Building fire crowd evacuation intelligent path planning system and method based on extended ground field and multi-target SBOA

Abstract

The invention relates to a building fire crowd evacuation intelligent path planning system and method based on an extended ground field and a multi-target SBOA. The invention relates to the technical field of building crowd evacuation simulation, and the method comprises the steps: obtaining a cellular set based on a space structure of a building; building pedestrian basic information is collected based on a questionnaire and video observation mode, the pedestrian panic degree is judged according to the pedestrian basic information, and a panic field is obtained; based on an FDS smoke diffusion model, a dangerous field is obtained according to the real-time smoke concentration of the fire; based on a classic ground CA model, the panic field and the dangerous field, an extended ground CA model is established, and a current pedestrian target evacuation path is obtained; and based on a multi-target SBOA optimization algorithm, according to the current pedestrian target evacuation path, optimizing the evacuation path, and generating an optimal evacuation path. According to the method, the relation between evacuation groups is actually reflected, a more reasonable evacuation path is planned, the evacuation time is shortened, the safety is improved, casualties are reduced, and the evacuation efficiency is improved.

Description

Technical Field

[0001] The invention relates to the technical field of building crowd evacuation simulation technology, and is an intelligent path planning system and method for building fire crowd evacuation based on extended ground field and multi-objective SBOA. Background Art

[0002] With the increase of group activities and the frequent occurrence of accidents, crowded places often have major hidden dangers. Therefore, studying the behavioral laws of crowd evacuation through simulation technology has very important practical significance for the evacuation of crowds in the event of a disaster. On the one hand, the evacuation model is the basis for studying the evolution of crowd movement and crowd behavior. On the other hand, it can provide a theoretical basis for risk assessment, emergency plans and emergency decisions in safety management. At present, the CA model has attracted the interest of many people because of its more accurate and natural description of movement, and has become one of the most widely used evacuation models. However, the current CA model does not fully consider the impact of people's panic behavior and smoke concentration, which makes the simulation results of indoor evacuation not realistic enough, and the evacuation path selection of people is not intelligent enough to restore reality.

[0003] The SBOA optimization algorithm is a swarm intelligence optimization algorithm proposed by Fu Youfa in 2024. Its basic principle is to simulate the survival ability of secretary birds and has good global search capabilities. Its bionic principle is: the exploration phase simulates the process of secretary birds hunting prey; the utilization phase simulates the escape of secretary birds from predators. In this phase, the secretary birds observe the environment and choose the most appropriate way to reach a safe shelter. These two phases are iteratively repeated under the constraints of the termination criteria to find the optimal solution to the optimization problem.

[0004] The most disastrous group behavior of evacuated people is the stampede caused by panic, which often leads to a large number of casualties. For this reason, researchers have conducted a large number of evacuation simulations in the hope of finding out the characteristics of panic transmission. However, simulations are pre-organized events and cannot fully reflect the panic of the evacuated people at the disaster site. In addition, panic transmission is complex, time-varying, and cannot be replicated. Therefore, it is necessary to understand the impact of panic on pedestrians, establish a panic transmission model, study the characteristics of panic transmission, and consider the impact of panic on pedestrians' choice of paths during the evacuation process, so as to provide methodological support for preventing overcrowding and stampedes. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention provides an intelligent path planning system and method for crowd evacuation in building fires based on extended field and multi-objective SBOA, aiming to solve the problem that in the prior art, only evacuation timeliness is considered when dynamically planning evacuation paths, but evacuation safety and the impact of crowd psychology on evacuation are not taken into account.

[0006] The present invention provides the following technical solutions:

[0007] A system and method for intelligent path planning of crowd evacuation in building fire based on extended ground field and multi-objective SBOA, the method comprising the following steps:

[0008] Step 1: Based on the spatial structure of the building, obtain the cell set, collect basic information of pedestrians in the building through questionnaires and video observations, and randomly initialize the pedestrian positions;

[0009] Step 2: Determine the panic level of pedestrians based on their basic information and establish a panic scene; establish a danger scene based on the FDS smoke diffusion model and the real-time smoke concentration of the fire;

[0010] Step 3: Based on the classic ground-field CA model, panic field and danger field, an extended ground-field CA model is established for macroscopic path planning to obtain the current pedestrian target evacuation path;

[0011] Step 4: Based on the multi-objective SBOA optimization algorithm, optimize the evacuation path according to the current pedestrian target evacuation path and generate the optimal evacuation path.

[0012] Preferably, the panic field includes: individual comprehensive panic factor, panic propagation factor and panic factor affected by smoke. The individual comprehensive panic factor is determined according to basic information of pedestrians, the panic propagation factor is determined according to pedestrian psychology and the law of pedestrian panic emotion propagation, and the panic factor affected by smoke is determined according to smoke concentration.

[0013] Basic information of pedestrians includes their education, age, social experience, number of drills they have participated in, and their knowledge of evacuation;

[0014] The danger field is established according to the smoke concentration. Specifically, the danger field describes that pedestrians prefer to stay away from the danger source, and the danger field is set proportional to the distance from the danger source;

[0015] Danger sources include fire sources, obstacles, and cellular grids whose smoke concentration exceeds the maximum smoke concentration that the human body can withstand.

[0016] Preferably, the panic field describes the effect of panic emotion on pedestrian evacuation. The panic field Pa fg It includes: the panic factor of pedestrians themselves, the factors affected by the panic emotions of people around them, and the factors affecting the panic emotions of smoke concentration. The panic scene Pa is expressed by the following formula: fg :

[0017]

[0018] Among them, A i is the individual comprehensive panic factor, B ji is the panic spreading factor, Ci is the panic factor affected by smoke concentration. a , k b , k c They represent the adjustment coefficients respectively, and r is the pedestrian viewing range.

[0019] Preferably, based on the basic information of building pedestrians, an individual comprehensive panic factor A is set. i , expressed by the following formula:

[0020]

[0021] Among them, x 1 represents the number of pedestrians participating in the exercise, x 2 represents the coefficient of pedestrians’ understanding of evacuation knowledge; 1 represents the age of the pedestrian, y 2 represents the pedestrian's social experience coefficient; a and b represent weight coefficients, a∈[0, 1], b∈[-0.1, 0]; γ represents the pedestrian's familiarity coefficient with the environment.

[0022] Preferably, the panic spreading factor is based on pedestrian psychology and the law of pedestrian panic spreading, and the following panic spreading factors are determined: panic spreading factor B ji The formula is:

[0023]

[0024] Among them, D ij represents the distance between pedestrians i and j, E ji It is expressed as the emotional influence attribute, which indicates the influence of pedestrian j on pedestrian i; λ indicates the intensity of personal emotional expression; ω indicates the coefficient of personal self-judgment ability; the coefficient of self-judgment ability is affected by the level of education. The higher the level of education, the stronger the self-judgment ability.

[0025] Preferably, based on pedestrian psychology, pedestrians may panic due to the harmful gases in the smoke and reduced visibility. According to the smoke concentration at a height of 1.4m and the maximum smoke concentration that the human body can withstand, the panic factor C affected by the smoke concentration is obtained. i , the formula is:

[0026]

[0027] Among them, C i t is the smoke concentration at a height of 1.4m, C cr The maximum smoke concentration that the human body can withstand;

[0028] The danger field is set to be proportional to the distance to the danger source, which includes fire sources, obstacles, and cell grids whose smoke concentration exceeds the maximum concentration that the human body can withstand. The danger field Hij The formula is:

[0029]

[0030] Among them, C i ≥C cr is regarded as a danger source, and its coordinate is marked as x 1，2，3 .....

[0031] Preferably, the extended ground field CA model includes a static field, a dynamic field, a panic field and a danger field;

[0032] The extended ground-field CA model calculates the probability of each pedestrian's moving direction. The pedestrian moves to the cell grid with the highest probability of moving direction. The cell grid where the pedestrian is located determines the cell grid to move next based on the transfer in nine directions, and then plans the pedestrian evacuation path.

[0033] In one time step, the probability P of the pedestrian's moving direction f，g The formula is:

[0034] P f,g =Uexp(k S S fg +k D D fg +k pa Pa fg +k H H fg )(1-η fg )ε fg

[0035] Among them, U represents the normalization factor, k S , k D , k Pa , k H Respectively represent static field weight, dynamic field weight, panic field weight, and danger field weight; k S +k D +k Pa +k H =1;S fg , D fg ,Pa fg , H fg Respectively represent the static field, dynamic field, panic field and danger field at time interval t; η fg Indicates whether the cell is occupied by pedestrians. If the cell is occupied by pedestrians, its value is 1, and if the cell is empty, its value is 0; ε fg Indicates whether the cell is occupied by an obstacle. If the cell is occupied by an obstacle, its value is 0, and if the cell is empty, its value is 1;

[0036] The multi-objective SBOA optimization algorithm establishes a fitness function based on the distance between pedestrians and exits, the distance between pedestrians and dangerous sources, the vertical distance between pedestrians and walls, and the ambient temperature. The solution with the largest fitness value obtained through iteration is the optimal solution, that is, the optimal path for pedestrians. The fitness function is:

[0037]

[0038] Among them, D a Indicates the distance from the pedestrian position to the final evacuation exit, D b Indicates the distance from the pedestrian to the danger source, D w represents the vertical distance from the pedestrian to the wall, T represents the temperature influence coefficient of the pedestrian; m 1 and m 2 Represents the weight coefficient, m 1 、m 2 ∈[0, 1]; r is the pedestrian’s field of view;

[0039] The temperature influence coefficient of pedestrians is as follows:

[0040]

[0041] Among them, T s is the fire scene temperature (℃); T 0 is room temperature (20°C); T e1 The temperature that causes discomfort is set at 30°C; T e2 The temperature for causing damage is set at 60°C; T d The lethal temperature is set to 120°C. When the temperature exceeds 120°C, T=0.

[0042] An intelligent path planning system for crowd evacuation in building fire based on extended ground field and multi-objective SBOA, the system comprising:

[0043] The information acquisition and initialization module is used to obtain basic information about pedestrians in the building through questionnaires and video observations, obtain cell sets based on the spatial structure of the building, and randomly initialize the positions of pedestrians;

[0044] The panic field module is used to obtain the comprehensive panic factor of all pedestrians, the panic propagation factor, the panic factor affected by smoke concentration, and establish a panic field based on the basic information of pedestrians;

[0045] Danger field module, which is used to set the distance between pedestrians and danger sources in proportion to the distance between pedestrians and danger sources. Danger sources include fire sources, obstacles, and cellular grids whose smoke concentration exceeds the maximum concentration that the human body can withstand.

[0046] Model building and path planning module, used to build an extended ground field CA model based on panic field, danger field, and classic ground field CA model; perform macroscopic path planning based on the extended ground field CA model to obtain the target evacuation path;

[0047] The optimization module is used to replace the target evacuation path with the second stage C of the multi-objective SBOA optimization algorithm 1 The optimal solution among the possibilities is found, and the optimal solution is iterated based on the multi-objective SBOA optimization algorithm as the optimal evacuation path.

[0048] A computer-readable storage medium stores a computer program, which is executed by a processor to implement an intelligent path planning method for crowd evacuation in building fires based on an extended field and multi-objective SBOA.

[0049] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, an intelligent path planning method for crowd evacuation in a building fire based on an extended ground field and multi-objective SBOA is implemented.

[0050] The present invention has the following beneficial effects:

[0051] Compared with the prior art, the present invention has the following advantages:

[0052] The present invention obtains a cell set based on the spatial structure of the building; collects basic information of pedestrians in the building based on questionnaires and video observations, and determines the panic level of pedestrians based on the basic information of pedestrians to obtain a panic field; obtains a danger field based on the real-time smoke concentration of the fire based on the FDS smoke diffusion model; establishes an extended ground field CA model based on the classic ground field CA model and the panic field and danger field to obtain the current pedestrian target evacuation path; based on the multi-objective SBOA optimization algorithm, further optimizes the evacuation path based on the current pedestrian target evacuation path to generate the optimal evacuation path. This method can actually reflect the relationship between evacuation groups, plan a more reasonable evacuation path, shorten the evacuation time, improve safety, reduce casualties, and improve evacuation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 Shown is a flow chart of the intelligent path planning method for crowd evacuation during building fires based on the extended ground-field CA model and multi-objective SBOA optimization algorithm of the present invention;

[0055] Figure 2 Shown is a block diagram of the intelligent path planning system for crowd evacuation during building fires based on the extended ground-field CA model and multi-objective SBOA optimization algorithm of the present invention.

[0056] Figure 3 Shown is a flow chart of the SBOA optimization algorithm of the present invention. DETAILED DESCRIPTION

[0057] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. 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.

[0058] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0059] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0060] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0061] The present invention is described in detail below in conjunction with specific embodiments. Specific embodiment one:

[0063] according to Figures 1 to 3 As shown, the specific optimization technical solution adopted by the present invention to solve the above technical problems is: the present invention relates to an intelligent path planning system and method for building fire crowd evacuation based on extended field and multi-objective SBOA.

[0064] An intelligent path planning method for crowd evacuation in a building fire based on an extended ground field and multi-objective SBOA, the method comprising the following steps:

[0065] The method comprises the following steps:

[0066] Step 1: Based on the spatial structure of the building, obtain the cell set, collect basic information of pedestrians in the building through questionnaires and video observations, and randomly initialize the pedestrian positions;

[0067] Step 2: Determine the panic level of pedestrians based on their basic information and establish a panic scene; establish a danger scene based on the FDS smoke diffusion model and the real-time smoke concentration of the fire;

[0068] Step 3: Based on the classic ground-field CA model, panic field and danger field, an extended ground-field CA model is established for macroscopic path planning to obtain the current pedestrian target evacuation path;

[0069] Step 4: Based on the multi-objective SBOA optimization algorithm, optimize the evacuation path according to the current pedestrian target evacuation path and generate the optimal evacuation path. Specific embodiment 2:

[0071] The difference between the second embodiment of the present application and the first embodiment is that:

[0072] The panic scene includes: individual comprehensive panic factor, panic transmission factor and panic factor affected by smoke. The individual comprehensive panic factor and panic transmission factor are determined based on the basic information of pedestrians, and the panic factor affected by smoke is determined based on the smoke concentration.

[0073] Basic information of pedestrians includes their education, age, social experience, number of drills they have participated in, and their knowledge of evacuation;

[0074] The danger field is established according to the smoke concentration. Specifically, the danger field describes that pedestrians prefer to stay away from the danger source, and the danger field is set proportional to the distance from the danger source;

[0075] Danger sources include fire sources, obstacles, and cellular grids whose smoke concentration exceeds the maximum smoke concentration that the human body can withstand. Specific embodiment three:

[0077] The difference between the third embodiment of the present application and the second embodiment is that:

[0078] Panic field describes the impact of panic emotions on pedestrian evacuation. fg It includes: the panic factor of pedestrians themselves, the factors affected by the panic emotions of people around them, and the factors affecting the panic emotions of smoke concentration. The panic scene Pa is expressed by the following formula: fg :

[0079]

[0080] Among them, A i is the individual comprehensive panic factor, B ji is the panic spreading factor, C i is the panic factor affected by smoke concentration. a , k b , k c They represent the adjustment coefficients respectively, and r is the pedestrian viewing range. Specific embodiment four:

[0082] The difference between the fourth embodiment of the present application and the third embodiment is that:

[0083] Based on the basic information of building pedestrians, set the individual comprehensive panic factor A i , expressed by the following formula:

[0084]

[0085] Among them, x 1 represents the number of pedestrians participating in the exercise, x 2 represents the coefficient of pedestrians’ understanding of evacuation knowledge; 1 represents the age of the pedestrian, y 2 represents the pedestrian's social experience coefficient; a and b represent weight coefficients, a∈[0, 1], b∈[-0.1]; γ represents the pedestrian's familiarity coefficient with the environment. Specific embodiment five:

[0087] The difference between the fifth embodiment of the present invention and the fourth embodiment is that:

[0088] Panic propagation factors are based on pedestrian psychology and the law of pedestrian panic transmission. The following panic propagation factors are determined: Panic propagation factor B ji The formula is:

[0089]

[0090] Among them, D ij represents the distance between pedestrians i and j, E ji It is expressed as the emotional influence attribute, which indicates the influence of pedestrian j on pedestrian i; λ indicates the intensity of personal emotional expression; ω indicates the coefficient of personal self-judgment ability; the coefficient of self-judgment ability is affected by the level of education. The higher the level of education, the stronger the self-judgment ability. Specific embodiment six:

[0092] The difference between the sixth embodiment of the present invention and the fifth embodiment is that:

[0093] Based on pedestrian psychology, pedestrians will panic due to the harmful gases in the smoke and the reduced visibility. According to the smoke concentration at a height of 1.4m and the maximum smoke concentration that the human body can withstand, the panic factor C affected by the smoke concentration is obtained. i , the formula is:

[0094]

[0095] Among them, C i t is the smoke concentration at a height of 1.4m, C cr The maximum smoke concentration that the human body can withstand;

[0096] The danger field is set to be proportional to the distance to the danger source, which includes fire sources, obstacles, and cell grids whose smoke concentration exceeds the maximum concentration that the human body can withstand. The danger field H ij The formula is:

[0097]

[0098] Among them, C i ≥C cr is regarded as a danger source, and its coordinate is marked as x 1，2，3 ..... Specific embodiment seven:

[0100] The difference between the seventh embodiment of the present invention and the sixth embodiment is that:

[0101] The extended ground field CA model includes static field, dynamic field, panic field and danger field;

[0102] The extended ground-field CA model calculates the probability of each pedestrian's moving direction. The pedestrian moves to the cell grid with the highest probability of moving direction. The cell grid where the pedestrian is located determines the cell grid to move next based on the transfer in nine directions, and then plans the pedestrian evacuation path.

[0103] In one time step, the probability P of the pedestrian's moving direction f，g The formula is:

[0104] P f,g =Uexp(k S S fg +k D D fg +k pa Pa fg +k H H fg )(1-η fg )ε fg

[0105] Among them, U represents the normalization factor, k S , kD , k Pa , k H Respectively represent static field weight, dynamic field weight, panic field weight, and danger field weight; k S +k D +k Pa +k H =1;S fg , D fg ,Pa fg , H fg Respectively represent the static field, dynamic field, panic field and danger field at time interval t; η fg Indicates whether the cell is occupied by pedestrians. If the cell is occupied by pedestrians, its value is 1, and if the cell is empty, its value is 0; ε fg Indicates whether the cell is occupied by an obstacle. If the cell is occupied by an obstacle, its value is 0, and if the cell is empty, its value is 1;

[0106] The multi-objective SBOA optimization algorithm establishes a fitness function based on the distance between pedestrians and exits, the distance between pedestrians and dangerous sources, the vertical distance between pedestrians and walls, and the ambient temperature. The solution with the largest fitness value obtained through iteration is the optimal solution, that is, the optimal path for pedestrians. The fitness function is:

[0107]

[0108] Among them, D a Indicates the distance from the pedestrian position to the final evacuation exit, D b Indicates the distance from the pedestrian to the danger source, D w represents the vertical distance from the pedestrian to the wall, T represents the temperature influence coefficient of the pedestrian; m 1 and m 2 Represents the weight coefficient, m 1 、m 2 ∈[0, 1]; r is the pedestrian’s field of view;

[0109] The temperature influence coefficient of pedestrians is as follows:

[0110]

[0111] Among them, T s is the fire scene temperature (℃); T 0 is room temperature (20°C); T e1 The temperature that causes discomfort is set at 30°C; T e2 The temperature for causing damage is set at 60°C; T d The lethal temperature is set to 120°C. When the temperature exceeds 120°C, T=0. Specific embodiment eight:

[0113] The difference between the eighth embodiment of the present invention and the seventh embodiment is that:

[0114] The present invention provides an intelligent path planning system for crowd evacuation in building fires based on an extended ground field and multi-objective SBOA, the system comprising:

[0115] The information acquisition and initialization module is used to obtain basic information about pedestrians in the building through questionnaires and video observations, obtain cell sets based on the spatial structure of the building, and randomly initialize the positions of pedestrians;

[0116] The panic field module is used to obtain the comprehensive panic factor of all pedestrians, the panic propagation factor, the panic factor affected by smoke concentration, and establish a panic field based on the basic information of pedestrians;

[0117] Danger field module, which is used to set the distance between pedestrians and danger sources in proportion to the distance between pedestrians and danger sources. Danger sources include fire sources, obstacles, and cellular grids whose smoke concentration exceeds the maximum concentration that the human body can withstand.

[0118] Model building and path planning module, used to build an extended ground field CA model based on panic field, danger field, and classic ground field CA model; perform macroscopic path planning based on the extended ground field CA model to obtain the target evacuation path;

[0119] The optimization module is used to replace the target evacuation path with the second stage C of the multi-objective SBOA optimization algorithm 1 The optimal solution among the possibilities is found, and the optimal solution is iterated based on the multi-objective SBOA optimization algorithm as the optimal evacuation path. Specific embodiment nine:

[0121] The difference between the ninth embodiment of the present invention and the eighth embodiment is that:

[0122] The present invention provides a computer-readable storage medium on which a computer program is stored. The program is executed by a processor to implement an intelligent path planning method for crowd evacuation in building fires based on an extended ground field and multi-objective SBOA. Specific embodiment ten:

[0124] The difference between the tenth embodiment of the present invention and the ninth embodiment is that:

[0125] The present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, an intelligent path planning method for crowd evacuation in a building fire based on an extended ground field and multi-objective SBOA is implemented. Specific embodiment eleven:

[0127] The difference between the eleventh embodiment of the present invention and the tenth embodiment is that:

[0128] The SBOA optimization algorithm is a swarm intelligence optimization algorithm proposed by Fu Youfa in 2024. Its basic principle is to simulate the survival ability of secretary vultures and has good global search capabilities.

[0129] Combination Figure 1 The present invention proposes an intelligent path planning method for crowd evacuation during building fires based on an extended ground-field CA model and a multi-objective SBOA optimization algorithm, the method comprising:

[0130] Step 1: Based on the spatial structure of the building, obtain a cell set; collect basic information of pedestrians in the building through questionnaires and video observations, and initialize the pedestrian positions;

[0131] The step 1 is specifically as follows:

[0132] Step 1.1: The building spatial structure information includes the building plan shape, the total building area, the area of ​​each room in the building, the number of building exits, exit positions, exit widths, the number of building rooms, the number of exits in each room of the building, exit positions, exit widths and obstacle positions in each room of the building.

[0133] Step 1.2: The basic information of building pedestrians includes the pedestrians' education, age, social experience, number of drills participated in, and degree of understanding of evacuation knowledge.

[0134] Step 2: Determine the panic level of pedestrians based on the basic information of the pedestrians and establish a panic field; establish a danger field based on the FDS smoke diffusion model and the real-time fire smoke concentration;

[0135] Step 2.1: The panic field describes the panic of the pedestrian, the impact of the panic of the people around him, and the impact of the smoke concentration on the panic. The panic field Pa can be described by the following formula: fg :

[0136] (1)

[0137] Among them, A i is the individual comprehensive panic factor, B ji is the panic spreading factor, C i is the panic factor affected by smoke concentration. a , k b , k c They represent the adjustment coefficients respectively, and r is the pedestrian viewing range.

[0138] Step 2.2: Based on the basic information of the building pedestrians, set the individual comprehensive panic factor A i ;

[0139] (2)

[0140] Among them, x 1 represents the number of pedestrians participating in the exercise, x 2 represents the coefficient of pedestrians’ understanding of evacuation knowledge; 1 represents the age of the pedestrian, y 2 represents the pedestrian's social experience coefficient; a and b represent weight coefficients, a∈[0, 1], b∈[-0.1, 0]; γ is the pedestrian's familiarity coefficient with the environment.

[0141] Step 2.3: Based on pedestrian psychology and the law of panic spread, the following panic spread factor is set: ji The formula is:

[0142] (3)

[0143] Among them, D ij represents the distance between pedestrians i and j, E ji It is represented as the emotional influence attribute, which indicates the influence of pedestrian j on pedestrian i; λ indicates the intensity of personal emotional expression; ω indicates the coefficient of personal self-judgment ability; the self-judgment ability coefficient is affected by the education level, and the higher the education level, the stronger the self-judgment ability;

[0144] Step 2.4: Based on pedestrian psychology, pedestrians will panic due to the influence of harmful gases in smoke and reduced visibility. According to the smoke concentration at a height of 1.4m and the maximum smoke concentration that the human body can withstand, the panic factor C affected by the smoke concentration is obtained. i , the formula is:

[0145] (4)

[0146] Among them, C i t is the smoke concentration at a height of 1.4m, C cr It is the maximum smoke concentration that the human body can withstand.

[0147] Step 2.5: The danger field describes that pedestrians prefer to stay away from danger, and the space area away from danger is more attractive. Therefore, the danger field is set to be proportional to the distance from the danger source. The danger source includes fire source, obstacles, and cell grids whose smoke concentration exceeds the maximum concentration that the human body can withstand. The danger field H ij The formula is:

[0148] (5)

[0149] Among them, C i ≥C cr is regarded as a danger source, and its coordinate is marked as x 1，2，3 .....

[0150] Step 3: According to the classic ground-field CA model, based on the panic field and the danger field, an extended ground-field CA model is established, and the extended ground-field CA model is used to perform macroscopic path planning to obtain the target evacuation path;

[0151] The step 3 is specifically as follows:

[0152] Step 3.1: The present invention proposes an extended ground field CA model to simulate evacuation dynamics. Panic field and danger field are added based on the classic ground field CA model, and the classic ground field CA model includes a static field and a dynamic field. The probability of each pedestrian's moving direction is calculated, and the pedestrian moves to the cell with the highest probability of moving direction. The calculation of the probability of each pedestrian's moving direction takes the cell where the pedestrian is located as an example, and it can determine the next moving cell based on the transfer in 9 directions. In one time step, the probability of the pedestrian's moving direction is:

[0153] (6) P f,g =Uexp(k S S fg +k D D fg +k pa Pa fg +k H H fg )(1-η fg )ε fg

[0154] Among them, U represents the normalization factor, k S , k D , k Pa , k H Respectively represent static field weight, dynamic field weight, panic field weight, and danger field weight; k S +k D +k Pa +k H =1;S fg , D fg ,Pa fg , H fg They represent the static field, dynamic field, panic field and danger field at time interval t respectively. fg Indicates whether the cell is occupied by pedestrians. If the cell is occupied by pedestrians, its value is 1, and if the cell is empty, its value is 0; ε fg Indicates whether the cell is occupied by an obstacle. If the cell is occupied by an obstacle, its value is 0, and if the cell is empty, its value is 1;

[0155] Step 3.2: In the process of calculating the probability of the pedestrian's moving direction, set the static field S fg , Dynamic field D fg , Panic scenefg 、Hazardous areas fg Specifically, the static field S fg Describes the shortest distance from the cell (f, g) to the nearest exit. It remains unchanged during the simulation. The shorter the distance from the cell (f, g) to the exit, the smaller the static field S fg The larger the static field S fg The formula is:

[0156] (7)

[0157] There are n exits in the building, d* fg represents the shortest distance from cell (f, g) to the nearest exit.

[0158] Step 3.3: The dynamic field D fg It describes the behavior of pedestrians following others during movement, representing the virtual traces left by pedestrians, which corresponds to the phenomenon that people tend to follow the traces of others, that is, the herd effect. In the initial state, the dynamic field D of the initial state of all cells fg = 0. When the simulation starts, in each time step, the dynamic field D of the original cell after the pedestrian moves fg = 1. Then the dynamic field D fg = 1 decays to D with probability α fg = 0, and the dynamic field D fg Diffusion to neighboring cells with probability β. The following formula can be used to describe the dynamic field D fg :

[0159] Whenever a pedestrian moves from one cell to an adjacent cell, the dynamic field value of the cell occupied by the pedestrian is increased by 1:

[0160] (8) D fg =D fg t +1

[0161] The dynamic field D fg The diffusion and attenuation parameters of the value are controlled by α and β respectively:

[0162] (9)

[0163] Where α represents the attenuation probability and β represents the diffusion probability.

[0164] Step 4: Use the multi-objective SBOA optimization algorithm to further optimize the target evacuation path and generate the optimal evacuation path.

[0165] The step 4 is specifically as follows:

[0166] Step 4.1: The target evacuation path x output by the extended ground field CA model is optimized using the multi-objective SBOA optimization algorithm. aim Further optimization is performed, specifically, the target evacuation path x aim Replace the second stage C of the multi-objective SBOA optimization algorithm 1 The best possible solution x best , let the candidate solutions evacuate around the target path x aim Iteration, the multi-objective SBOA algorithm first stage, second stage C 2 The possibility is still iterated according to the original algorithm position update rules;

[0167] Step 4.2: The multi-objective SBOA optimization algorithm establishes a fitness function based on the distance between the pedestrian and the exit, the distance between the pedestrian and the danger source, the vertical distance between the pedestrian and the wall, and the ambient temperature. The iterative fitness value is the optimal solution, that is, the optimal path for the pedestrian. The fitness function is:

[0168] (10)

[0169] Among them, D a Indicates the distance from the pedestrian position to the final evacuation exit, D b Indicates the distance from the pedestrian to the danger source, D w represents the vertical distance from the pedestrian to the wall, T represents the temperature influence coefficient of the pedestrian; m 1 and m 2 Represents the weight coefficient, m 1 、m 2 ∈[0, 1]; r is the pedestrian’s field of view.

[0170] Furthermore, the pedestrian temperature influence coefficient is specifically:

[0171] (11)

[0172] Among them, T s is the fire scene temperature (℃); T 0 is room temperature (20°C); T e1 is the temperature causing discomfort (set to 30°C); T e2 is the temperature causing damage (set to 60℃); T d is the lethal temperature (set to 120°C). When the temperature exceeds 120°C, T=0.

[0173] Step 4.3: The algorithm iteration process is specifically as follows:

[0174] The SBOA optimization algorithm simulates the behavior of secretary birds in hunting prey and avoiding predators. It is divided into two stages, the exploration stage and the utilization stage, and optimizes the candidate solutions respectively. The steps are as follows:

[0175] Step 4.3.1: Randomly initialize candidate solutions, the formula is:

[0176] (12)X s,q =lb q +r×(ub q -lb j ),s=1,2,...,N,q=1,2,...,Dim

[0177] Step 4.3.2: The exploration phase includes a phase of simulating a secretary bird searching for prey, a phase of the secretary bird surrounding the prey to consume the prey's physical strength, and a phase of the secretary bird attacking the prey.

[0178] Step 4.3.2.1: During the prey-seeking phase of the secretary bird, the position update formula is expressed as follows:

[0179] (13)

[0180] (14)

[0181] Among them, t represents the current number of iterations, T represents the maximum number of iterations, and X s new,P1 represents the new state of the s-th secretary bird in the first stage, x random_1 and x random_2 is a random candidate solution for the first stage iteration. R 1 Represents a randomly generated array of dimension 1×Dim in the interval [0, 1], where Dim is the dimension of the solution space. s，q new,P1 represents the value of its qth dimension, F s new,P1 Represents the fitness value of its objective function.

[0182] Step 4.3.2.2: During the prey consumption phase of the secretary bird, the position update formula is expressed as follows:

[0183] (15) RB = random(1, Dim)

[0184] (16)

[0185] (17) Here the formula x i Changed to x s (Changed)

[0186] Among them, random(1, Dim) represents an array of dimension 1×Dim randomly generated from a standard normal distribution (mean value is 0 and standard deviation is 1), x best Indicates the current optimal value.

[0187] Step 4.3.2.2: During the stage of the secretary bird attacking its prey, the position update formula is expressed as follows:

[0188] (18)

[0189] (19) Here the formula x i Changed to x s (Changed)

[0190] In order to improve the optimization accuracy of the algorithm, weighted Levy flight is used, denoted as "RL":

[0191] (20)RL=0.5×Levy(Dim)

[0192] Where Levy(Dim) represents the Levy flight distribution function. The calculation method is as follows:

[0193] (twenty one)

[0194] Where s is a fixed constant of 0.01 and η is a fixed constant of 1.5. u and v are random numbers in the interval [0, 1]. The formula for σ is as follows:

[0195] (twenty two)

[0196] Wherein, Γ represents the gamma function, and the value of η is 1.5.

[0197] Step 4.3.3: The utilization stage includes two possibilities, namely, the secretary bird uses the environment to camouflage and the secretary bird flies away or escapes, and the two possibilities are equally likely events, respectively represented by C 1 , C 2 Indicates that the position update formula is as follows:

[0198] (twenty three)

[0199] (twenty four)

[0200] Among them, x aim represents the target evacuation path obtained by the above extended ground field CA model, r = 0.5, R2 represents an array of dimension (1×Dim) randomly generated from a normal distribution, x random represents the random candidate solution of the current iteration, and K represents a random selection of an integer 1 or 2, which can be calculated as follows:

[0201] (25) K = round(1 + rand(1, 1))

[0202] Among them, rand(1, 1) means randomly generating a random number between (0, 1).

[0203] In summary, the flowchart of SBOA is as follows: Figure 3 As shown, based on the iteration of this process, the calculation is terminated when the algorithm reaches the maximum number of iterations; specifically, after using the SBOA optimization algorithm for macroscopic path planning, the current optimal value is obtained. If the current optimal value is better than the optimal value of the previous iteration, an update operation is performed, otherwise no update operation is performed, and the iteration operation is continued until the conditions are met, and finally the global optimal value and the best fitness value are obtained, which can be used as the global evacuation path.

[0204] The present invention also proposes an intelligent path planning system for crowd evacuation during building fires based on an extended ground-field CA model and a multi-objective SBOA optimization algorithm, the system comprising:

[0205] The information acquisition and initialization module is used to obtain basic information of pedestrians in the building through questionnaires and video observation, obtain cell sets based on the spatial structure of the building, and randomly initialize the positions of the pedestrians;

[0206] A panic field module is used to obtain the comprehensive panic factor of all pedestrians, the panic propagation factor, the panic factor affected by smoke concentration, and establish a panic field based on the basic information of pedestrians;

[0207] A danger field module is used for setting the danger field in direct proportion to the distance between pedestrians and danger sources based on the distance between the danger source and pedestrians. The danger sources include fire sources, obstacles, and cellular grids whose smoke concentration exceeds the maximum concentration that the human body can withstand.

[0208] Model building and path planning module: used to build an extended ground field CA model based on panic field, danger field and classic ground field CA model; perform macroscopic path planning based on the extended ground field CA model to obtain the target evacuation path;

[0209] Algorithm optimization module, used to replace the target evacuation path with the second stage C of the multi-objective SBOA optimization algorithm 1 The optimal solution among the possibilities is obtained, and the optimal solution is iterated based on the multi-objective SBOA optimization algorithm as the optimal evacuation path.

[0210] The above is only a preferred implementation of a system and method for intelligent path planning for crowd evacuation in building fires based on an extended ground field and multi-target SBOA. The protection scope of a system and method for intelligent path planning for crowd evacuation in building fires based on an extended ground field and multi-target SBOA is not limited to the above embodiments. All technical solutions under this idea belong to the protection scope of the present invention. It should be pointed out that for those skilled in the art, several improvements and changes without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. An intelligent path planning method for crowd evacuation in building fire based on extended ground field and multi-objective SBOA, characterized by: The method comprises the following steps: Step 1: Based on the spatial structure of the building, obtain the cell set, collect basic information of pedestrians in the building through questionnaires and video observations, and randomly initialize the pedestrian positions; Step 2: Determine the panic level of pedestrians based on their basic information and establish a panic scene; establish a danger scene based on the FDS smoke diffusion model and the real-time smoke concentration of the fire; Step 3: Based on the classic ground-field CA model, panic field and danger field, an extended ground-field CA model is established for macroscopic path planning to obtain the current pedestrian target evacuation path; Step 4: Based on the multi-objective SBOA optimization algorithm, optimize the evacuation path according to the current pedestrian target evacuation path and generate the optimal evacuation path.

2. The method according to claim 1, characterized in that: The panic field includes: individual comprehensive panic factor, panic transmission factor and panic factor affected by smoke. The individual comprehensive panic factor is determined based on the basic information of pedestrians, the panic transmission factor is determined based on pedestrian psychology and the law of pedestrian panic emotion transmission, and the panic factor affected by smoke is determined based on smoke concentration. Basic information of pedestrians includes their education, age, social experience, number of drills they have participated in, and their knowledge of evacuation; The danger field is established according to the smoke concentration. Specifically, the danger field describes that pedestrians prefer to stay away from the danger source, and the danger field is set proportional to the distance from the danger source; Danger sources include fire sources, obstacles, and cellular grids whose smoke concentration exceeds the maximum smoke concentration that the human body can withstand.

3. The method according to claim 2, characterized in that: Panic field describes the impact of panic emotions on pedestrian evacuation. fg It includes: the panic factor of pedestrians themselves, the factors affected by the panic emotions of people around them, and the factors affecting the panic emotions of smoke concentration. The panic scene Pa is expressed by the following formula: fg : Among them, A i is the individual comprehensive panic factor, B ji is the panic spreading factor, C i is the panic factor affected by smoke concentration, k a , k b , k c They represent the adjustment coefficients respectively, and r is the pedestrian viewing range.

4. The method according to claim 1, characterized in that: Based on the basic information of building pedestrians, set the individual comprehensive panic factor A i , expressed by the following formula: Among them, x1 represents the number of times the pedestrian participated in the exercise, x2 represents the coefficient of the pedestrian's understanding of evacuation knowledge; y1 represents the pedestrian's age, y2 represents the pedestrian's social experience coefficient; a and b represent weight coefficients, a∈[0,1], b∈[-0.1,0]; γ represents the pedestrian's familiarity with the environment coefficient.

5. The method according to claim 4, characterized in that: Panic propagation factors are based on pedestrian psychology and the law of pedestrian panic propagation. The following panic propagation factors are determined: Panic propagation factor B ji The formula is: Among them, D ij represents the distance between pedestrians i and j, E ji It is expressed as the emotional influence attribute, which indicates the influence of pedestrian j on pedestrian i; λ indicates the intensity of personal emotional expression; ω indicates the coefficient of personal self-judgment ability; the coefficient of self-judgment ability is affected by the level of education. The higher the level of education, the stronger the self-judgment ability.

6. The method according to claim 1, characterized in that: Based on pedestrian psychology, pedestrians will panic due to the harmful gases in the smoke and the reduced visibility. According to the smoke concentration at a height of 1.4m and the maximum smoke concentration that the human body can withstand, the panic factor C affected by the smoke concentration is obtained. i , the formula is: Among them, C i t is the smoke concentration at a height of 1.4m, C cr The maximum smoke concentration that the human body can withstand; The danger field is set to be proportional to the distance to the danger source, which includes fire sources, obstacles, and cell grids whose smoke concentration exceeds the maximum concentration that the human body can withstand. The danger field H ij The formula is: Among them, C i ≥C cr is regarded as a danger source, and its coordinate is marked as x 1，2，3 .....

7. The method according to claim 6, characterized in that: The extended ground field CA model includes static field, dynamic field, panic field and danger field; The extended ground-field CA model calculates the probability of each pedestrian's moving direction. The pedestrian moves to the cell grid with the highest probability of moving direction. The cell grid where the pedestrian is located determines the cell grid to move next based on the transfer in nine directions, and then plans the pedestrian evacuation path. In one time step, the probability P of the pedestrian's moving direction f，g The formula is: P f,g =Uexp(k S S fg +k D D fg +k pa Dad fg +k H H fg )(1-η fg )ε fg Among them, U represents the normalization factor, k S , k D , k Pa , k H Respectively represent static field weight, dynamic field weight, panic field weight, and danger field weight; k S +k D +k Pa +k H =1;S fg , D fg ,Pa fg , H fg Respectively represent the static field, dynamic field, panic field and danger field at time interval t; η fg Indicates whether the cell is occupied by pedestrians. If the cell is occupied by pedestrians, its value is 1, and if the cell is empty, its value is 0; ε fg Indicates whether the cell is occupied by an obstacle. If the cell is occupied by an obstacle, its value is 0, and if the cell is empty, its value is 1; The multi-objective SBOA optimization algorithm establishes a fitness function based on the distance between pedestrians and exits, the distance between pedestrians and dangerous sources, the vertical distance between pedestrians and walls, and the ambient temperature. The solution with the largest fitness value obtained through iteration is the optimal solution, that is, the optimal path for pedestrians. The fitness function is: Among them, D a Indicates the distance from the pedestrian position to the final evacuation exit, D b Indicates the distance from the pedestrian to the danger source, D w represents the vertical distance from the pedestrian to the wall, T represents the temperature influence coefficient of the pedestrian; m1 and m2 represent weight coefficients, m1, m2∈[0,1]; r is the pedestrian's field of view; The temperature influence coefficient of pedestrians is as follows: Among them, T s is the fire scene temperature; T0 is the room temperature 20℃; T e1 The temperature that causes discomfort is set at 30°C; T e2 The temperature for causing damage is set at 60°C; T d The lethal temperature is set to 120°C. When the temperature exceeds 120°C, T=0.

8. An intelligent path planning system for crowd evacuation in building fires based on extended ground field and multi-objective SBOA, characterized by: The system comprises: The information acquisition and initialization module is used to obtain basic information about pedestrians in the building through questionnaires and video observations, obtain cell sets based on the spatial structure of the building, and randomly initialize the positions of pedestrians; The panic field module is used to obtain the comprehensive panic factor of all pedestrians, the panic propagation factor, the panic factor affected by smoke concentration, and establish a panic field based on the basic information of pedestrians; Danger field module, which is used to set the distance between pedestrians and danger sources in proportion to the distance between pedestrians and danger sources. Danger sources include fire sources, obstacles, and cellular grids whose smoke concentration exceeds the maximum concentration that the human body can withstand. Model building and path planning module, used to build an extended ground field CA model based on panic field, danger field, and classic ground field CA model; perform macroscopic path planning based on the extended ground field CA model to obtain the target evacuation path; The optimization module is used to replace the target evacuation path with the optimal solution in the second stage C1 possibility of the multi-objective SBOA optimization algorithm, and iterate the optimal solution based on the multi-objective SBOA optimization algorithm as the optimal evacuation path.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: The program is executed by a processor to implement the method according to claims 1-7.

10. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method of claims 1-7 is implemented.

Images