Personnel evacuation path optimization method under emergency based on optimization algorithm

Abstract

The invention discloses a method for optimizing a personnel evacuation path under an emergency based on an optimization algorithm. The method comprises the following steps: firstly, establishing an indoor emergency evacuation map, generating a grid evacuation environment, and introducing Cubic chaotic mapping into firework population initialization in the grid evacuation environment; calculating the number of explosive fireworks on the basis of the initialized firework population; then, the firework explosion range is calculated based on a Piecewise chaos method; carrying out free variation operation on firework positions; selecting an optimal individual position by adopting a mixed selection strategy combining elite selection and a roulette strategy; and finally, storing the optimal individual position and finally forming an optimal path. According to the method, a free variation operation and a mixed selection strategy are designed, wherein the mixed selection strategy is realized by adopting a roulette strategy and elite selection mixed weighting. Compared with a traditional firework algorithm, the method is more suitable for the evacuation environment in an actual public place, and the evacuation path optimization efficiency and accuracy can be further improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of emergency management, and in particular relates to a method for optimizing the evacuation path of personnel in an emergency event based on an optimization algorithm. Background Art

[0002] Emergency evacuation refers to quickly guiding people to a safe area in the event of an emergency. When an emergency occurs in a public place, it is necessary to quickly guide people to evacuate the area, so accurate and efficient evacuation path optimization is particularly important. In traditional evacuation scenarios, evacuees complete the evacuation by observing evacuation signs and exit instructions. However, this evacuation efficiency is low and the route planning is unreasonable. Therefore, it is necessary to combine more information to study and explore the optimization method of personnel evacuation path based on optimization algorithm.

[0003] With the continuous improvement of optimization algorithm research, new intelligent optimization algorithms are emerging, which provides an opportunity to solve the optimization method of personnel evacuation path under emergency events. Among many intelligent optimization algorithms, the fireworks algorithm has been widely used due to its strong global search ability and parallel search characteristics. However, considering the complexity of actual evacuation scenarios, the traditional fireworks algorithm has problems such as weak local optimization ability, low search efficiency, and long optimization path. Summary of the invention

[0004] In view of the above problems, the present invention proposes a method for optimizing the evacuation path of personnel under emergency events based on an optimization algorithm. First, the present invention introduces the Cubic chaotic map into the initialization of the fireworks algorithm and designs a new initialization formula to make the distribution in the initial space more reasonable and uniform to avoid falling into the local optimum. Secondly, in view of the problem that the traditional fireworks algorithm does not consider the obstacle information, the present invention designs an adaptive factor in the fitness formula of the algorithm to effectively avoid obstacles and improve the accuracy of path optimization under emergency evacuation. Then, the present invention introduces the Piecewise chaos method to calculate the explosion range of fireworks, improves the global search accuracy of the algorithm and avoids falling into the local optimal solution. Finally, in order to avoid the algorithm from falling into the optimal solution of the starting area, the present invention designs a free mutation operation and a hybrid selection strategy, in which the hybrid strategy adopts a roulette strategy and an elite selection hybrid weighted implementation. Compared with the traditional fireworks algorithm, the present invention is more in line with the actual evacuation environment in public places and can further improve the efficiency and accuracy of evacuation path optimization.

[0005] The above purpose is achieved through the following technical solutions:

[0006] A method for optimizing the evacuation path of personnel in an emergency based on an optimization algorithm, the method comprising the following steps:

[0007] Step 1: Establish an indoor emergency evacuation map, generate a grid evacuation environment, and introduce the Cubic chaotic map into the fireworks population initialization in the grid evacuation environment;

[0008] Step 2: Calculate the number of fireworks that explode based on the fireworks population initialized in step 1;

[0009] Step 3: Calculate the explosion range of fireworks based on the Piecewise chaos method;

[0010] Step 4: Perform free variation operation on the position of fireworks;

[0011] Step 5: Use a hybrid selection strategy that combines elite selection and roulette strategy to select the optimal individual position;

[0012] Step 6: Save the optimal individual position x obtained in step 5 i,j , and finally form the optimal path.

[0013] Furthermore, the specific operation of step one includes the following sub-steps:

[0014] Sub-step 1.1, obtaining a two-dimensional plane map of the indoor layout, dividing the map into grids based on the map, and generating a grid evacuation environment;

[0015] Sub-step 1.2, introduce the Cubic chaos map into the fireworks initialization:

[0016]

[0017] where t m represents the mth sequence generated by Cubic chaos; t m-1 represents the m-1th sequence generated by Cubic chaos; m represents the number of sequences; a=2.595 represents the chaos coefficient;

[0018] Sub-step 1.3, fireworks population initialization: According to the sequence generated by the Cubic chaotic map in sub-step 1.2, it is converted into the search space of the fireworks population to generate the initialized fireworks population:

[0019] Y i =L min +t m (L max -L min )

[0020] Among them, Y i represents the i-th initialized fireworks population; L min Indicates the minimum value of the initialized fireworks population; L max Indicates the maximum value of the initialized fireworks population.

[0021] Furthermore, the calculation of the number of exploded fireworks in step 2 is expressed as follows using a fitness function based on an adaptive factor:

[0022]

[0023] in, Indicates that the previous generation of fireworks generates child fireworks x i The horizontal axis of Indicates that the previous generation of fireworks generates child fireworks x i The vertical coordinate of It means that the previous generation of fireworks x is generated i-1 The horizontal axis of It means that the previous generation of fireworks x is generated i-1 The vertical coordinate of f(x i ) represents individual fireworks x i The fitness function value; n represents the fireworks population, and the specific calculation is as follows:

[0024] n=L max -L min

[0025] Based on the above fitness function value, the number of explosion fireworks S generated by the i-th firework is obtained i as follows:

[0026]

[0027] In the formula, x i represents the position of the ith firework; H is the total number of fireworks; f max represents the maximum fitness value of fireworks; δ=5 represents a constant, f(x j ) represents individual fireworks x j The fitness function value of .

[0028] Furthermore, the calculation of the fireworks explosion range based on the Piecewise chaos method in step 3 is expressed by a piecewise function:

[0029]

[0030] Among them, A i represents the explosion amplitude of the i-th firework, In the formula, B a Indicates the maximum explosion direction; f min Indicates the minimum fitness value of fireworks; C i Represents the explosion radius of fireworks; γ takes values ​​in [0,0.5] and is a piecewise control factor used to divide the piecewise function into four parts;

[0031] Based on the above number of fireworks and explosion range, the position update formula of the i-th firework in the z direction is as follows:

[0032]

[0033] In the formula, Indicates the position value of the i-th firework in the z direction, Represents the Gaussian variation value of the fireworks in the k-th dimension, where k is the selected dimension; rand(-1,1) is a uniform random number in the interval [-1,1].

[0034] Furthermore, the strategy for freely varying the position of fireworks described in step 4 is as follows:

[0035] x i,j =x l,j ·(x l,j -x b,j )·rand(0,1)

[0036] Among them, x i,j represents the fireworks position with the best adaptability value, also called the optimal position, x b,j represents the position of the current optimal firework in the jth dimension; x l,j Indicates the position of the current worst firework in the jth dimension; rand(0,1) represents a random number uniformly distributed in the interval [0,1].

[0037] Furthermore, the probability of an individual being selected under the mixed selection strategy described in step 5 is expressed as follows:

[0038]

[0039] Among them, ρ(x i ) represents individual fireworks x i The probability value of being selected, σ represents a constant used to adjust the probability value, f min Represents the minimum fitness value of fireworks; β is a minimum parameter to prevent division by zero.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] (1) To address the problem that uneven initialization distribution is prone to falling into local optimality, the present invention introduces Cubic chaotic mapping into the fireworks algorithm initialization and designs a new initialization formula to make the distribution in the initial space more reasonable and uniform to avoid falling into local optimality.

[0042] (2) The present invention addresses the problem that the traditional fireworks algorithm does not consider obstacle information. The present invention designs an adaptive factor in the fitness formula of the algorithm to effectively avoid obstacles and improve the accuracy of path optimization in emergency evacuation;

[0043] (3) The present invention introduces the Piecewise chaos method to calculate the explosion range of fireworks, thereby improving the global search accuracy of the algorithm and avoiding falling into the local optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 A flowchart for realizing the method of the present invention;

[0045] Figure 2 This is the optimization result of the personnel evacuation path of the traditional fireworks algorithm;

[0046] Figure 3 The optimization result of the personnel evacuation path of the method of the present invention;

[0047] Figure 4 It is a comparison curve of the optimization results of the traditional fireworks algorithm and the method of the present invention. DETAILED DESCRIPTION

[0048] The present invention is further described below in conjunction with the accompanying drawings and specific examples.

[0049] like Figure 1 As shown, the method for optimizing the evacuation path of personnel in an emergency based on an optimization algorithm of the present invention comprises the following steps:

[0050] Step 1: Fireworks initialization based on Cubic chaos map

[0051] ① Establish an indoor emergency evacuation map. Obtain a two-dimensional plane map of the indoor layout, divide it into grids on this basis, and generate a grid evacuation environment;

[0052] ②Cubic chaotic mapping calculation. When the traditional fireworks algorithm performs path optimization in a grid environment, a random function is usually used to generate the initial fireworks population. This method makes the number of initialized fireworks population unevenly distributed and has poor traversability. Based on the above, considering that the Cubic chaotic mapping has better chaotic traversability and better generated sequence uniformity, the present invention introduces the Cubic chaotic mapping into the fireworks initialization.

[0053]

[0054] where t m represents the mth sequence generated by Cubic chaos; t m-1 represents the m-1th sequence generated by Cubic chaos; m represents the number of sequences; a=2.595 represents the chaos coefficient.

[0055] ③ Fireworks population initialization. According to the sequence generated by the above Cubic chaotic map, it is converted into the search space of the fireworks population to generate the initialized fireworks population:

[0056] Y i =L min +t m (L max -L min )

[0057] Among them, Y i represents the i-th initialized fireworks population; L min Indicates the minimum value of the initialized fireworks population; L max Indicates the maximum value of the initialized fireworks population.

[0058] Step 2: Calculation of the number of explosion fireworks based on adaptive factors

[0059] The number of exploded fireworks is calculated based on the above-mentioned initialized fireworks population. Generally, only good fireworks will explode further. Therefore, it is necessary to find the number of fireworks with good quality. In the traditional fireworks algorithm, the selection of the current evacuation path node position is based on the fitness function of the current fireworks and the distance between the starting position as the main evaluation factor. However, in the actual evacuation scenario, the path nodes will change due to obstacles such as surrounding buildings and facilities. Therefore, the present invention takes the information of obstacles into account on the basis of the traditional fireworks algorithm. The specific implementation is: the present invention adds an adaptive factor to the fitness formula of the traditional fireworks algorithm, the purpose is to remove the fireworks that fall on obstacles. The fitness function based on the adaptive factor is expressed as follows:

[0060]

[0061] in, Indicates that the previous generation of fireworks generates child fireworks x i The horizontal axis of Indicates that the previous generation of fireworks generates child fireworks x i The vertical coordinate of It means that the previous generation of fireworks x is generated i-1 The horizontal axis of It means that the previous generation of fireworks x is generated i-1 The vertical coordinate of f(x i ) represents individual fireworks x i The fitness function value; n represents the fireworks population, which is related to the number of initialized fireworks:

[0062] n=L max -L min

[0063] The fitness function value of the algorithm determines the number of fireworks that explode. Based on the above fitness function value, the number of fireworks that explode when the ith firework is generated can be obtained as follows:

[0064]

[0065] In the formula, x i represents the position of the ith firework; H is the total number of fireworks; f max represents the maximum fitness value of fireworks; δ=5 represents a constant, f(x j ) represents individual fireworks x j The fitness function value of .

[0066] Step 3: Calculation of fireworks explosion range based on Piecewise chaos method

[0067] Calculate the explosion range of fireworks. The present invention introduces a piecewise chaos method to calculate the explosion range of fireworks, improve the global search accuracy of the algorithm and avoid falling into a local optimal solution.

[0068] The Piecewise chaos method is used to generate the explosion radius, specifically:

[0069]

[0070] Among them, A i represents the explosion amplitude of the i-th firework, B a Indicates the maximum explosion direction; f min Indicates the minimum fitness value of fireworks. C i Indicates the explosion radius of fireworks; γ takes values ​​in [0,0.5] and is a piecewise control factor used to divide the four-part function of the piecewise function. The present invention selects γ = 0.3. The chaotic orbit state range is (0,1).

[0071] Based on the above number of fireworks and explosion range, the position update formula in the direction of the i-th firework is as follows:

[0072]

[0073] In the formula, Indicates the position value of the i-th firework in the z direction, Represents the Gaussian variation value of the fireworks in the k-th dimension, where k is the selected dimension; rand(-1,1) is a uniform random number in the interval [-1,1].

[0074] Step 4: Free mutation operation

[0075] Based on the above, the present invention proposes a free mutation strategy. Considering that the Gaussian mutation method used in the traditional fireworks algorithm is prone to overlap, the present invention provides a free mutation strategy as follows:

[0076] x i,j =x l,j ·(x l,j -x b,j)·rand(0,1)

[0077] Among them, x b,j represents the position of the current optimal firework in the jth dimension; x l,j Indicates the position of the current worst firework in the jth dimension; rand(0,1) represents a random number uniformly distributed in the interval [0,1].

[0078] Step 5: Mixed Selection Strategy

[0079] The traditional fireworks algorithm does not take into account the possibility that the offspring may mutate and produce differences in superior offspring. Therefore, in terms of selection strategy, the present invention proposes a selection strategy that combines elite selection and roulette strategy. In this case, the probability of an individual being selected is expressed as follows:

[0080]

[0081] Among them, f mi represents the minimum fitness value of fireworks; σ represents a constant used to adjust the probability value, and β is a minimum parameter to prevent division by zero. This idea can further improve the local search ability of the algorithm while retaining the optimal solution.

[0082] Step 6: Save the waypoints

[0083] Save the optimal path position x i,j , and finally form the optimal path.

[0084] Experimental verification:

[0085] In order to verify the effectiveness of the personnel evacuation path optimization method based on optimization algorithm proposed in the present invention under emergency events, the present invention designed a personnel evacuation simulation experiment. The simulated evacuation environment is an indoor room. Figure 1 That is, the evacuation environment plan, and the evacuation area is gridded. The evacuees are in the upper left corner of the evacuation area, and the evacuation exit is in the lower right corner. Based on the above evacuation environment, the traditional fireworks algorithm and the method of the present invention are used to conduct a simulation experiment on evacuation path optimization.

[0086] Simulation results analysis: Figure 1 The method of the present invention realizes the flow chart. Figure 2 Optimization results of personnel evacuation paths using the traditional fireworks algorithm. Figure 3 The personnel evacuation path optimization result of the method of the present invention. Figure 4 Comparison curve of path optimization results of two methods. Figure 4It can be seen that the traditional fireworks algorithm takes 34.7252 seconds in the entire route optimization process, while the method of the present invention takes 31.7431 seconds, which shows that the present invention is more efficient. In terms of average path length, the traditional fireworks algorithm has a path length of 34.7252 meters after optimization, and the method of the present invention has a path length of 31.7431 meters after optimization, which shows that the path optimization result of the present invention is the best.

Claims

1. A method for optimizing the evacuation path of personnel in an emergency based on an optimization algorithm, characterized in that: The method comprises the following steps: Step 1: Establish an indoor emergency evacuation map, generate a grid evacuation environment, and introduce the Cubic chaotic map into the fireworks population initialization in the grid evacuation environment; Step 2: Calculate the number of fireworks that explode based on the fireworks population initialized in step 1; Step 3: Calculate the explosion range of fireworks based on the Piecewise chaos method; Step 4: Perform free variation operation on the position of fireworks; Step 5: Use a hybrid selection strategy that combines elite selection and roulette strategy to select the optimal individual position; Step 6: Save the optimal individual position x obtained in step 5 i,j , and finally form the optimal path.

2. The method for optimizing the evacuation path of personnel in an emergency based on an optimization algorithm according to claim 1, characterized in that: The specific operations of step one include the following sub-steps: Sub-step 1.1, obtaining a two-dimensional plane map of the indoor layout, dividing the map into grids based on the map, and generating a grid evacuation environment; Sub-step 1.2, introduce the Cubic chaos map into the fireworks initialization: where t m represents the mth sequence generated by Cubic chaos; t m-1 represents the m-1th sequence generated by Cubic chaos; m represents the number of sequences; a=2.595 represents the chaos coefficient; Sub-step 1.3, fireworks population initialization: According to the sequence generated by the Cubic chaotic map in sub-step 1.2, it is converted into the search space of the fireworks population to generate the initialized fireworks population: Y i =L min +t m (L max -L min ) Among them, Y i represents the i-th initialized fireworks population; L min Indicates the minimum value of the initialized fireworks population; L max Indicates the maximum value of the initialized fireworks population.

3. The method for optimizing the evacuation path of personnel in an emergency based on an optimization algorithm according to claim 2, characterized in that: The calculation of the number of explosion fireworks in step 2 is expressed as follows using a fitness function based on an adaptive factor: in, Indicates that the previous generation of fireworks generates child fireworks x i The horizontal axis of Indicates that the previous generation of fireworks generates child fireworks x i The vertical coordinate of It means that the previous generation of fireworks x is generated i-1 The horizontal axis of It means that the previous generation of fireworks x is generated i-1 The vertical coordinate of f(x i ) represents individual fireworks x i The fitness function value; n represents the fireworks population, and the specific calculation is as follows: n=L max -L min Based on the above fitness function value, the number of explosion fireworks S generated by the i-th firework is obtained i as follows: In the formula, x i represents the position of the ith firework; H is the total number of fireworks; f max represents the maximum fitness value of fireworks; δ=5 represents a constant, f(x j ) represents individual fireworks x j The fitness function value of .

4. The method for optimizing the evacuation path of personnel in an emergency based on an optimization algorithm according to claim 3, characterized in that: The calculation of the fireworks explosion range based on the Piecewise chaos method in step 3 is expressed by a piecewise function: Among them, A i represents the explosion amplitude of the i-th firework, In the formula, B a Indicates the maximum explosion direction; f min Indicates the minimum fitness value of fireworks; C i Represents the explosion radius of fireworks; γ takes values ​​in [0,0.5] and is a piecewise control factor used to divide the piecewise function into four parts; Based on the above number of fireworks and explosion range, the position update formula of the i-th firework in the z direction is as follows: In the formula, Indicates the position value of the i-th firework in the z direction, Represents the Gaussian variation value of the fireworks in the k-th dimension, where k is the selected dimension; rand(-1,1) is a uniform random number in the interval [-1,1].

5. The method for optimizing the evacuation path of personnel in an emergency based on an optimization algorithm according to claim 4, characterized in that: The strategy for free variation of fireworks position described in step 4 is as follows: x i,j =x l,j ·(x l,j -x b,j )·rand(0,1) Among them, x i,j represents the fireworks position with the best adaptability value, also called the optimal position, x b,j represents the position of the current optimal firework in the jth dimension; x l,j Indicates the position of the current worst firework in the jth dimension; rand(0,1) represents a random number uniformly distributed in the interval [0,1].

6. The method for optimizing the evacuation path of personnel in an emergency based on an optimization algorithm according to claim 5, characterized in that: The probability of an individual being selected under the mixed selection strategy described in step 5 is expressed as follows: Among them, ρ(x i ) represents individual fireworks x i The probability value of being selected, σ represents a constant used to adjust the probability value, f min Represents the minimum fitness value of fireworks; β is a minimum parameter to prevent division by zero.

Images