Ship Fire Evacuation Route Planning Method Based on Ant Colony Algorithm

Abstract

A ship fire evacuation path planning method based on the ant colony algorithm, which belongs to the field of crowd emergency evacuation technology. The purpose of the present invention is to solve the problems that the traditional path planning method is prone to falling into local optimal solutions, does not consider the personnel density and environmental complexity, resulting in low evacuation efficiency, and the path search takes a long time. The present invention generates a ship path network according to the nodes and paths in the ship; locates the positions of each person on the ship, and then uses the positions of the safe points in the ship path network as the evacuation end points. The evacuation speed is calculated according to the position states and environmental states of each person on the ship, and then the evacuation speed is used as the heuristic function of the ant colony algorithm, and evacuation path planning is carried out based on the heuristic function to obtain an evacuation path set. The method of the present invention can be applied to the field of crowd emergency evacuation technology.

Description

Technical Field

[0001] The present invention belongs to the technical field of crowd emergency evacuation, and particularly relates to a ship fire evacuation path planning method based on the ant colony algorithm. Background Technique

[0002] In recent years, waterway sightseeing tourism including large ocean liners, Three Gorges of the Yangtze River scenic cruise ships, etc. has started to become popular. In this case, once natural disasters or man-made emergencies occur, the irrational and uncoordinated behaviors of the crowd may lead to heavy casualties. Many scholars at home and abroad have studied path planning algorithms. Among them, the ant colony algorithm is one of the intelligent bionic algorithms. It is a probabilistic algorithm used to find the optimal path. Essentially, it is a heuristic global optimization algorithm in evolutionary algorithms. Because of its good robustness, self-organization, positive feedback, and distributed computing and other characteristics, it has obvious advantages when facing the complex environment inside the ship; at the same time, its characteristics of "wide application range" and "strong plasticity" enable it to modify the model according to the specific situation of ship evacuation, improving the fit of the algorithm to the problem of personnel evacuation in the complex environment of the ship.

[0003] In addition, the traditional evacuation method in the ship environment relies on wayfinding signs and the experience of staff. However, with the increasing complexity of the cruise ship structure, the traditional evacuation method is prone to select the same evacuation path, resulting in personnel congestion and reducing the evacuation efficiency, which is not suitable for the planning of emergency evacuation routes. Therefore, the ship evacuation path planning method in extreme environments should be continuously updated and improved. However, when using the ant colony algorithm to solve complex path problems, there are problems such as being easily trapped in local optimal solutions, long search time, and insufficient global optimization ability, and the influence of personnel density and environmental complexity on path selection is not considered, resulting in limited improvement in evacuation efficiency. Therefore, in order to better improve the fire personnel evacuation efficiency, it is very necessary to propose a new path planning method. Summary of the Invention

[0004] The purpose of the present invention is to propose a ship fire evacuation path planning method based on the ant colony algorithm to solve the problems that the traditional path planning method is easily trapped in local optimal solutions, does not consider personnel density and environmental complexity resulting in low evacuation efficiency, and the time required for path search is long.

[0005] The technical solution adopted by the present invention to solve the above technical problems is:

[0006] A ship fire evacuation path planning method based on the ant colony algorithm, the method specifically includes the following steps:

[0007] Step 1: Generate a ship path network C based on the nodes and paths in the ship, and denote the generated ship path network as C[R, N], where R represents the set of road segments in the ship path network C, and N represents the set of nodes in the ship path network C;

[0008] Step 2: Locate the positions of each person on the ship, and then use the safe point position F in the ship path network C k as the evacuation end point, calculate the evacuation speed according to the position status and environmental status of each person on the ship, and then use the evacuation speed as the heuristic function of the ant colony algorithm, and perform evacuation path planning based on the heuristic function to obtain an evacuation path set.

[0009] Furthermore, the node set includes cabin nodes, deck nodes, safe points, escape passage nodes, and emergency equipment storage points.

[0010] Furthermore, the positioning technology used for locating the positions of each person on the ship is image positioning or personnel terminal wifi positioning.

[0011] Furthermore, in Step 2, calculate the evacuation speed according to the position status and environmental status of each person on the ship, and then use the evacuation speed as the heuristic function of the ant colony algorithm, and perform evacuation path planning based on the heuristic function to obtain an evacuation path set; specifically:

[0012] Step 201: Initialize the number of ants as ω, the pheromone factor as α, the heuristic function factor as β, the maximum number of iterations as iter_max, the minimum number of iterations as iter_min, the initial position of the ant as G k and the end position of the ant as F k ;

[0013] Step 202: Initialize the number of iterations Z = 1;

[0014] Step 203: Initialize the time t = 0;

[0015] Step 204: Calculate the population density ρ ij (t) on the road segment between node i and node j at time t according to the personnel positioning result:

[0016]

[0017] where f ij (t) is the number of people on the road segment between node i and node j at time t, and V ij represents the capacity of the road segment between node i and node j;

[0018] Then calculate the environmental complexity θ ij on the road segment between node i and node j:

[0019]

[0020] Among them, a, b, and c respectively represent the weight coefficient of the number of obstacles, the weight coefficient of the obstacle position score, and the weight coefficient of the road section turning frequency, N obs represents the number of obstacles on the road section between node i and node j, P s′ represents the score of the s'-th obstacle on the road section between node i and node j, s' = 1, 2, …, N obs , F turn represents the number of turns required on the road section between node i and node j;

[0021] Step 205: According to ρ ij (t) and θ ij calculate the evacuation speed v ij (t) on the road section between node i and node j at time t;

[0022]

[0023] Among them, v0 represents the evacuation speed under ideal conditions, ε represents the weight coefficient of the influence of crowd density on speed, and e is the base of the natural logarithm;

[0024] Step 206: Take the evacuation speed v ij (t) as the heuristic function, and calculate the probability that ant k selects the road section between node i and node j at time t

[0025]

[0026] Among them, τ ij (t) represents the pheromone concentration on the road section between node i and node j at time t, τ is (t) represents the pheromone concentration on the road section between node i and node s at time t, v is (t) represents the evacuation speed on the road section between node i and node s at time t, allow k represents the set of alternative nodes of ant k, and s represents the nodes that have not been assigned to ant k;

[0027] Then, according to the calculated probability, ant k selects the road section at time t;

[0028] Step 207: Update the local pheromone:

[0029]

[0030] Among them, τ ij (t + 1) represents the pheromone concentration on the road section between node i and node j at time t + 1, Denote the pheromone released by the k-th ant on the section between node i and node j, ω’ represents the number of ants choosing the section between node i and node j at time t, and rou ij (t + 1) represents the pheromone evaporation factor on the section between node i and node j at time t+1;

[0031] Step 208, determine whether all ants have reached the end point;

[0032] If all ants have reached the end point, then execute Step 209;

[0033] Otherwise, let t = t + 1, and return to execute Step 204;

[0034] Step 209, calculate the global pheromone concentration of the path from the starting point to the end point of each ant respectively, and then select the paths with the global pheromone concentration greater than the threshold in the paths obtained in the current iteration as the paths selected in the current iteration;

[0035] Then, from the paths selected in the current iteration, screen out the path with the maximum global pheromone concentration; and update the pheromone concentration according to the path screening result, and use the updated pheromone concentration as the pheromone concentration at the beginning of the next iteration;

[0036] Then calculate the total evacuation time T of the paths with the global pheromone concentration greater than the threshold selected in the current iteration Z , and denote the total evacuation time of the paths with the global pheromone concentration greater than the threshold selected in the most recent iteration stored in the path set R b as T Z-1 ;

[0037] Step 210, determine whether T Z -T Z-1 <δ;

[0038] If T Z -T Z-1 <δ, then add the paths with the global pheromone concentration greater than the threshold selected in the current iteration to the path set R b , and continue to execute Step 211;

[0039] If T Z -T Z-1 ≥δ, then execute Step 212;

[0040] Step 211, determine whether Z is less than the minimum number of iterations;

[0041] If Z is less than the minimum number of iterations, then let Z = Z + 1, and return to execute Step 203;

[0042] Otherwise, output the path set R b ;

[0043] Step 212: Determine whether Z is equal to the maximum number of iterations;

[0044] If Z is less than the maximum number of iterations, then set Z = Z + 1, and return to execute Step 203;

[0045] If Z is equal to the maximum number of iterations, then output the path set R b 。

[0046] Furthermore, the calculation methods of the weight coefficients a, b, and c are as follows:

[0047] a = 0.2 × N obs + 0.8

[0048] b = 0.05 × V obs

[0049]

[0050] where V obs is the total volume of obstacles on the section between node i and node j, and the unit is cubic meters; L ij is the length of the section between node i and node j, and α K represents the angle of the K-th turn.

[0051] Furthermore, the value of the pheromone factor is set to 1 to 5, the value of the heuristic function factor is set to 2 to 5, the minimum number of iterations is set to 100, and the maximum number of iterations is set to 1000.

[0052] Furthermore, the calculation method is as follows:

[0053]

[0054] where Q is a time-varying function, and D ij represents the equivalent length of the section between node i and node j assigned to ant k;

[0055]

[0056] where Z is the current number of iterations, Z max is the maximum number of iterations, and Q1 is a constant;

[0057]

[0058] where is the temperature influence coefficient on the section between node i and node j, λ is the visibility influence coefficient on the section between node i and node j, and λ ∈ [0, 1].

[0059] Furthermore, update the pheromone concentration according to the path screening result; specifically:

[0060]

[0061] Among them, L represents the path with the largest global pheromone concentration screened in the current iteration, and ij∈L means that the section between node i and node j is on path L; τ i ′ j represents the pheromone concentration on the section between node i and node j at the end of the current iteration; τ i ′ j ′ represents the updated pheromone concentration on the section between node i and node j; rou i ′ j represents the pheromone evaporation factor on the section between node i and node j at the end of the current iteration; R is the reward coefficient; represents the average crowd flow on the section between node i and node j when the k-th ant selects the section between node i and node j, and ω″ is the total number of ants that selected the section between node i and node j in the previous iteration; is the historical optimal average crowd flow on the section between node i and node j; is an intermediate variable.

[0062] Furthermore, the value of the reward coefficient R is 2.

[0063] Even further, the and are calculated as follows:

[0064]

[0065] Among them, f ij k represents the total number of people on the section between node i and node j when the k-th ant selects the section between node i and node j; T represents the evacuation time on the section between node i and node j after the k-th ant selects the section between node i and node j; P is the total number of people to be evacuated; T b is the evacuation time of the path with the minimum evacuation time in the set R b ; ij∈R b means that the section between node i and node j is in the set R b .

[0066] The beneficial effects of the present invention are:

[0067] The present invention applies the advantages of the ant colony algorithm in the field of dynamic programming to the problem of ship personnel evacuation, which can effectively solve the problem of changing actual ship environmental conditions and continuously optimize the evacuation route. By combining personnel density and environmental complexity, an evacuation speed is proposed as the heuristic function for improving the ant colony algorithm. By avoiding being trapped in local optima, the accuracy and search efficiency of the ant colony algorithm can be improved, thereby enhancing the evacuation efficiency, effectively ensuring the safety of the evacuees on board, and increasing the applicability of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 is a flowchart of a ship fire evacuation route planning method based on the ant colony algorithm of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0069] DETAILED DESCRIPTION OF THE INVENTION 1: Combined with Figure 1 describe this embodiment. A ship fire evacuation route planning method based on the ant colony algorithm described in this embodiment specifically includes the following steps:

[0070] Step 1: Generate a ship path network C according to the nodes and paths in the ship, and denote the generated ship path network as C[R, N], where R represents the set of road segments in the ship path network C (each road segment connects two nodes), and N represents the set of nodes in the ship path network C;

[0071] Step 2: Locate the positions of each person on the ship, and then use the position F of the safe point in the ship path network C k as the evacuation end point. Calculate the evacuation speed according to the position status and environmental status (i.e., environmental complexity) of each person on the ship, and then use the evacuation speed as the heuristic function of the ant colony algorithm, and perform evacuation route planning based on the heuristic function to obtain an evacuation route set.

[0072] DETAILED DESCRIPTION OF THE INVENTION 2: The difference between this embodiment and DETAILED DESCRIPTION OF THE INVENTION 1 is that: the set of nodes includes cabin nodes, deck nodes, safe points, escape passage nodes, and emergency equipment storage points.

[0073] Other steps and parameters are the same as those in DETAILED DESCRIPTION OF THE INVENTION 1.

[0074] DETAILED DESCRIPTION OF THE INVENTION 3: The difference between this embodiment and DETAILED DESCRIPTION OF THE INVENTION 1 or 2 is that: for the positioning of the positions of each person on the ship, the positioning technology used is image positioning or personnel terminal wifi positioning.

[0075] Other steps and parameters are the same as those in DETAILED DESCRIPTION OF THE INVENTION 1 or 2.

[0076] The positioning technologies that can be adopted in the present invention include, but are not limited to, image positioning and personnel terminal wifi positioning.

[0077] Embodiment 4: The difference between this embodiment and any one of Embodiments 1 to 3 is that in Step 2, the evacuation speed is calculated based on the position status and environmental status of each person on the ship, and then the evacuation speed is used as the heuristic function of the ant colony algorithm, and the evacuation path planning is carried out based on the heuristic function to obtain the evacuation path set; specifically:

[0078] Step 201: Initialize the number of ants as ω, the pheromone factor as α, the heuristic function factor as β, the maximum number of iterations as iter_max, the minimum number of iterations as iter_min, the initial position of the ant as G k and the end position of the ant as F k ;

[0079] Step 202: Initialize the iteration number Z = 1;

[0080] Step 203: Initialize the time t = 0;

[0081] Step 204: Calculate the crowd density ρ ij (t) on the section between node i and node j at time t according to the personnel positioning result:

[0082]

[0083] where fi ij (t) is the number of people on the section between node i and node j at time t, and V ij represents the capacity of the section between node i and node j;

[0084] Then calculate the environmental complexity θ ij :

[0085]

[0086] where a, b, and c respectively represent the weight coefficients of the number of obstacles, the weight coefficient of the obstacle position score, and the weight coefficient of the section turning frequency, and N obs represents the number of obstacles on the section between node i and node j, P s′ represents the score of the s'-th obstacle on the section between node i and node j (the score of the obstacle located in the middle of the section is 3, and the score of the obstacle located at other positions on the section is 1), s' = 1, 2,..., N obs , F turn represents the number of turns required on the section between node i and node j;

[0087] Step 205: Calculate the evacuation speed vij(t) on the section between node i and node j at time t according to ρ ij (t) and θ ij ;

[0088]

[0089] Among them, v0 represents the evacuation speed under ideal conditions, v0 = 1.2 m / s, ε represents the weight coefficient of the influence of crowd density on speed (which can be selected according to the actual situation), and e is the base of the natural logarithm;

[0090] Step 206: Take the evacuation speed v ij (t) as the heuristic function, and calculate the probability that ant k selects the section between node i and node j at time t according to the heuristic function

[0091]

[0092] Among them, τ ij (t) represents the pheromone concentration on the section between node i and node j at time t (the pheromone concentration at time 0 is obtained through initialization), τ is (t) represents the pheromone concentration on the section between node i and node s at time t, v is (t) represents the evacuation speed on the section between node i and node s at time t, allow k represents the set of alternative nodes of ant k (that is, the set of feasible nodes to which the ant can be assigned at time t, that is, in the ship route network, the set of nodes that have a section with node i), and s represents the node that has not been assigned to ant k;

[0093] Then, according to the calculated probability, ant k selects the section at time t (the ant selects the section with the largest corresponding probability);

[0094] Step 207: Update the local pheromone:

[0095]

[0096] Among them, τ ij (t + 1) represents the pheromone concentration on the section between node i and node j at time t + 1, represents the pheromone released by the k-th ant on the section between node i and node j, ω' represents the number of ants that select the section between node i and node j at time t, rou ij (t + 1) represents the pheromone evaporation factor on the section between node i and node j at time t + 1 (obtained according to the temperature influence coefficient and visibility influence coefficient at time T + 1);

[0097] Step 208: Determine whether all ants have reached the end point;

[0098] If all ants have reached the end point, then execute Step 209;

[0099] Otherwise, let t = t + 1, and return to execute step 204; the number of people on the section between node i and node j at the next moment is also updated accordingly, that is, f ij (t + 1)= f ij (t)+ n;

[0100] Step 209: Calculate the global pheromone concentration of the path from the starting point to the ending point for each ant respectively, and then select the paths with a global pheromone concentration greater than the threshold among the paths obtained in the current iteration (for any ant, the sum of the pheromone concentrations of each section in the evacuation path obtained by this ant is used as the global pheromone concentration of the path obtained by this ant) as the paths selected in the current iteration;

[0101] Then, from the paths selected in the current iteration, screen out the path with the largest global pheromone concentration; and update the pheromone concentration according to the path screening result, and use the updated pheromone concentration as the pheromone concentration at the beginning of the next iteration (that is, when t = 0 in the next iteration) (the initial pheromone concentration in the first iteration can be randomly initialized);

[0102] Then calculate the total evacuation time T of the paths with a global pheromone concentration greater than the threshold selected in the current iteration Z , and denote the total evacuation time of the paths with a global pheromone concentration greater than the threshold selected in the most recent iteration stored in the path set R b as T Z-1 ; that is, after obtaining the evacuation times of each path selected in the current iteration respectively, then calculate the total evacuation time T of each path selected in the current iteration Z ; Similarly, calculate the total evacuation time T of the paths selected in the most recent iteration stored in the path set R b ; Z-1 ;

[0103] Step 210: Determine whether T Z - T Z-1 < δ (if it is the first iteration, there is no need to make a comparison, and directly use the global pheromone updated in the first iteration as the initial pheromone for the next iteration to start the next iteration);

[0104] If T Z - T Z-1 < δ, then add the paths with a global pheromone concentration greater than the threshold selected in the current iteration to the path set R b , and continue to execute step 211;

[0105] If T Z - T Z-1 ≥ δ, then execute step 212;

[0106] Step 211: Determine whether Z is less than the minimum number of iterations;

[0107] If Z is less than the minimum number of iterations, then set Z = Z + 1 and return to execute Step 203;

[0108] Otherwise, output the path set R b ;

[0109] Step 212: Determine whether Z is equal to the maximum number of iterations;

[0110] If Z is less than the maximum number of iterations, then set Z = Z + 1 and return to execute Step 203;

[0111] If Z is equal to the maximum number of iterations, then output the path set R b 。

[0112] Other steps and parameters are the same as those in any one of the specific embodiments one to three.

[0113] Since the personnel are distributed at various positions on the ship, the entire ship can be divided into multiple local areas. In this way, each local area can be used as a starting point for evacuation, and path planning can be carried out separately for each evacuation starting point. Moreover, during one iteration process, the previously selected sections will affect the personnel density on the subsequent sections, and subsequent sections will be selected based on the updated personnel density until all ants reach the evacuation end point. After evacuating for a period of time according to the currently planned evacuation path, the evacuation path can be re-planned according to the actual situation to continuously optimize the evacuation path.

[0114] Specific Embodiment Five: The difference between this embodiment and any one of the specific embodiments one to four is that the calculation methods of the weight coefficients a, b, and c are as follows:

[0115] a = 0.2×N obs + 0.8

[0116] b = 0.05×V obs

[0117]

[0118] where V obs is the total volume of obstacles on the section between node i and node j, with the unit of cubic meters; L ij is the length of the section between node i and node j, and α K represents the angle of the Kth turn.

[0119] Other steps and parameters are the same as those in any one of the specific embodiments one to four.

[0120] Embodiment Six: The difference between this embodiment and any one of Embodiments One to Five is that: the value of the pheromone factor is set to 1 to 5, the value of the heuristic function factor is set to 2 to 5, the minimum number of iterations is set to 100, and the maximum number of iterations is set to 1000.

[0121] Other steps and parameters are the same as those in any one of Embodiments One to Five.

[0122] The values of these parameters can be arbitrarily selected within the numerical range set in the present invention.

[0123] Embodiment Seven: The difference between this embodiment and any one of Embodiments One to Six is that: the calculation method is:

[0124]

[0125] where Q is a time-varying function, and D ij represents the equivalent length of the section between node i and node j to which ant k is assigned;

[0126]

[0127] where Z is the current number of iterations, and Z max is the maximum number of iterations (i.e., iter_max), Q1 is a constant, and its value range is 100 to 200;

[0128]

[0129] where is the temperature influence coefficient on the section between node i and node j, λ is the visibility influence coefficient on the section between node i and node j, and λ ∈ [0, 1].

[0130] Other steps and parameters are the same as those in any one of Embodiments One to Six.

[0131] The time-varying function represents the change of pheromone intensity with time and is used to facilitate obtaining the global optimal solution. The value-taking situations of the temperature influence coefficient and the visibility influence coefficient are shown in Table 1 and Table 2 respectively:

[0132] Table 1 Value-taking situation of the temperature influence coefficient

[0133]

[0134] Table 2 Value-taking situation of the visibility influence coefficient

[0135]

[0136] The value of the pheromone evaporation factor is determined according to the highest risk level among the risk levels corresponding to temperature and visibility, as shown in Table 3 specifically:

[0137] Table 3 Value-taking situation of pheromone evaporation factor

[0138]

[0139] Specific Embodiment Eight: The difference between this embodiment and any one of Embodiments One to Seven is that: the pheromone concentration is updated according to the path screening result; specifically:

[0140]

[0141] Among them, L represents the path with the largest global pheromone concentration screened in the current iteration, and ij∈L indicates that the section between node i and node j is on path L; τ i ′ j represents the pheromone concentration on the section between node i and node j at the end of the current iteration; τ″ ij represents the pheromone concentration on the section between node i and node j after update; rou i ′ j represents the pheromone evaporation factor on the section between node i and node j at the end of the current iteration; R is the reward coefficient; represents the average population flow on the section between node i and node j when the kth ant selects the section between node i and node j, and ω″ is the total number of ants that selected the section between node i and node j in the previous iteration process; is the historical optimal average population flow on the section between node i and node j; is an intermediate variable.

[0142] Other steps and parameters are the same as any one of Embodiments One to Seven.

[0143] Specific Embodiment Nine: The difference between this embodiment and any one of Embodiments One to Eight is that: the value of the reward coefficient R is 2.

[0144] Other steps and parameters are the same as any one of Embodiments One to Eight.

[0145] Specific Embodiment Ten: The difference between this embodiment and any one of Embodiments One to Nine is that: the and are calculated as follows:

[0146]

[0147] Among them, It represents the total number of people on the section between node i and node j when the k-th ant selects the section between node i and node j; T represents the evacuation time on the section between node i and node j after the k-th ant selects the section between node i and node j; P is the total number of people to be evacuated; T b is the set R b the evacuation time of the path with the minimum evacuation time in; ij ∈ R b represents the section between node i and node j in the set R b in (that is, in the set R b at least one path in includes the section between node i and node j).

[0148] Other steps and parameters are the same as one of the specific embodiments one to nine.

[0149] The above calculation examples of the present invention are only for explaining in detail the calculation model and calculation process of the present invention, rather than limiting the implementation manners of the present invention. For those of ordinary skill in the art, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation manners here. Any obvious changes or modifications derived from the technical solutions of the present invention are still within the protection scope of the present invention.

Claims

1. A method for ship fire evacuation route planning based on the ant colony algorithm, characterized in that The method specifically includes the following steps: Step 1: Generate a ship path network C based on the nodes and paths in the ship. Denote the generated ship path network as C[R, N], where R represents the set of road segments in the ship path network C, and N represents the set of nodes in the ship path network C; Step 2: Locate the positions of all personnel on the ship, and then use the safe point positions F in the ship path network C k as the evacuation end points. Calculate the evacuation speed based on the position states and environmental states of all personnel on the ship, and then use the evacuation speed as the heuristic function of the ant colony algorithm. Based on the heuristic function, conduct evacuation path planning to obtain an evacuation path set; In Step 2, calculate the evacuation speed according to the position states of the personnel on the ship and the environmental state, then use the evacuation speed as the heuristic function of the ant colony algorithm, and perform evacuation path planning based on the heuristic function to obtain an evacuation path set; specifically: Step 201: Initialize the number of ants as ω, the pheromone factor as α, the heuristic function factor as β, the maximum number of iterations as iter_max, the minimum number of iterations as iter_min, the initial position of the ants as G k and the end position of the ants as F k ; Step 202: Initialize the iteration number Z = 1; Step 203: Initialize the time t = 0; Step 204: Calculate the population density ρ(t) of the road section between node i and node j at time t according to the personnel positioning result ij (t):: Among them, f ij (t) is the number of people on the road section between node i and node j at time t, and V ij represents the capacity of the road section between node i and node j; Then calculate the environmental complexity θ on the road section between node i and node j ij : Among them, a, b, and c respectively represent the weight coefficient of the number of obstacles, the weight coefficient of the obstacle position score, and the weight coefficient of the road section turning frequency, N obs represents the number of obstacles on the road section between node i and node j, P s′ represents the score of the s'-th obstacle on the road section between node i and node j, s' = 1, 2, …, N obs , F turn represents the number of turns required on the road section between node i and node j; Step 205, according to ρ ij (t) and θ ij calculate the evacuation speed v ij (t) on the road section between node i and node j at time t; Among them, v0 represents the evacuation speed under ideal conditions, ε represents the weight coefficient of the influence of crowd density on speed, and e is the base of the natural logarithm; Step 206: Take the evacuation speed v ij (t) as the heuristic function, and calculate the probability that ant k selects the section between node i and node j at time t according to the heuristic function Among them, τ ij (t) represents the pheromone concentration on the road section between node i and node j at time t, τ is (t) represents the pheromone concentration on the road section between node i and node s at time t, v is (t) represents the evacuation speed on the road section between node i and node s at time t, allow k represents the set of alternative nodes of ant k, and s represents the node that has not been assigned to ant k; Then, according to the calculated probability, ant k selects the road segment at time t; Step 207: Update the local pheromone: Among them, τ ij (t + 1) represents the pheromone concentration on the road section between node i and node j at time t + 1, represents the pheromone released by the k-th ant on the road section between node i and node j, ω' represents the number of ants that choose the road section between node i and node j at time t, rou ij (t + 1) represents the pheromone evaporation factor on the road section between node i and node j at time t + 1; Step 208: Determine whether all ants have reached the end point; If all ants have reached the end point, execute Step 209; Otherwise, set t = t + 1, and return to execute Step 204; Step 209: Calculate the global pheromone concentration of the path from the starting point to the end point for each ant respectively, then select the paths with a global pheromone concentration greater than the threshold in the paths obtained in the current iteration as the paths selected in the current iteration; Then, select the path with the maximum global pheromone concentration from the paths selected in the current iteration; and update the pheromone concentration according to the path selection result, and use the updated pheromone concentration as the pheromone concentration at the beginning of the next iteration; Then calculate the total evacuation time T of the paths where the global pheromone concentration selected in the current iteration is greater than the threshold Z , and denote the total evacuation time of the paths where the global pheromone concentration selected in the most recent iteration stored in the path set R b is greater than the threshold as T Z-1 ; Step 210, determine whether it satisfies T Z -T Z-1 <δ; If T Z -T Z-1 < δ, then add the paths with the global pheromone concentration greater than the threshold selected in the current iteration to the path set R b , and continue to execute step 211; If T Z -T Z-1 ≥ δ, then perform step 212; Step 211: Determine whether Z is less than the minimum number of iterations; If Z is less than the minimum number of iterations, set Z = Z + 1, and return to execute Step 203; Otherwise, the output path set R b ; Step 212: Determine whether Z is equal to the maximum number of iterations; If Z is less than the maximum number of iterations, set Z = Z + 1, and return to execute Step 203; If Z is equal to the maximum number of iterations, output the path set R b .

2. The method for planning a ship fire evacuation path based on the ant colony algorithm according to claim 1, characterized in that The node set includes cabin nodes, deck nodes, safety points, escape passage nodes, and emergency equipment storage points.

3. The method for planning a ship fire evacuation path based on the ant colony algorithm according to claim 1, wherein The positioning technology used to locate the positions of the personnel on the ship is image positioning or personnel terminal wifi positioning.

4. The method for planning a ship fire evacuation path based on the ant colony algorithm according to claim 1, characterized in that The calculation methods of the weight coefficients a, b, and c are: a = 0.2×N obs + 0.8 b = 0.05 × V obs Among them, V obs is the total volume of obstacles on the road section between node i and node j, with the unit of cubic meters; L ij is the length of the road section between node i and node j, and α K represents the angle of the K-th turn.

5. The method for planning a ship fire evacuation path based on the ant colony algorithm according to claim 1, characterized in that The value of the pheromone factor is set to 1 - 5, the value of the heuristic function factor is set to 2 - 5, the minimum number of iterations is set to 100, and the maximum number of iterations is set to 1000.

6. The method for planning the ship fire evacuation path based on the ant colony algorithm according to claim 4, characterized in that, The calculation method is as follows: where Q is a time-varying function, D ij represents the equivalent length of the section between node i and node j to which ant k is assigned; where Z is the current iteration number, Z max is the maximum number of iterations, and Q1 is a constant; Among them, is the temperature influence coefficient on the road section between node i and node j. λ is the visibility influence coefficient on the road section between node i and node j, and λ ∈ [0, 1].

7. The method for planning the ship fire evacuation path based on the ant colony algorithm according to claim 6, characterized in that, The updating of the pheromone concentration according to the path selection result; specifically: Among them, L represents the path with the largest global pheromone concentration selected in the current iteration. ij ∈ L means that the section between node i and node j is on path L; τ′ ij represents the pheromone concentration on the section between node i and node j at the end of the current iteration; τ″ ij represents the updated pheromone concentration on the section between node i and node j; rou′ ij represents the pheromone evaporation factor on the section between node i and node j at the end of the current iteration; R is the reward coefficient; represents the average population flow on the section between node i and node j when the k-th ant selects the section between node i and node j. ω″ is the total number of ants that selected the section between node i and node j in the previous iteration process; is the historical optimal average population flow on the section between node i and node j; is an intermediate variable.

8. The method for planning the ship fire evacuation path based on the ant colony algorithm according to claim 7, wherein, The value of the reward coefficient R is 2.

9. The method for planning the ship fire evacuation path based on the ant colony algorithm according to claim 7, characterized in that, The said and are calculated as follows: Among them, represents the total number of people on the section between node i and node j when the k-th ant selects the section between node i and node j; T represents the evacuation time on the section between node i and node j after the k-th ant selects the section between node i and node j; P is the total number of people to be evacuated; T b is the evacuation time of the path with the minimum evacuation time in the set R b ; ij ∈ R b indicates that the section between node i and node j is in the set R b .

Images