A simulation system for evaluating evacuation of people in buildings

Abstract

The present invention discloses a simulation system for evaluating the evacuation of personnel in a building in the field of human flow simulation, including a building import module for importing building drawings, a pedestrian import module for adding pedestrians in the building drawings, pedestrians are provided with action logic, and the action logic includes sight logic and scene logic; a special scene import module is used to mark special scene tags in the building drawings, and the special scene tags include signboards, elevators, stairs, exits and gates. The elevators, stairs and exits are all provided with an attraction value for increasing the probability of the pedestrian model approaching, and the attraction value decreases as the number of pedestrians attracted increases. The sight logic is provided for pedestrians with a sight range, and pedestrians will read the distribution of pedestrians and special scene tags within the sight range, and the probability of approaching the corresponding special scene tag is calculated according to the attraction value of each special scene tag. The technical solution of the present invention can increase the judgment logic of pedestrians in some special situations and further restore the evacuation process.

Description

Technical Field

[0001] The invention belongs to the field of human flow simulation, in particular to a simulation system for evaluating the evacuation of people in a building. Background Art

[0002] Sudden events of large gatherings under complex circumstances may cause crowding and trampling accidents. Individual pedestrians in the evacuation site often cannot find the best escape direction due to unfamiliarity with the environment. The research focus in the field of pedestrian flow is to evaluate the evacuation capacity of the assembly when an emergency occurs and provide effective evacuation strategies through pedestrian flow evacuation models. The research on pedestrian flow characteristics mainly consists of two aspects: micro-behavioral characteristics of pedestrians and macro-group characteristics. The research on micro-behavioral characteristics of pedestrians mainly analyzes and studies individual pedestrian behavior and its influencing factors, such as pedestrian fear, pedestrian slow start effect, overtaking behavior, herding effect, group behavior effect, etc. The research on macro-behavioral characteristics of pedestrian flow mainly analyzes and studies pedestrian group behavior and its evaluation, analyzes and studies the crowding mechanism of pedestrian groups and the characteristics of cluster behavior, etc. At present, typical pedestrian flow evacuation models include cellular automaton models, lattice gas models, social force models, fluid dynamics models, etc. Among them, cellular automata, as a discrete model that can simulate individual behavior of pedestrians and present their microscopic characteristics and macroscopic behavior, have been widely used in pedestrian flow simulation research. They play an important role in reducing indoor pedestrian evacuation congestion, avoiding pedestrian stampede accidents, and improving the basic theory and application level of rapid evacuation in sudden gatherings.

[0003] In order to solve the above problems, patent publication number CN106354978A discloses a method for simulating indoor multi-exit pedestrian flow considering evacuation guidance. First, the indoor space is gridded, and then the evacuated personnel are initialized. The evacuated pedestrians are arranged in random blank cells with uniform random distribution and represented by dots of different colors from the guiding personnel; then the pedestrians in each cell are searched, and the guiding force on each pedestrian, the force between pedestrians, the attraction of visible exits, and the repulsion of obstacles are analyzed to determine the four directions that each pedestrian may go to, thereby obtaining the direction matrix A; finally, the direction matrix is ​​analyzed, and each pedestrian participates in the competition to obtain the actual moving direction; the pedestrian positions are updated until all pedestrians reach the exit.

[0004] The above-mentioned indoor multi-exit pedestrian flow simulation method considering evacuation guidance restores the evacuation process by analyzing the guiding force on each pedestrian, the force between pedestrians, the attraction of visible exits and the repulsion of obstacles. However, pedestrians have intellectual logic and will judge special situations instead of just going there because of attraction. For example, when evacuating through an elevator, the initial attraction of the elevator is greater than that of other channels, but the evacuation efficiency of the elevator is limited. When the number of people waiting at the elevator entrance accumulates too much, some people will change to other channels for evacuation. Summary of the invention

[0005] In order to solve the problem in the prior art that attraction is difficult to reflect the intelligence analysis of pedestrians, the purpose of the present invention is to provide a simulation system for evaluating the evacuation of personnel in buildings, which can increase the judgment logic of pedestrians in some special situations and further restore the evacuation process.

[0006] In order to achieve the above object, the technical solution of the present invention is as follows: a simulation system for evaluating evacuation of personnel in a building,

[0007] Architectural import module: used to import architectural drawings, including dxf and dwg format files;

[0008] Pedestrian import module: used to add pedestrians to architectural drawings. Pedestrians have action logic, which includes sight logic and scene logic.

[0009] Special scene import module: used to mark special scene labels in architectural drawings. Special scene labels include signboards, elevators, stairs, exits and gates. Elevators, stairs and exits are all equipped with attraction values ​​to increase the probability of pedestrian models approaching. The attraction value of the exit is greater than the attraction value of the elevator and greater than the attraction value of the building. Elevators, stairs, exits and gates will detect the number of pedestrians attracted, and the attraction value decreases as the number of pedestrians attracted increases. Elevators are equipped with rated operating time and rated load capacity. The elevator takes away the rated load of pedestrians every rated operating time. Stairs and gates are equipped with movement speed attenuation values. When pedestrians are at stairs and gates, their movement speed will be reduced according to the movement speed attenuation value.

[0010] The sight logic sets a sight range for pedestrians, and pedestrians will read the distribution of pedestrians and special scene labels within the sight range;

[0011] The scene logic is that pedestrians will behave differently according to the special scene tags within their field of vision. Specifically, when there is a sign within the field of vision, pedestrians will read the direction indicated by the sign and be attracted to move in the indicated direction. When there is an elevator, staircase, exit or gate within the field of vision, the attraction value of the special scene tag is read, and the probability of approaching the corresponding special scene tag is calculated based on the attraction value of each special scene tag.

[0012] The above solution achieves the following beneficial effects: users can import architectural drawings through the building import module to simulate the internal structure of the building. Pedestrians can be added to the architectural drawings to simulate the flow of pedestrians during evacuation.

[0013] Signs, elevators, stairs, exits, and gates are all special scene labels. Signs are used to simulate the effect of route signs or evacuation staff, guide pedestrians to their target routes, and make pedestrians move in the right direction. Elevators, stairs, exits, and gates are different evacuation channels. In a non-crowded state, pedestrians will choose exits to complete evacuation, but when one exit is crowded, they will choose other exits or reach other floors. Gates are usually set in front of exits, and there are usually multiple gates. In reality, pedestrians often prioritize in front of gates with fewer people. Therefore, different attraction values ​​are set for elevators, stairs, exits, and gates. At the same time, the attraction value will decrease as the number of pedestrians around the special scene label increases, in order to simulate the intelligent judgment behavior of pedestrians.

[0014] Pedestrians' intelligent judgment is limited to their cognition, and their familiarity with building structures varies. By setting the sight range to simulate the cognitive range of pedestrians, pedestrians will use the special scene tags within their sight range as logical references, collect the special scene tags within the sight range, and calculate the probability of approaching the corresponding special scene tags based on their attractiveness, so that pedestrians can be more restored, rather than evacuating along the best route in the entire building.

[0015] Compared with the existing technology, by marking special scene tags, more scene simulations are expanded to restore the real evacuation scene; through dynamic changes in attraction, pedestrians are given a certain intelligence and will choose the direction of movement according to the situation of the line of sight or the number of pedestrians in each channel, making the simulation more realistic.

[0016] Furthermore, when importing a multi-story building, the building import module imports the architectural drawings of each floor and marks the interconnected openings between floors. After pedestrians touch the connecting openings, they will move to the architectural drawings of the corresponding connecting openings.

[0017] Beneficial effect: By importing multi-story building drawings into multiple planes and switching to the building drawings of the next layer by moving pedestrians to the connecting port, the calculation on the z-axis is reduced, making the overall computing power requirement smaller.

[0018] Furthermore, the pedestrian model movement is simulated using a linear acceleration algorithm and a steering acceleration algorithm.

[0019] Beneficial effect: There are acceleration and deceleration processes when the human body moves, so the movement simulation through linear acceleration algorithm and steering acceleration algorithm is more realistic.

[0020] Furthermore, the pedestrian model is provided with random acceleration values ​​and magnitudes, ranging from 80% to 120% of the standard value.

[0021] Beneficial effects: In reality, everyone's physical fitness is different, so the movement state is different. By setting random acceleration values ​​and pedestrian model sizes, the physical differences between pedestrians can be simulated.

[0022] Furthermore, the action logic also includes crowding logic, which detects the number of pedestrian models within the surrounding area of ​​the pedestrian model and corrects the moving speed of the pedestrian model by 10% to 20% according to the number of pedestrian models.

[0023] Beneficial effect: It is difficult for pedestrians to maintain normal movement in crowded conditions. Therefore, by detecting the number of pedestrian models within the surrounding area of ​​the pedestrian model, it is determined whether it is in a crowded environment and its speed is corrected.

[0024] Furthermore, the action logic also includes a guidance logic. The guidance logic divides pedestrians into attracted, sub-attracted and unattracted states. When there are no signs, elevators, stairs, exits and gates within the pedestrian's field of vision, the pedestrian will follow the pedestrian in the attracted state. At this time, the pedestrian's state is sub-attracted. When there are no signs, elevators, stairs, exits, gates and pedestrians in the attracted state within the pedestrian's field of vision, the pedestrian will follow the pedestrian in the sub-attracted state. When there are no signs, elevators, stairs, exits, gates, pedestrians in the attracted state and pedestrians in the sub-attracted state within the pedestrian's field of vision, the pedestrian will move in a random non-wall direction.

[0025] Beneficial effects: In real situations, there are situations where pedestrians do not know the evacuation direction, and pedestrians usually follow other pedestrians. Therefore, if there are no signs within the pedestrian's field of vision, pedestrians will follow pedestrians who know the evacuation direction, and the driving effect will be transmitted through people, so pedestrians in the following state are secondarily attracted, and pedestrians who do not know the evacuation direction will follow the secondarily attracted pedestrians. Simulating pedestrian behavior under group psychology, pedestrians who have no signs within their field of vision can only search by themselves until they are attracted.

[0026] Furthermore, the special scenario import module is also used to add, delete or move signboards in real time during the simulation process.

[0027] Beneficial effects: Compared with other special scene labels or buildings themselves, mobile signs have strong functions, low construction costs and are easy to deploy. Therefore, users can add, delete or move signs in real time to explore feasible and easy-to-reach evacuation plans.

[0028] Furthermore, it also includes an evacuation effect evaluation module, which is used to collect the distribution of pedestrians in the building drawings in real time, and output a pedestrian density heat map according to the distribution of pedestrians. The pedestrian density heat map includes the current density, maximum density, regional service level index, evacuation congestion index, and records the time values ​​required for the pedestrian evacuation to reach 50%, 95%, and 100%.

[0029] Beneficial effects: Density heat maps can analyze the population density distribution during the period through color depth, making it easier for users to observe the evacuation effect and analyze the degree of congestion during the process. The evaluation of evacuation effect is based on multiple evacuation purposes. The pedestrian evacuation volume of 50%, 95% and 100% respectively represents the emergency evacuation effect in the early stage of evacuation, the time required for crowd evacuation to complete, and the time required for complete evacuation. Multi-indicator data can more comprehensively evaluate the evacuation effect.

[0030] Furthermore, the personnel have a pre-action time before responding to the evacuation action, and the pre-action time follows the Weibull distribution:

[0031] F(x)=(ac[1-exp(-x c )] a-1 ·exp(-x c )x c-1 -loc) / scale

[0032] Where x is the pre-action time, a and c are two parameters in the Weibull probability density distribution function, loc is the position translation coefficient, and scale is the scaling coefficient.

[0033] Beneficial effects: The pre-action time can simulate the evacuation start-up process of pedestrians, making the pedestrian model more realistic.

[0034] Furthermore, it also includes a zone evacuation module, which is used to divide the architectural drawings into multiple areas, and each area has a preset evacuation start time; the area division includes an exit division mode and a crowd division mode; in the exit division mode, the pedestrian model in the area only searches for exits to evacuate within the area, and in the crowd division mode, the pedestrian model is only restricted by the evacuation start time. The zone evacuation module is used to simulate zone evacuation, stage evacuation and zone-to-stage evacuation.

[0035] Beneficial effects: The zoned evacuation module can simulate more evacuation modes, evacuate the crowd in batches and through different exits, and can simulate situations under different evacuation plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 Schematic diagram of an embodiment of the present invention.

[0037] Figure 2 Schematic diagram of pedestrian logic.

[0038] Figure 3 Schematic diagram of pedestrian movement.

[0039] Figure 4 This is a schematic diagram of zone evacuation. DETAILED DESCRIPTION

[0040] The following is further described in detail through specific implementation methods:

[0041] Embodiment 1

[0042] The embodiment is basically as shown in the attached Figure 1 To Attachment Figure 4 As shown:

[0043] A simulation system for evaluating evacuation of personnel in a building. Building import module: used to import building drawings. The types of building drawings include dxf and dwg format files.

[0044] Pedestrian import module: used to add pedestrians to architectural drawings. Pedestrians have action logic, which includes sight logic and scene logic.

[0045] Special scene import module: used to mark special scene labels in architectural drawings. Special scene labels include signboards, elevators, stairs, exits and gates. Elevators, stairs and exits are all equipped with attraction values ​​to increase the probability of pedestrian models approaching. The attraction value of the exit is greater than the attraction value of the elevator and greater than the attraction value of the building. Elevators, stairs, exits and gates will detect the number of pedestrians attracted, and the attraction value decreases as the number of pedestrians attracted increases. Elevators are equipped with rated operating time and rated load capacity. The elevator takes away the rated load of pedestrians every rated operating time. Stairs and gates are equipped with movement speed attenuation values. When pedestrians are at stairs and gates, their movement speed will be reduced according to the movement speed attenuation value.

[0046] The sight logic sets a sight range for pedestrians, and pedestrians will read the distribution of pedestrians and special scene labels within the sight range;

[0047] The scene logic is that pedestrians will behave differently according to the special scene tags within their field of vision. Specifically, when there is a sign within the field of vision, pedestrians will read the direction indicated by the sign and be attracted to move in the indicated direction. When there is an elevator, staircase, exit or gate within the field of vision, the attraction value of the special scene tag is read, and the probability of approaching the corresponding special scene tag is calculated based on the attraction value of each special scene tag.

[0048] The specific implementation process is as follows: users can import architectural drawings through the building import module to simulate the internal structure of the building. Pedestrians are added to the architectural drawings to simulate the flow of pedestrians during evacuation.

[0049] Signs, elevators, stairs, exits, and gates are all special scene tags. Signs are used to simulate the effect of route signs or evacuation staff, guide pedestrians to their target routes, and make pedestrians move in the right direction. Elevators, stairs, exits, and gates are different evacuation channels. In a non-crowded state, pedestrians will choose the exit to complete the evacuation, but when one exit is crowded, they will choose other exits or reach other floors. Gates are usually set in front of the exits, and there are usually multiple gates. In reality, pedestrians often prioritize in front of gates with fewer people. Therefore, different attraction values ​​are set for elevators, stairs, exits, and gates. At the same time, the attraction value will decrease as the number of pedestrians around the special scene tags increases, so as to simulate the intelligent judgment behavior of pedestrians. At the same time, special scenes have operation logic. Specifically, elevators have running time and maximum passenger capacity to simulate the actual transportation process. In the case of stairs and gates, pedestrians need to follow the movement rules of stairs and gates, so the speed will be reduced.

[0050] Pedestrians' intelligent judgment is limited to their cognition, and their familiarity with building structures varies. By setting the sight range to simulate the cognitive range of pedestrians, pedestrians will use the special scene tags within their sight range as logical references, collect the special scene tags within the sight range, and calculate the probability of approaching the corresponding special scene tags based on their attractiveness, so that pedestrians can be more restored, rather than evacuating along the best route in the entire building.

[0051] This invention expands more scene simulations by marking special scene tags and restoring real evacuation scenes; through dynamic changes in attraction, pedestrians are given a certain intelligence and will choose the direction of movement based on the line of sight or the number of pedestrians in each channel, making the simulation more realistic.

[0052] Embodiment 2

[0053] The difference from the above embodiment is that when the building import module imports a multi-story building, it imports the architectural drawings of each floor, marks the connecting openings between floors, and pedestrians will move to the architectural drawings of the corresponding connecting openings after touching the connecting openings.

[0054] The specific implementation process is as follows: by importing multi-story building drawings into multiple planes, and by having pedestrians move to the connecting port to switch to the building drawings of the next floor, the calculation on the z-axis is reduced, making the overall computing power requirements smaller.

[0055] Embodiment 3

[0056] The difference from the above embodiment is that the pedestrian model movement adopts a linear acceleration algorithm and a steering acceleration algorithm to perform movement simulation.

[0057] The specific implementation process is as follows: when the human body moves, there are acceleration and deceleration processes, so the movement simulation through linear acceleration algorithm and steering acceleration algorithm is more realistic.

[0058] Embodiment 4

[0059] The difference from the above embodiment is that the pedestrian model is provided with random acceleration values ​​and magnitudes, and the value range is within 80% to 120% of the standard value.

[0060] The specific implementation process is as follows: In reality, everyone's physical fitness is different, so the movement state is different. By setting random acceleration values ​​and pedestrian model sizes, the physical differences between pedestrians are simulated.

[0061] Embodiment 5

[0062] The difference from the above embodiment is that the action logic also includes crowding logic, which detects the number of pedestrian models within the surrounding area of ​​the pedestrian model and corrects the moving speed of the pedestrian model by 10% to 20% according to the number of pedestrian models.

[0063] The specific implementation process is as follows: It is difficult for pedestrians to maintain normal movement in a crowded environment. Therefore, by detecting the number of pedestrian models around the pedestrian model, it is determined whether it is in a crowded environment and its speed is corrected.

[0064] Embodiment 6

[0065] The difference from the above embodiments is that the action logic also includes guidance logic, which divides pedestrians into attracted, sub-attracted and unattracted states. When there are no signs, elevators, stairs, exits and gates within the pedestrian's field of vision, the pedestrian will follow the pedestrian in the attracted state. At this time, the pedestrian's state is sub-attracted. When there are no signs, elevators, stairs, exits, gates and pedestrians in the attracted state within the pedestrian's field of vision, the pedestrian will follow the pedestrian in the sub-attracted state. When there are no signs, elevators, stairs, exits, gates, pedestrians in the attracted state and pedestrians in the sub-attracted state within the pedestrian's field of vision, the pedestrian will move in a random non-wall direction.

[0066] The specific implementation process is as follows: In reality, there are situations where pedestrians do not know the evacuation direction, and pedestrians usually follow other pedestrians. Therefore, if there are no instructions within the pedestrian's field of vision, pedestrians will follow pedestrians who know the evacuation direction, and the driving effect will be transmitted through people, so pedestrians in the following state are secondarily attracted, and pedestrians who do not know the evacuation direction will follow the secondarily attracted pedestrians. Simulating pedestrian behavior under group psychology, pedestrians who have no instructions within their field of vision can only search by themselves until they are attracted.

[0067] Embodiment 7

[0068] The difference from the above embodiment is that the special scene import module is also used to add, delete or move signboards in real time during the simulation process.

[0069] The specific implementation process is as follows: Compared with other special scene labels or buildings themselves, mobile signs have strong functions, low construction costs and are easy to deploy. Therefore, users can add, delete or move signs in real time to explore feasible and easy-to-reach evacuation plans.

[0070] Embodiment 8

[0071] The difference from the above embodiment is that it also includes an evacuation effect evaluation module, which is used to collect the distribution of pedestrians in the building drawings in real time, and output a pedestrian density heat map according to the distribution of pedestrians. The pedestrian density heat map includes the current density, maximum density, regional service level index, evacuation congestion index, and records the time values ​​required for the pedestrian evacuation to reach 50%, 95%, and 100%.

[0072] The specific implementation process is as follows: The density heat map can analyze the population density distribution within the period through color depth, which is convenient for users to observe the evacuation effect and analyze the degree of congestion during the process. The evaluation of the evacuation effect is based on multiple evacuation purposes. The pedestrian evacuation volume of 50%, 95% and 100% respectively represents the emergency evacuation effect in the early stage of evacuation, the time required for crowd evacuation to complete, and the time required for complete evacuation. Multi-indicator data can evaluate the evacuation effect more comprehensively.

[0073] Embodiment 9

[0074] The difference from the above embodiment is that: the personnel are provided with a pre-action time before responding to the evacuation action, and the pre-action time follows the Weibull distribution:

[0075] F(x)=(ac[1-exp(-x c )] a-1 ·exp(-x c )x c-1 -loc) / scale

[0076] Wherein, x is the pre-action time, a and c are two parameters in the Weibull probability density distribution function, loc is the position translation coefficient, and scale is the scaling coefficient. In this embodiment, the value of a is 116.399, c is 0.366, loc is -1.118, and scale is 0.241.

[0077] The specific implementation process is as follows: the pre-action time can simulate the evacuation start-up process of pedestrians, making the pedestrian model more realistic.

[0078] Embodiment 10

[0079] The difference from the above embodiment is that: it also includes a zone evacuation module, which is used to divide the architectural drawings into multiple areas, and each area has a preset evacuation start time; the area division includes an exit division mode and a crowd division mode; in the exit division mode, the pedestrian model in the area only searches for exits to evacuate within the area, and in the crowd division mode, the pedestrian model is only restricted by the evacuation start time. The zone evacuation module is used to simulate zone evacuation, stage evacuation and zone-to-stage evacuation.

[0080] The specific implementation process is as follows: The zoned evacuation module can simulate more evacuation modes, evacuate the crowd in batches and through exits, and can simulate situations under different evacuation plans.

[0081] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A simulation system for assessing evacuation of personnel in a building, characterized in that: Architectural import module: used to import architectural drawings, including dxf and dwg format files; Pedestrian import module: used to add pedestrians to architectural drawings. Pedestrians have action logic, which includes sight logic and scene logic. Special scene import module: used to mark special scene labels in architectural drawings. Special scene labels include signboards, elevators, stairs, exits and gates. Elevators, stairs and exits are all equipped with attraction values ​​to increase the probability of pedestrian models approaching. The attraction value of the exit is greater than the attraction value of the elevator and greater than the attraction value of the building. Elevators, stairs, exits and gates will detect the number of pedestrians attracted, and the attraction value decreases as the number of pedestrians attracted increases. Elevators are equipped with rated operating time and rated load capacity. The elevator takes away the rated load of pedestrians every rated operating time. Stairs and gates are equipped with movement speed attenuation values. When pedestrians are at stairs and gates, their movement speed will be reduced according to the movement speed attenuation value. The sight logic sets a sight range for pedestrians, and pedestrians will read the distribution of pedestrians and special scene labels within the sight range; The scene logic is that pedestrians will behave differently according to the special scene tags within their field of vision. Specifically, when there is a sign within the field of vision, pedestrians will read the direction indicated by the sign and be attracted to move in the indicated direction. When there is an elevator, staircase, exit or gate within the field of vision, the attraction value of the special scene tag is read, and the probability of approaching the corresponding special scene tag is calculated based on the attraction value of each special scene tag.

2. The simulation system for assessing evacuation of personnel in a building according to claim 1, characterized in that: When importing a multi-story building, the building import module imports the architectural drawings of each floor and marks the connecting openings between floors. After pedestrians touch the connecting openings, they will move to the architectural drawings of the corresponding connecting openings.

3. The simulation system for assessing evacuation of personnel in a building according to claim 2, characterized in that: The pedestrian model movement is simulated using linear acceleration algorithm and steering acceleration algorithm.

4. The simulation system for assessing evacuation of personnel in a building according to claim 3, characterized in that: The pedestrian model is set with random acceleration values ​​and magnitudes ranging from 80% to 120% of the standard value.

5. The simulation system for assessing evacuation of personnel in a building according to claim 4, characterized in that: The action logic also includes crowding logic, which detects the number of pedestrian models within the surrounding area of ​​the pedestrian model and corrects the moving speed of the pedestrian model by 10% to 20% according to the number of pedestrian models.

6. The simulation system for assessing evacuation of personnel in a building according to claim 5, characterized in that: The action logic also includes guidance logic. The guidance logic divides pedestrians into attracted, sub-attracted and unattracted states. When there are no signs, elevators, stairs, exits and gates within the pedestrian's field of vision, the pedestrian will follow the pedestrian in the attracted state. At this time, the pedestrian's state is sub-attracted. When there are no signs, elevators, stairs, exits, gates and pedestrians in the attracted state within the pedestrian's field of vision, the pedestrian will follow the pedestrian in the sub-attracted state. When there are no signs, elevators, stairs, exits, gates, pedestrians in the attracted state and pedestrians in the sub-attracted state within the pedestrian's field of vision, the pedestrian will move in a random non-wall direction.

7. The simulation system for assessing evacuation of personnel in a building according to claim 6, characterized in that: The special scene import module is also used to add, delete or move signs in real time during the simulation.

8. The simulation system for assessing evacuation of personnel in a building according to claim 7, characterized in that: It also includes an evacuation effect evaluation module, which is used to collect the distribution of pedestrians in the building drawings in real time, and output a pedestrian density heat map based on the distribution of pedestrians. The pedestrian density heat map includes the current density, maximum density, regional service level index, evacuation congestion index, and records the time values ​​required for the pedestrian evacuation to reach 50%, 95%, and 100%.

9. The simulation system for assessing evacuation of personnel in a building according to claim 8, characterized in that: Personnel have a pre-action time before responding to the evacuation action, and the pre-action time follows the Weibull distribution: F(x)=(ac[1-exp(-x c )] a-1 ·exp(-x c )x c-1 -loc) / scale Where x is the pre-action time, a and c are two parameters in the Weibull probability density distribution function, loc is the position translation coefficient, and scale is the scaling coefficient.

10. The simulation system for assessing evacuation of personnel in a building according to claim 9, characterized in that: It also includes a zone evacuation module, which is used to divide the building drawings into multiple areas. Each area has a preset evacuation start time. The area division includes an exit division mode and a crowd division mode. In the exit division mode, the pedestrian model in the area only searches for exits to evacuate within the area. In the crowd division mode, the pedestrian model is only restricted by the evacuation start time. The zone evacuation module is used to simulate zone evacuation, stage evacuation and zone-to-stage evacuation.

Images