections with pedestrians in different groups. The dynamic field strength value only acts within the current time step, and the dynamic field strength of the next time step Need to recalculate.

[0045] Preferably, the probability of a pedestrian moving within a ramp is expressed as follows:

[0046] ;

[0047] in, represents the normalization parameter, ensuring and ; and is the sensitivity parameter of static field strength and dynamic field strength, is the crowd density in the target grid, which is used to determine whether the target grid has reached the maximum density. Used to determine whether the target grid is occupied by walls and obstacles.

[0048] Preferably, in step S34, the following is specifically included:

[0049] S341. Define the factors affecting ramp risk as: ramp gradient, average speed of all pedestrians in the grid, congestion risk value of the grid and relative position of the grid. Use the AHP method to assign corresponding weights to all factors and construct a ramp risk assessment equation.

[0050] S342, in each time step, extracting the number of pedestrians, pedestrian speed and pedestrian position parameters in each grid from the ramp evacuation model based on the speed-density rule, and calculating the speed curl and congestion degree as intermediate parameters for further evaluating the congestion risk value of the grid;

[0051] S343. The ramp slope, the average speed of all pedestrians in the grid, the congestion risk value of the grid and the relative position of the grid are normalized and input into the constructed ramp risk assessment equation to obtain the final ramp risk assessment model, which is used to calculate the ramp risk assessment value in each grid.

[0052] Preferably, in step S341, the factors affecting the ramp risk are ranked from high to low in importance as follows: the congestion risk value of the grid, the average speed of all pedestrians in the grid, the slope of the ramp, and the relative position of the grid. The slope of the ramp and the relative position of the grid are determined by the geometric parameters of the ramp and remain constant throughout the evacuation process; the congestion risk value of the grid and the average speed of all pedestrians in the grid are continuously updated with the distribution and dynamic changes of pedestrians.

[0053] Preferably, in step S342, it specifically includes:

[0054] (1) By analyzing the pedestrian's x-axis and y-axis velocities, a velocity curl is proposed. The concept of velocity curl is used to measure the complexity of pedestrian paths. If pedestrians often change direction during walking, the velocity curl is large; if the pedestrian path is straight, the velocity curl is small. The velocity curl is specifically defined as:

[0055] ;

[0056] In each grid, calculate the offset of all pedestrian movement directions separately ; The larger the value, the stronger the pedestrian's willingness to change the direction of movement. Value minus minimum The value is defined as the velocity curl of this grid;

[0057] (2) The congestion degree is calculated based on the obtained velocity curl. The congestion degree is reflected by the relationship between the velocity curl within the grid and the average speed of all pedestrians in the grid. The larger the average speed of pedestrians in the grid, the smaller the congestion degree, and vice versa:

[0058] ;

[0059] (3) Based on the obtained congestion degree, the congestion risk value of the grid is finally evaluated. The congestion risk value is reflected in the relationship between the congestion degree and the pedestrian density within the grid. The greater the crowd density in the area, the greater the crowd risk, and vice versa:

[0060] ;

[0061] Where r is the radius of a circle centered at the grid center and tangent to the grid boundary, j represents all pedestrians in the grid, represents the position coordinates of pedestrian j, is the position coordinate of the grid, represents the square of the Euclidean distance between pedestrian j and the grid, is an exponential decay term, which indicates that the influence of pedestrian j on the grid position decays with distance; pedestrians far from the center of the grid have little impact on the congestion risk of the grid.

[0062] Preferably, in step S343, the ramp risk assessment model is expressed as:

[0063] ;

[0064] in, represents the risk value in the grid (i, j), The value represents the weight of each factor. represents the normalized value of the average speed of all pedestrians in the grid (i, j), represents the normalized value of the congestion risk of the grid (i, j), Represents the normalized value of the relative position of the grid (i, j), A normalized value representing the slope of the ramp.

[0065] The beneficial effects of the present invention are:

[0066] 1) The present invention can improve simulation accuracy and computational efficiency: by rationally designing a speed-density-based ramp evacuation model, the dynamic behavior of pedestrians in a ramp environment can be accurately simulated, and the microscopic interactions between pedestrians and environmental impacts can be fully considered. In particular, in ramp scenarios, the acceleration, deceleration, and behavioral patterns of pedestrians are effectively simulated, thereby significantly improving the simulation accuracy. At the same time, the present invention also reduces the amount of calculation and improves the simulation efficiency by optimizing the calculation method while ensuring the simulation accuracy. In a high-density pedestrian environment, the present invention can quickly complete real-time simulation, providing technical support for the simulation of large-scale evacuation scenarios.

[0067] 2) The present invention can simultaneously perform pedestrian ramp evacuation simulation and ramp risk assessment: The risk assessment model of the present invention can analyze the risk factor in the ramp in real time, and combine the distribution data of pedestrians with environmental factors to provide relevant personnel with a visual risk distribution. This technology can not only provide support for the safety management of public places, but also help managers plan reasonable evacuation strategies in advance before large-scale crowd-intensive activities, thereby improving the level of safety management. Through the model evaluation results, managers can better understand the risk tolerance of different ramps and take appropriate countermeasures to reduce casualties caused by improper evacuation.

[0068] 3) The present invention has a significant contribution in economic effect: The present invention has a high application value in public safety management, especially in large public places (such as subway stations, shopping malls, stadiums, etc.) and special scenarios (such as disaster drills, emergency response, etc.), and can provide scientific decision-making support. Through the precise simulation of ramps and evacuation processes, the limitations of traditional manual evaluation methods are avoided, saving a lot of manual resources and time costs. At the same time, because this method can help improve evacuation efficiency and reduce crowding and trampling risks, it can effectively reduce the economic losses caused by evacuation accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is a schematic flow chart of the steps of the ramp area public safety risk assessment method based on crowd simulation of the present invention;

[0070] Figure 2 It is a schematic diagram of a simulation example of bidirectional pedestrian flow convection in an embodiment of the present invention;

[0071] Figure 3 Schematic diagram of an example of simultaneous risk assessment of bidirectional pedestrian flow in an embodiment of the present invention. DETAILED DESCRIPTION

[0072] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0073] The present invention provides a technical solution: a method for assessing public safety risks in ramp areas based on crowd simulation, such as Figure 1 As shown, the following steps are included:

[0074] S1. Obtain the spatial geometric data of the ramp evacuation scene and the initial data of pedestrians.

[0075] The spatial geometric data of the ramp scene is obtained through architectural drawings or shooting equipment, and the distribution data of pedestrians is initialized according to the crowd size required for evacuation simulation.

[0076] S2. Extract the geometric parameters of the ramp, the ramp slope and the evacuation exit settings from the spatial geometric data of the ramp evacuation scene, and extract the pedestrian's movement pattern, the number of pedestrians and the initial position coordinates of the pedestrians from the initial data of the pedestrians.

[0077] Step S3: Using the geometric parameters of the ramp, the slope of the ramp, the evacuation exit settings, the movement pattern of pedestrians, the number of pedestrians and the initial position coordinates of pedestrians extracted in step S2, a ramp evacuation model based on speed-density rules, pedestrian speed heterogeneity rules, pedestrian ramp interaction rules and ramp risk assessment model are constructed. Specifically, the following steps are included:

[0078] Step S31: According to the geometric parameters of the ramp and the evacuation exit settings, the ramp scene is initialized to a discrete uniform grid, and the ramp scene is displayed in a top view. During the initialization process, the grid graphics do not directly reflect the slope characteristics of the ramp, and the size of the gridded ramp area is set according to the actual slope size.

[0079] The size of each grid is set to 0.5m × 0.5m. Pedestrians are abstracted as discrete individuals occupying a specific grid unit with a size of 0.25m × 0.25m. Pedestrians can only be located in one grid at a time. A grid can be occupied by a wall or obstacle, multiple pedestrians, or empty.

[0080] Step S32: Initialize pedestrians, place pedestrians in the constructed ramp scene according to their initial position coordinates, and assign them initial speeds according to the pedestrian speed heterogeneity rule, thereby constructing a ramp evacuation model based on the speed-density rule. Specifically, the steps include:

[0081] S321, define the static field strength SFF rule. The SFF remains unchanged throughout the simulation process and does not change over time. The SFF value represents the distance from the grid to the evacuation exit. The grid close to the evacuation exit is more attractive to pedestrians, reflecting the pedestrians' willingness to find an exit. That is, the static field strength defines the basic movement rules for pedestrians to evacuate to the exit;

[0082] (1) The SFF value of the grid occupied by walls and obstacles is set to ∞, that is, pedestrians cannot move into the walls and obstacles;

[0083] (2) The SFF values ​​of all grids where evacuation exits are located are set to 1. Starting from these exit grids, all non-exit, non-wall and non-obstacle grids are traversed until all grids are assigned corresponding SFF values. The calculation rule of the SFF value is: the SFF value of the adjacent non-diagonal grid is the SFF value of the current grid + 1, and the SFF value of the adjacent diagonal grid is the SFF value of the current grid + 1.4;

[0084] (3) In this embodiment, the evacuation exit is composed of multiple exit grids, so the SFF values ​​of all exit grids are set to 1. For multiple exit grids, each exit grid and its adjacent grids are traversed in turn until all non-wall and obstacle grids are traversed. For each non-wall and obstacle grid, the SFF values ​​from all exit grids to the grid are calculated, and the minimum value of the SFF values ​​from different exit grids to the grid is selected as the SFF value of the grid; the specific calculation method is:

[0085] ;

[0086] Where, en is the nth exit grid, SFF en Indicates the SFF value from the nth export grid to this grid;

[0087] S322. Define the speed change rules for pedestrians and introduce speed heterogeneity rules to simulate the speed differences of individuals with different purposes in a group, including the following:

[0088] (1) The free speed of pedestrians is divided into three types: fast, normal and slow; the speeds are 1.45m / s, 1.34m / s and 1.04m / s respectively.

[0089] (2) According to the speed heterogeneity rule, pedestrians with fast speed and slow speed are mixed with pedestrians with normal speed respectively, thus forming groups with three different travel purposes. Specifically, there are three travel purpose combinations: i) peak commuting: fast-speed pedestrians (80%), normal-speed pedestrians (20%); ii) daily commuting: normal-speed pedestrians (100%); iii) leisure travel: slow-speed pedestrians (80%), normal-speed pedestrians (20%);

[0090] (3) The speed of pedestrians follows the speed-density rule. The speed change rule of pedestrians based on the speed-density rule is expressed by the following formula:

[0091] ;

[0092] in, represents the free speed of pedestrians; Indicates the travel purpose of pedestrians, and the three travel purposes correspond to peak commuting, daily commuting and leisure travel The values ​​are 1.20, 1.11 and 0.84 respectively; is the crowd density within the pedestrian moving target grid, is the maximum crowd density in the grid, , indicating that a grid can accommodate multiple pedestrians until the crowd density in the grid reaches the maximum value, thus prohibiting pedestrians from continuing to enter; when and The smaller the difference between them, the slower the pedestrian’s speed is, reflecting the restricted mobility in a high-density environment.

[0093] At this point, the basic movement rules of pedestrians have been defined, which are applicable to the evacuation simulation of all buildings or plane scenes. Next, the movement rules applicable to the ramp area will be defined. In this embodiment, pedestrians involve two movement modes (uphill and downhill). When pedestrians of different modes move at the same time, convection effects will occur and different movement characteristics will be exhibited. To simplify the processing, the movement rules of pedestrians in the two movement modes on the ramp will be set separately.

[0094] Step S33: Based on the ramp evacuation model based on the speed-density rule constructed in step S32, the pedestrians are divided into uphill groups and downhill groups according to their movement patterns, and basic pedestrian ramp interaction rules and corresponding movement rules are set for each group of pedestrians. Specifically, the steps include:

[0095] S331. Calculate the static field strength in the ramp scene according to the SFF value calculation method defined in step S321. According to the ramp geometry parameters and the evacuation exit settings, the evacuation exit settings are subdivided into uphill evacuation exits and downhill evacuation exits; a grid has two SFF values ​​at the same time, and only the SFF value of the represented motion mode is referenced when calculating the static field strength. Pedestrians moving uphill and downhill follow the corresponding static field strength when moving:

[0096] ;

[0097] in, and are the static field strengths of pedestrians going uphill and downhill, respectively. For grid (i, j), only the 8 adjacent grids are traversed; and Respectively represent the maximum distances of uphill and downhill pedestrians from the exit in the current grid; and The SFF value of one of the eight adjacent grids of the grid (i, j) representing the uphill and downhill directions, respectively. The closer the grid is to the exit, the greater the static field strength is;

[0098] S332. According to different movement modes, i.e., uphill and downhill, corresponding speed change rules are assigned to pedestrians: when pedestrians are moving downhill, their speed gradually increases with the increase of downhill distance until a stable higher speed is reached; while when pedestrians are moving uphill, their speed gradually decreases with the increase of uphill distance until a stable lower speed is reached. That is, the speed change rules for pedestrians on the ramp are:

[0099] ;

[0100] in, represents the speed of the uphill pedestrian moving to the target grid, represents the speed of the downhill pedestrian moving to the target grid; is the movement distance of pedestrians on the ramp; parameter and Represents the slope factor, which is used to control the changing trend of pedestrian speed under different slope conditions. and The former is the downhill acceleration factor, and the latter is the uphill deceleration factor, and the corresponding values ​​are and ; The closer the value is to 1, the less obvious the deceleration effect is. When , it means there is no deceleration effect. Similarly, The closer the value is to 1, the less obvious the deceleration effect is. When , it means there is no acceleration effect. Therefore, The smaller the value, the stronger the deceleration effect. The larger the value, the stronger the acceleration effect. and The specific value needs to be obtained based on the actual experimental data. Figure 2 and Figure 3 The simulation process of evacuation under a 9° ramp is shown. The ramp area is 43 meters long and 5 meters wide, corresponding to , .

[0101] S333. Define the basic ramp interaction rules in the ramp. For pedestrians in the same group, that is, they have the same movement pattern, there is attraction between them; for pedestrians in different groups, that is, they have different movement patterns, there is repulsion between them; therefore, when pedestrians with different movement patterns flow together, a self-organized "lane" phenomenon will be formed; the dynamic field strength that reflects the pedestrian interaction rules in the ramp is defined as:

[0102] ;

[0103] in Indicates the perception range of pedestrians, which is expressed as a circular area with a radius of 1m, indicating that the movement of pedestrians will be affected by other pedestrians within their perception range; and They represent pedestrians with different motion modes. If the main movement is an uphill pedestrian, then represents an uphill pedestrian, then represents pedestrians walking downhill, and vice versa; and The calculation of the dynamic field strength only involves pedestrians within the sensing range; and It represents the number of pedestrians with the same motion mode within the statistical perception range and the number of pedestrians with different motion modes within the perception range; and They represent the vector velocities of pedestrians with the same motion mode and pedestrians with different motion modes within the sensing range respectively; It represents the sum of the vector velocities of pedestrians with the same motion mode within the perception range, and is used to characterize the overall motion trend of pedestrians with the same motion mode, reflecting the attraction between pedestrians in the same group. It reflects the repulsive force between different groups of pedestrians; and The sensitivity parameters representing the attraction force between pedestrians in the same group and the repulsion force between pedestrians in different groups are used to adjust the weights of the two forces;

[0104] Pedestrians going uphill or downhill tend to follow the same group of pedestrians while moving, and avoid intersections with pedestrians in different groups. , The dynamic field strength value only applies to the current time step. The dynamic field strength of the next time step Need to recalculate.

[0105] The probability of a pedestrian moving within the ramp is expressed as follows:

[0106] ;

[0107] in, represents the normalization parameter, ensuring and ; and is the sensitivity parameter of static field strength and dynamic field strength, and in this embodiment, the value is K s = K d = 0.5, is the crowd density in the target grid, which is used to determine whether the target grid has reached the maximum density. Used to determine whether the target grid is occupied by walls and obstacles.

[0108] Step S34: construct a ramp risk assessment model. Specifically, it includes the following:

[0109] S341. Define the factors affecting ramp risk as: ramp gradient, average speed of all pedestrians in the grid, congestion risk value of the grid and relative position of the grid. Use the AHP method to assign corresponding weights to them and construct a ramp risk assessment equation. The ramp risk influencing factors are evaluated by the Delphi expert evaluation method, and the importance is ranked from high to low as follows: congestion risk value in the grid, average speed of all pedestrians in the grid, ramp gradient and relative position of the grid. Then, the four key factors are evaluated by the AHP method, and the weights are: 0.583, 0.215, 0.145, 0.057 respectively.

[0110] S342. In each time step, the number of pedestrians, pedestrian speed and pedestrian position parameters in each grid are extracted from the ramp evacuation model based on the speed-density rule, and the speed curl and congestion degree are calculated as intermediate parameters to further evaluate the congestion risk value of the grid.

[0111] Specifically include:

[0112] (1) By analyzing the pedestrian's x-axis and y-axis velocities, a velocity curl is proposed. The concept of velocity curl is used to measure the complexity of pedestrian paths. If pedestrians often change direction during walking, the velocity curl is large; if the pedestrian path is straight, the velocity curl is small. The velocity curl is specifically defined as:

[0113] ;

[0114] In each grid, calculate the offset of all pedestrian movement directions separately . The larger the value, the stronger the pedestrian's willingness to change the direction of movement. Value minus minimum The value is defined as the velocity curl of this grid;

[0115] (2) The congestion degree is calculated based on the obtained velocity curl. The congestion degree is reflected by the relationship between the velocity curl within the grid and the average speed of all pedestrians in the grid. The larger the average speed of pedestrians in the grid, the smaller the congestion degree, and vice versa:

[0116] ;

[0117] (3) Based on the obtained congestion degree, the congestion risk value of the grid is finally evaluated. The congestion risk value is reflected in the relationship between the congestion degree and the pedestrian density within the grid. The greater the crowd density in the area, the greater the crowd risk, and vice versa:

[0118] ;

[0119] Where r is the radius of a circle centered at the grid center and tangent to the grid boundary, j represents all pedestrians in the grid, represents the position coordinates of pedestrian j, is the position coordinate of the grid, represents the square of the Euclidean distance between pedestrian j and the grid, is an exponential decay term, which indicates that the influence of pedestrian j on the grid position decays with distance; pedestrians far from the center of the grid have little impact on the congestion risk of the grid.

[0120] S343. The ramp slope, the average speed of all pedestrians in the grid, the congestion risk value of the grid and the relative position of the grid are normalized and input into the constructed ramp risk assessment equation to obtain the final ramp risk assessment model, which is used to calculate the ramp risk assessment value in each grid.

[0121] The final ramp risk assessment model is expressed as:

[0122] ;

[0123] in, represents the risk value in the grid (i, j), The value represents the weight of each factor. represents the normalized value of the average speed of all pedestrians in the grid (i, j), represents the normalized value of the congestion risk of the grid (i, j), Represents the normalized value of the relative position of the grid (i, j), represents the normalized value of the slope. In this embodiment, the normalized slope value of the 9° slope is The relative position of the grid is the shortest vertical distance between the current grid and the entrance and exit of the ramp. The closer the grid is to the middle of the ramp, the larger the relative position value will be.

[0124] Step S4: Perform a pedestrian evacuation simulation calculation for a ramp scene according to the ramp evacuation model based on the speed-density rule, the pedestrian speed heterogeneity rule, and the pedestrian ramp interaction rule established in step S3, and record all information at each time step.

[0125] Step S5: Input the information obtained in step S4 into the ramp risk assessment model to calculate the risk value of each grid unit at each time step, that is, the real-time local risk value.

[0126] Step S6: After the pedestrian evacuation simulation is completed, the simulation results of the entire evacuation process and the ramp risk assessment value are output.

[0127] Figure 2The evacuation simulation snapshot at the 40th time step is shown. There are 400 pedestrians in the scene, and the ratio of uphill pedestrians to downhill pedestrians is 1:1. The blue pedestrians in the figure are downhill pedestrians, moving from left to right, and the rose-red pedestrians are uphill pedestrians, moving from right to left. It can be observed that under the setting of this embodiment, the convective pedestrians show obvious self-organized "lane" phenomenon. Figure 3 The risk assessment diagram at the same time step is shown. After normalization, the risk value range is [0,1]. The highest risk value in the figure is 0.68, which appears at the initial convection of pedestrians on both sides, and the lowest is 0, which appears at both ends of the ramp.

[0128] Through simulation-based evacuation results and visual real-time risk assessment, a scientific basis can be provided for evacuation management and emergency response. By predicting pedestrian movement behavior in ramp areas and their interaction with environmental factors, potential risk areas and key bottlenecks can be effectively identified, so that reasonable, multi-scenario evacuation planning can be formulated in advance to improve public safety.

[0129] In the embodiments provided in the embodiments of the present invention, it should be noted that, in this article, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0130] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0131] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0132] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0133] The "first\second" mentioned in the embodiments is only to distinguish similar objects, and does not represent a specific order for the objects. It is understandable that the "first\second" can be interchanged with the specific order or sequence where permitted. It should be understood that the objects distinguished by "first\second" can be interchanged where appropriate, so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.

[0134] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for public safety risk assessment in ramp areas based on crowd simulation, characterized in that: The steps include: S1. Obtain the spatial geometric data of the ramp evacuation scene and the initial data of pedestrians; S2, extracting the geometric parameters of the ramp, the slope of the ramp and the evacuation exit settings from the spatial geometric data of the ramp evacuation scene, and extracting the movement pattern of the pedestrians, the number of pedestrians and the initial position coordinates of the pedestrians from the initial data of the pedestrians; Step S3: using the geometric parameters of the ramp, the slope of the ramp, the evacuation exit settings, the movement pattern of pedestrians, the number of pedestrians and the initial position coordinates of pedestrians extracted in step S2, construct a ramp evacuation model based on the speed-density rule, the pedestrian speed heterogeneity rule, the pedestrian ramp interaction rule and the ramp risk assessment model; constructing the ramp risk assessment model specifically includes the following: S341. Define the factors affecting ramp risk as: ramp gradient, average speed of all pedestrians in the grid, congestion risk value of the grid and relative position of the grid. Use the AHP method to assign corresponding weights to all factors and construct a ramp risk assessment equation. The factors affecting ramp risk are ranked from high to low in importance: the congestion risk value of the grid, the average speed of all pedestrians in the grid, the slope of the ramp, and the relative position of the grid. The slope of the ramp and the relative position of the grid are determined by the geometric parameters of the ramp and remain constant throughout the evacuation process. The congestion risk value of the grid and the average speed of all pedestrians in the grid are continuously updated with the distribution and dynamic changes of pedestrians. S342, in each time step, extracting the number of pedestrians, pedestrian speed and pedestrian position parameters in each grid from the ramp evacuation model based on the speed-density rule, and calculating the speed curl and congestion degree as intermediate parameters for further evaluating the congestion risk value of the grid; S343, the ramp slope, the average speed of all pedestrians in the grid, the congestion risk value of the grid and the relative position of the grid are normalized and input into the constructed ramp risk assessment equation to obtain the final ramp risk assessment model; the ramp risk assessment model is expressed as: Among them, R ij represents the risk value in the grid (i, j), and the β value represents the weight value of each factor. Represents the normalized value of the average speed of all pedestrians in grid (i, j), Cd′ ij Represents the normalized value of the congestion risk of the grid (i, j), Dis′ ij represents the normalized value of the relative position of the grid (i, j), and θ′ represents the normalized value of the slope of the ramp; Step S4, performing a pedestrian evacuation simulation calculation for a ramp scene according to the ramp evacuation model based on the speed-density rule, the pedestrian speed heterogeneity rule, and the pedestrian ramp interaction rule established in step S3, and recording all information at each time step; Step S5, input the information obtained in step S4 into the ramp risk assessment model, and calculate the risk value of each grid unit at each time step, that is, the real-time local risk value; Step S6: After the pedestrian evacuation simulation is completed, the simulation results of the entire evacuation process and the ramp risk assessment value are output.

2. The method for public safety risk assessment in ramp areas based on crowd simulation according to claim 1 is characterized in that: In step S3, the following steps are specifically included: Step S31: Initialize the ramp scene into a discrete uniform grid according to the geometric parameters of the ramp and the evacuation exit settings, and display the ramp scene in a top view; during the initialization process, the grid pattern does not directly reflect the slope characteristics of the ramp, and the size of the gridded ramp area is set according to the actual slope size; Step S32: Initialize pedestrians, place pedestrians in the constructed ramp scene according to their initial position coordinates, and assign them initial speeds according to the pedestrian speed heterogeneity rule, thereby constructing a ramp evacuation model based on the speed-density rule; Step S33: Based on the ramp evacuation model based on the speed-density rule constructed in step S32, the pedestrians are divided into an uphill group and a downhill group according to their movement patterns, and basic pedestrian ramp interaction rules and corresponding movement rules are set for each group of pedestrians; Step S34: construct a ramp risk assessment model.

3. The method for public safety risk assessment in ramp areas based on crowd simulation according to claim 2 is characterized in that: In step S32, the following steps are specifically included: S321. Define the static field strength SFF rule. The SFF value represents the distance from the grid to the evacuation exit. The grid close to the evacuation exit is more attractive to pedestrians, reflecting the pedestrians' willingness to find the exit. That is, the static field strength defines the basic movement rule for pedestrians to evacuate to the exit. (1) The SFF value of the grid occupied by walls and obstacles is set to ∞, that is, pedestrians cannot move into the walls and obstacles; (2) The SFF value of the grid where the evacuation exit is located is set to 1. Starting from the exit grid, all non-exit, non-wall and non-obstacle grids are traversed until all grids are assigned corresponding SFF values. The SFF value calculation rule is: the SFF value of the adjacent non-diagonal grid is the SFF value of the current grid + 1, and the SFF value of the adjacent diagonal grid is the SFF value of the current grid + 1.4; (3) For multiple exit grids, traverse each exit grid and its adjacent grids in turn until all non-wall and obstacle grids are traversed. For each non-wall and obstacle grid, calculate the SFF value from all exit grids to the grid, and select the minimum value of the SFF values ​​from different exit grids to the grid as the SFF value of the grid. The specific calculation method is: SFF=min{SFF e1 ,SFF e2 ,…,SFF en } Where, en is the nth exit grid, SFF en Indicates the SFF value from the nth export grid to this grid; S322. Define the speed change rules for pedestrians and introduce speed heterogeneity rules to simulate the speed differences of individuals with different purposes in a group, including the following: (1) The free speed of pedestrians is divided into three types: fast, normal and slow; (2) According to the speed heterogeneity rule, pedestrians with fast and slow speeds are mixed with pedestrians with normal speeds, thus forming three groups of people with different travel purposes; (3) The speed of pedestrians follows the speed-density rule. The speed change rule of pedestrians based on the speed-density rule is expressed by the following formula: Among them, v M represents the free speed of pedestrians; γ represents the travel purpose of pedestrians; ρ is the crowd density in the target grid of pedestrian movement, ρ jam is the maximum crowd density in the grid, ρ≤ρ jam , indicating that a grid can accommodate multiple pedestrians until the crowd density in the grid reaches the maximum value, thus prohibiting pedestrians from entering further; when ρ and ρ jam The smaller the difference between them, the slower the pedestrian’s speed is, reflecting the restricted mobility in a high-density environment.

4. The method for public safety risk assessment in ramp areas based on crowd simulation according to claim 2 is characterized in that: In step S33, the following steps are specifically included: S331. According to the ramp geometry parameters and the evacuation exit settings, the evacuation exit settings are subdivided into uphill evacuation exits and downhill evacuation exits. A grid has two SFF values ​​at the same time. When calculating the static field strength, only the SFF value of the represented motion mode is referenced. Pedestrians going uphill and downhill follow the corresponding static field strength when moving: in, and are the static field strengths of pedestrians going uphill and downhill, respectively. For grid (i, j), only the 8 adjacent grids are traversed; and Respectively represent the maximum distances of uphill and downhill pedestrians from the exit in the current grid; and The SFF value of one of the eight adjacent grids of the grid (i, j) representing the uphill and downhill directions, respectively. The closer the grid is to the exit, the greater the static field strength is. S332. According to different movement modes, i.e., uphill and downhill, corresponding speed change rules are assigned to pedestrians: when pedestrians are moving downhill, their speed gradually increases with the increase of downhill distance until a stable higher speed is reached; while when pedestrians are moving uphill, their speed gradually decreases with the increase of uphill distance until a stable lower speed is reached. That is, the speed change rules for pedestrians on the ramp are: in, represents the speed of the uphill pedestrian moving to the target grid, represents the speed of pedestrians moving downhill to the target grid; parameters α and β represent slope factors, which are used to control the changing trend of pedestrian speed under different slope conditions, 0<β≤1, α≥1; d s is the movement distance of pedestrians on the ramp; S333. Define the basic ramp interaction rules in the ramp. For pedestrians in the same group, that is, they have the same movement pattern, there is attraction between them; for pedestrians in different groups, that is, they have different movement patterns, there is repulsion between them; therefore, when pedestrians with different movement patterns flow together, a self-organized "lane" phenomenon will be formed; the dynamic field strength that reflects the pedestrian interaction rules in the ramp is defined as: Where R represents the perception range of pedestrians, which is a circular area with a radius of 1m, indicating that the movement of pedestrians will be affected by other pedestrians within their perception range; x and y represent pedestrians with different motion modes. If the moving subject is an uphill pedestrian, x represents an uphill pedestrian, and y represents a downhill pedestrian, and vice versa; x∈R and y∈R indicate that the calculation of the dynamic field strength only involves pedestrians within the perception range; Sum x and Sum y It represents the number of pedestrians with the same motion mode within the statistical perception range and the number of pedestrians with different motion modes within the perception range; and They represent the vector velocities of pedestrians with the same motion mode and pedestrians with different motion modes within the sensing range respectively; It represents the sum of the vector velocities of pedestrians with the same motion mode within the perception range, and is used to characterize the overall motion trend of pedestrians with the same motion mode, reflecting the attraction between pedestrians in the same group. It reflects the repulsive force between different groups of pedestrians; μ and λ represent the sensitivity parameters of the attraction between pedestrians in the same group and the repulsive force between pedestrians in different groups, respectively, which are used to adjust the weights of the two forces; Pedestrians going uphill or downhill tend to follow the same group of pedestrians while moving, and avoid intersections with pedestrians in different groups. The dynamic field strength value only acts within the current time step, and the dynamic field strength of the next time step Need to recalculate.

5. The method for public safety risk assessment in ramp areas based on crowd simulation according to claim 4 is characterized in that: The probability of a pedestrian moving within the ramp is expressed as follows: Among them, N represents the normalization parameter, ensuring and K s and K d is the sensitivity parameter of static field strength and dynamic field strength, ρ is the crowd density in the target grid, which is used to determine whether the target grid has reached the maximum density, and η ij Used to determine whether the target grid is occupied by walls and obstacles.

6. The method for public safety risk assessment in ramp areas based on crowd simulation according to claim 1 is characterized in that: In step S342, it specifically includes: (1) By analyzing the pedestrian's x-axis and y-axis velocities, the velocity curl R is proposed. r The concept of velocity curl is used to measure the complexity of pedestrian paths. If pedestrians often change direction during walking, the velocity curl is large; if the pedestrian path is straight, the velocity curl is small. The velocity curl is specifically defined as: R r =max(r z )-min(r z ) In each grid, the offset r of all pedestrian movement directions is calculated separately z ; r z The larger the value, the stronger the pedestrian's willingness to change the direction of movement. z value minus the minimum r z The value is defined as the velocity curl of this grid; (2) The congestion degree is calculated based on the obtained velocity curl. The congestion degree is reflected by the relationship between the velocity curl within the grid and the average speed of all pedestrians in the grid. The larger the average speed of pedestrians in the grid, the smaller the congestion degree, and vice versa: (3) Based on the obtained congestion degree, the congestion risk value of the grid is finally evaluated. The congestion risk value is reflected in the relationship between the congestion degree and the pedestrian density within the grid. The greater the crowd density in the area, the greater the crowd risk, and vice versa: Where r is the radius of a circle centered at the grid center and tangent to the grid boundary, j represents all pedestrians in the grid, represents the position coordinates of pedestrian j, is the position coordinate of the grid, represents the square of the Euclidean distance between pedestrian j and the grid, is an exponential decay term, which indicates that the influence of pedestrian j on the grid position decays with distance; pedestrians far from the center of the grid have little impact on the congestion risk of the grid.

Images