A method for establishing an emergency resource scheduling model in a chemical industrial park and a scheduling method

Abstract

The application discloses a kind of establishment method and scheduling method of emergency resource scheduling model in chemical industry park, including determining the main objective function and secondary objective function of emergency resource scheduling based on chemical industry park information, then the two are converted into single-target problem emergency resource scheduling model.Introducing crossover operator and tabu list strategy to improve grey wolf algorithm and solve the model, output optimal resource scheduling strategy.The present application is based on the characteristics of emergency resource scheduling of sudden accident in chemical industry park, can intelligently screen, match and dispatch the resources in the park, so as to solve the technical problems of inaccurate resource scheduling, high cost and high failure rate when the existing chemical industry park occurs sudden event, can provide scientific and reasonable emergency resource scheduling strategy for decision maker, thereby improving emergency disposal and rescue ability.

Description

Technical Field

[0001] This invention relates to the field of emergency resource scheduling, and in particular to a method for establishing and scheduling an emergency resource scheduling model in a chemical industrial park. Background Technology

[0002] With the rapid development of my country's chemical industry, the construction of chemical industrial parks has become an indispensable part. While the entry of chemical plants into these parks improves production technology and promotes the industry's economy, it also brings challenges to safe production. Chemical industrial parks store large quantities of flammable and explosive materials. If a sudden accident such as a fire, explosion, or gas leak occurs within the park, it could trigger a catastrophic domino effect, resulting in significant personal injury and property damage. Given the sudden nature of such accidents, it is crucial to quickly formulate a reasonable emergency resource allocation plan when such an event occurs.

[0003] The emergency resource dispatch system of chemical industrial parks relies on its rapid response mechanism and adaptability to complex environments to cope with emergencies. Currently, most research on emergency resource dispatch by domestic and international scholars focuses on natural disaster scenarios, which is not well-suited to the context of emergency resource dispatch in chemical industrial parks. In reality, emergency resource dispatch in chemical industrial parks needs to consider multiple influencing factors. Chemical industrial parks are areas with a high concentration of hazardous chemicals, and hazardous chemical accidents exhibit a typical domino effect; a sudden safety accident at one enterprise, if not handled promptly, will simultaneously affect other enterprises within the park. Therefore, the occurrence of safety accidents in chemical industrial parks places more stringent demands on the timeliness and adequacy of emergency resources. Current known models and methods cannot meet practical needs, thus impacting emergency resource dispatch strategies.

[0004] Currently, there is an urgent need in this field for a method that can quickly, reasonably, and effectively transport emergency resources from reserve points to accident sites in the event of an accident, thereby minimizing losses and preventing the occurrence of other disasters. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention provides an emergency resource scheduling method for chemical industrial parks. It constructs a single-objective emergency resource scheduling model to address situations involving multiple reserve points, accident sites, and resources during emergency resource scheduling. An improved Grey Wolf algorithm is then applied to solve this model to obtain emergency resource scheduling strategies for sudden events, thereby improving emergency response and rescue capabilities.

[0006] This invention discloses a method for establishing an emergency resource dispatch model in a chemical industrial park. The applied chemical industrial park has several reserve points, each equipped with several emergency resources. These emergency resources can be transported to various accident sites via transport vehicles. The method includes:

[0007] Based on information from the chemical industrial park reflecting the status of reserve points, emergency resources, transport vehicles, and accident sites, the primary and secondary objective functions for emergency resource allocation are determined. The primary objective function aims to minimize the time required for emergency resource allocation to accident sites, while the secondary objective function aims to minimize the cost of emergency resource allocation to accident sites.

[0008] The ideal point method is used to transform the primary objective function and the secondary objective function into a single-objective emergency resource scheduling model, in which the emergency resource scheduling model aims to simultaneously minimize the emergency resource scheduling time and cost.

[0009] Furthermore, the chemical industrial parks in which this method is applied satisfy the following basic assumptions:

[0010] The storage capacity of various emergency resources at each reserve point is known;

[0011] The demand for various emergency resources at each accident site is known;

[0012] The transportation distances from each reserve point to each accident site are known.

[0013] The transport vehicles are identical and plentiful, and the vehicle speed and unit distance transport cost are known.

[0014] There was no traffic congestion or road damage.

[0015] Each reserve point is independent of the others and cannot allocate emergency resources to each other.

[0016] Furthermore, in the chemical industrial park where this method is applied, there are m reserve points A = {A1, A2, ..., A...} i ,...,A m}, n accident points B = {B1, B2, ..., B j ,...,B n}, u types of emergency resources C={C1,C2,...,C k ,...,C u};

[0017] The expression for the main objective function is as follows:

[0018]

[0019] In this expression, t i *H ij Preparation time, t i Preparation time required for each transport vehicle to load resources at each reserve point, H ij Reserve Point A i Towards accident point B j The number of transport vehicles dispatched, H ijThe calculation expression is as follows:

[0020]

[0021] In this expression, a k X represents the space coefficient of each item of the k-th resource in each transport vehicle. ijk Indicates reserve point A i Towards accident point B j The quantity of the k-th type of emergency resource to be allocated;

[0022] In the expression of the main objective function, T ij *G ij Refers to travel time, G ij Emergency resources are transferred from reserve point A. i To accident point B j The scheduling status is 1 if scheduled out, and 0 otherwise. G ij The calculation expression is as follows:

[0023]

[0024] T ij The calculation expression is as follows:

[0025] T ij =d ij / v

[0026] d ij This indicates that resources are from reserve point A. i To accident point B j The transport distance is v, where v represents the speed of the transport vehicle.

[0027] The expression for the secondary objective function is as follows:

[0028]

[0029] In this expression, s refers to the transportation cost per unit distance between vehicles, and d... ij Emergency resources are transferred from reserve point A. i To accident point B j The transport distance, H ij Reserve Point A i Towards accident point B j The number of transport vehicles dispatched.

[0030] Furthermore, the method also includes:

[0031] The following constraints are set to limit the scope of the emergency resource scheduling model:

[0032] Failure time constraint: This requires that all necessary emergency resources arrive at each accident site before the domino effect occurs. The expression for this constraint is as follows:

[0033]

[0034] Among them, ttf j This refers to the latest time to prevent a domino effect from occurring in storage tanks near the accident site;

[0035] Storage constraints at reserve points: Reserve point A i The total amount of the k-th type of resource allocated to each incident point is less than or equal to that of the reserve point A. i The expression for the total reserve quantity of the k-th resource is as follows:

[0036]

[0037] Among them, P ik Reserve Point A i The amount of reserves for the k-th resource;

[0038] Demand constraint at accident site: All reserve points are for accident site B. j The number of times the k-th type of resource is scheduled is equal to the incident point B. j The demand for the k-th resource is expressed as follows:

[0039]

[0040] Among them, Q jk Point B of the accident j The demand for the k-th resource;

[0041] Non-negative integer constraint on variables: representing reserve point A i Towards accident point B j The number of resources to be scheduled can be non-negative integers, and its expression is as follows:

[0042] X ijk ≥0 and is an integer,

[0043] Furthermore, the method of using ideal points to transform the primary objective function and secondary objective function into a single-objective emergency resource scheduling model also includes: performing normalization processing during the transformation to eliminate the influence of inconsistent dimensions.

[0044] The expression for the emergency resource scheduling model is as follows:

[0045]

[0046] Among them, f 1，min and f 1，max These are the optimal and worst solutions for the main objective function f1, respectively. 2，min and f 2，maxρ1 and ρ2 are the optimal and worst solutions of the secondary objective function, respectively. The weight coefficients ρ1 and ρ2 are the expected weights of f1 and f2, respectively, where ρ1 + ρ2 = 1 and 0 ≤ ρ1 ≤ 1, 0 ≤ ρ2 ≤ 1.

[0047] This invention also discloses a method for scheduling emergency resources in chemical industrial parks. This method solves the emergency resource scheduling model established by the aforementioned method for establishing an emergency resource scheduling model in chemical industrial parks using an improved gray wolf optimization algorithm. The scheduling method includes:

[0048] S1: Initialize population parameters;

[0049] S2: Calculate the fitness values ​​of all gray wolf individuals in the population, and select the three best gray wolf individuals as the first generation α wolf, β wolf, and δ wolf;

[0050] S3: Calculate the convergence factor a and coefficient vectors A and C, and update the positions of the remaining gray wolves;

[0051] S4: Use crossover operations to generate new gray wolf individuals and select the best ones to keep;

[0052] S5: Recalculate the fitness value of each gray wolf and select the top three individuals as the new generation of α wolf, β wolf, and δ wolf;

[0053] S6: Determine whether the selected gray wolf is in the taboo list. If it exists, exclude the gray wolf from the candidate solution and jump to S5. If it does not exist, proceed to the next step.

[0054] S7: Update the tabu list by adding the selected solution to the tabu list and replacing the earliest solution to enter the tabu list;

[0055] S8: Determine if the number of iterations has reached the maximum. If it has, jump to S9; otherwise, jump to S3 to start the next iteration.

[0056] S9: Output the optimal resource scheduling strategy.

[0057] Furthermore, in this method, a three-dimensional matrix of m*n*u is used to represent the position of an individual gray wolf, and the position of the individual gray wolf is shown in the following formula:

[0058]

[0059]

[0060] Where, φ l This represents the three-dimensional matrix position of the l-th individual gray wolf, where u represents the number of resource types. Let m represent the two-dimensional array corresponding to the k-th type of resource, m represent the number of reserve points, and n represent the number of accident points.

[0061] Furthermore, the method also includes:

[0062] Convert the positions of the gray wolves from real numbers to integers using the following encoding transformation method, which encodes the elements x of all individual gray wolves in the continuous space. i All are mapped to integers y i ,Right now:

[0063]

[0064] Where [x] is the floor function, and is the largest integer not greater than the real number x; L and U are respectively y i The lower and upper bounds of y i ∈[L,U]∪Z + .

[0065] Furthermore, step S4 specifically includes:

[0066] The crossover operation extracts some information from ordinary gray wolves and exchanges it with the corresponding location information of the alpha wolf. If the fitness value of the new gray wolf individual generated after the exchange is better, the gray wolf individual after the exchange is retained and the original gray wolf individual is eliminated; otherwise, the original individual is retained.

[0067] Furthermore, in steps S6 and S7, the tabu list is used to record the local optimal solutions selected in the previous several iterations, so as to avoid them in the next iteration to achieve the purpose of escaping local optima.

[0068] The present invention has at least the following beneficial effects:

[0069] Based on the characteristics of emergency resource scheduling in chemical industrial parks, this invention constructs a single-objective emergency resource scheduling model with the aim of minimizing time and cost, which is more reasonable than existing models.

[0070] Furthermore, this invention employs an improved gray wolf algorithm to obtain the optimal emergency resource scheduling strategy, enabling intelligent screening, matching, and scheduling of resources within the park. This solves the technical problems of inaccurate resource scheduling, high costs, and high error rates in existing chemical industrial parks during emergencies, providing decision-makers with a scientific and reasonable emergency resource scheduling strategy, thereby improving emergency response and rescue capabilities.

[0071] Other beneficial effects of the present invention will be described in detail in the Detailed Description of the Embodiments section. Attached Figure Description

[0072] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 This is a flowchart of the method for establishing an emergency resource scheduling model in a chemical industrial park, as disclosed in this invention.

[0074] Figure 2 This is a flowchart of the emergency resource scheduling method in a chemical industrial park disclosed in this invention.

[0075] Figure 3 This is a schematic diagram of the stochastic crossover process principle of the emergency resource scheduling method in chemical industrial parks disclosed in this invention. Detailed Implementation

[0076] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other implementation methods obtained by those skilled in the art without creative effort are within the scope of protection of this invention. When a fire occurs in a chemical industrial park, a scientific and reasonable emergency resource allocation plan must be formulated in the shortest possible time. This emergency resource allocation is characterized by urgency, and the emergency resources allocated to the chemical industrial park must meet its needs; otherwise, a domino effect can easily be triggered, leading to secondary accidents. The latest time for a domino effect to occur in a chemical industrial park accident, calculated based on a probability of no more than 10%, is generally about twenty-five minutes. When a fire occurs in a chemical industrial park, a large amount of fire-fighting and medical resources are required. At this time, multiple reserve depots need to cooperate in allocating emergency resources to meet rescue needs.

[0077] Example 1

[0078] In response to the characteristics of fire accidents in chemical industrial parks, this invention constructs a dual-objective emergency resource scheduling model for multiple reserve stations, multiple accident sites, and multiple resources. The primary objective is to minimize emergency resource scheduling time, and the secondary objective is to minimize transportation costs. The method for establishing this model is also described in [reference needed]. Figure 1 .

[0079] The scheduling relationship involved in this invention is represented as follows: There are m reserve points A = {A1, A2, ..., A...} i ,...,A m}, n accident points B = {B1, B2, ..., B j,...,B n}, u types of emergency resources C={C1,C2,...,C k ,...,C u Reserve Point A i Towards accident point B j The amount of resource k to be transferred is X ijk The symbols used are shown in Table 1 for clarity. To facilitate problem description, the symbols and meanings of the variables involved are presented in Table 1.

[0080] symbol meaning <![CDATA[A i ]]> The i-th storage point for various resources <![CDATA[B j ]]> The j-th accident site requiring rescue <![CDATA[Q jk ]]> <![CDATA[Accident point B j Demand for the k-th resource]]> <![CDATA[P ik ]]> <![CDATA[Reserve Point A i Reserve quantity of the k-th resource]]> <![CDATA[X ijk ]]> <![CDATA[Reserve point A i To accident point B j The quantity of the k-th resource transported]]> <![CDATA[a k ]]> The space coefficient of each item of resource k in each transport vehicle <![CDATA[H ij ]]> <![CDATA[Reserve Point A i To Accident Point B j The number of vehicles required to transport emergency resources]]> <![CDATA[G ij ]]> <![CDATA[Whether to dispatch emergency resources from reserve point A i to accident point B j > <![CDATA[T ij ]]> <![CDATA[The travel time of resources from reserve point A i to accident point B j > <![CDATA[t i ]]> Preparation time required for loading resources onto each vehicle at each reserve point <![CDATA[d ij ]]> <![CDATA[The transportation distance of resources from reserve point A i to accident point B j > s Transportation cost per unit distance of vehicle v Speed ​​of transport vehicles <![CDATA[ttf j ]]> The latest time to prevent a domino effect from occurring in storage tanks near the accident site

[0081] Table 1

[0082] In this embodiment, the following conditions should be met in the chemical industrial park: (1) the storage capacity of each reserve point for various emergency resources is known; (2) the demand for various emergency resources at each accident site is known; (3) the transportation distance from each reserve point to each accident site is known; (4) the transportation vehicles are the same and sufficient, and the vehicle speed and unit distance transportation cost are known; (5) there is no traffic jam or road damage; (6) each reserve point is independent of each other and cannot dispatch emergency resources to each other.

[0083] When calculating the shortest time target for emergency resource dispatch, first calculate the latest time when all the materials required for each accident point are delivered, and then sum the latest times of all accident points to obtain the shortest time target for the entire emergency resource dispatch process, as shown in expression (1):

[0084]

[0085] Among them, t i *H ij It's preparation time, T ij *G ij This refers to the travel time.

[0086] H ij This represents the number of transport vehicles dispatched from each reserve point to each accident site. Its calculation formula is as follows:

[0087]

[0088] Among them, H ij Representative reserve point A i Towards accident point B j The number of vehicles dispatched, a k X represents the space coefficient of each item of the k-th resource in each transport vehicle. ijk Indicates reserve point A i Towards accident point B j The quantity of the k-th type of resource to be scheduled.

[0089] When calculating travel time, use G. ijThis indicates that resources are from reserve point A. i To accident point B j The scheduling status of G ij The variable is 0-1; it is 1 when called and 0 otherwise, as shown in the following formula:

[0090]

[0091] Reserve Point A i Towards accident point B j The travel time is determined by the resource scheduling situation G. ij and travel time T ij Multiplying them together gives us T. ij The calculation expression is as follows:

[0092] T ij =d ij / v (4)

[0093] Where, d ij This indicates that resources are from reserve point A. i To accident point B j The transport distance is given by v, where v represents the speed of the transport vehicle.

[0094] This invention uses minimizing emergency resource allocation time as the primary objective and minimizing transportation costs as a secondary objective, which serves as the model's objective function. The model's objective function is shown below:

[0095]

[0096]

[0097] The constraints of the model are given in the following expressions (7) to (10):

[0098]

[0099]

[0100]

[0101]

[0102] Equation (7) is the failure time constraint, requiring that all necessary emergency resources arrive at each accident point before the domino effect occurs; Equation (8) is the storage capacity constraint at the reserve point, representing the storage capacity of reserve point A. i The total amount of the k-th type of resource allocated to each incident point is less than or equal to that of the reserve point A. i The total reserve quantity of the k-th resource; Equation (9) is the demand constraint at the accident point, indicating that all reserve points are accident point B. j The number of times the k-th type of resource is scheduled is equal to the incident point B.j The demand for the k-th resource; Equation (10) is a non-negative integer constraint on the variable, representing the reserve point A. i Towards accident point B j The range of values ​​for the number of resources to be scheduled is a non-negative integer.

[0103] Problem Transformation: The emergency resource scheduling problem in a chemical industrial park is a bi-objective optimization problem, requiring the shortest emergency scheduling time (f1) and the minimum transportation cost (f2). In solving multi-objective optimization problems, the objectives are often contradictory and conflicting, typically resulting in a set of non-dominated solutions, with no feasible solution simultaneously minimizing all objective functions. Emergency resource scheduling, however, requires quickly determining a specific scheduling strategy. To improve model accuracy and simplicity, this invention transforms the bi-objective optimization problem into a single-objective optimization problem using the ideal point method. The basic principle of the ideal point method is to construct an evaluation function. First, the optimal and worst solutions for each individual objective function are identified, i.e., positive and negative ideal points. Then, an ideal point is selected that minimizes the sum of distances to all optimal solutions.

[0104] f 1，min and f 1，max As the optimal and worst solutions of the objective function f1, f 2，min and f 2，max As the optimal and worst solutions for the objective function f2, considering that different objectives may have different priorities, weight coefficients ρ1 and ρ2 are introduced as the expected weights of f1 and f2, respectively, where ρ1 + ρ2 = 1 and 0 ≤ ρ1 ≤ 1, 0 ≤ ρ2 ≤ 1. Since the dimensions of each objective are different, normalization is required during the construction of the function to eliminate the influence of dimensions. The final constructed new optimization objective function is shown in the following equation:

[0105]

[0106] To obtain satisfactory solutions for the two objective functions—minimizing emergency dispatch time and minimizing transportation costs—and to get as close as possible to the ideal point, we need to minimize F. Using F as the new objective, we then solve the problem again by combining it with the model constraints.

[0107] Example 2

[0108] The emergency resource scheduling model constructed in this invention involves two objective functions and multidimensional decision variables, with these variables influencing each other. Its solution complexity is far higher than that of general scheduling and planning problems. General intelligent algorithms suffer from slow convergence speed and low solution quality. Therefore, this invention introduces an improved Grey Wolf Optimization Algorithm (TSGWO-GA) into the solution of this problem, effectively solving the problems of low solution efficiency and inaccurate resource scheduling in this emergency resource scheduling model.

[0109] The Gray Wolf Optimization (GWO) algorithm is inspired by the division of labor and cooperative foraging in wolves. It is a novel swarm intelligence algorithm that simulates the wolf hierarchy and predation behavior. The highest-ranking wolf is the alpha wolf, at the top of the food chain, responsible for leadership, decision-making, and other behaviors. Following them are the beta, delta, and ometa wolves. Although the beta and delta wolves are not the highest-ranking wolves, they can take over as new leaders when the alpha wolf loses leadership. The ometa wolf is the lowest-ranking wolf in the pack, responsible for balancing relationships within the group. The GWO algorithm treats each wolf as a potential solution, with alpha as the optimal solution, beta and delta as the second and third best solutions respectively, and ometa as an auxiliary reference. The GWO algorithm is an iterative optimization process that continuously updates the positions of the gray wolves. The process of gray wolves surrounding their prey can be represented by the following equation:

[0110] D = |C·X p (t)-X(t)| (12)

[0111] X(t+1)=X p (t)-A·D (13)

[0112] The above two formulas represent the distance between the individual and the prey, and the position update formula for the gray wolf, respectively. Where t is the current iteration number, A and C are coefficient vectors, and X... p Let A and C be the position vectors of the prey and the gray wolf, respectively. The formulas for calculating A and C are as follows:

[0113] A = 2a·r1-a (14)

[0114] C = 2·r² (15)

[0115] Where a is the convergence factor, and r1 and r2 are random numbers between [0, 1] as the number of iterations decreases linearly from 2 to 0.

[0116] Gray wolves can locate and surround prey. Once a gray wolf has identified the prey's location, β and δ, guided by α, lead the pack to surround the prey. The mathematical model for an individual gray wolf tracking its prey is described below:

[0117]

[0118] Among them, D α D β and D δ X represents the distances between α, β, and δ and other individuals, respectively; α X β and X δ C1, C2, and C3 represent the current positions of α, β, and δ, respectively; C1, C2, and C3 are random vectors, and X is the current position of the gray wolf.

[0119] The gray wolf gradually approaches its prey and updates its position. The wolf's position update process and result are represented by the following formula:

[0120]

[0121]

[0122] Gray wolves hunt by attacking, as shown in the following formula:

[0123] a=2-2·t / T (19)

[0124] Where t represents the current iteration number, and T is the set maximum iteration number. As the value of a decreases from 2 to 0, the corresponding value of A also varies in the interval [-a, a]. The larger the value of a, the more the gray wolves will move away from the prey, hoping to find a more suitable prey, thus prompting the wolf pack to conduct a global search (|A|>1). If the value of a is smaller, the gray wolves will move closer to the prey, prompting the wolf pack to conduct a local search (A|<1).

[0125] This invention provides an emergency resource scheduling method for chemical industrial parks. It uses an improved gray wolf optimization algorithm to find the optimal solution for the model obtained in the above embodiments. The specific details are as follows:

[0126] a) Encoding and conversion methods of solutions

[0127] In the improved gray wolf algorithm, the position of the individual gray wolf is the solution to the problem. This invention uses a three-dimensional matrix of m*n*u to represent the position of the individual gray wolf, as shown in the following formula:

[0128]

[0129]

[0130] Where, φ l The matrix represents the position of the l-th individual gray wolf, and u represents the number of resource types; where, Let m represent the two-dimensional array corresponding to the k-th type of resource, m represent the number of reserve points, and n represent the number of accident points.

[0131] The gray wolf population updated through iterative updates using the basic gray wolf algorithm is a real number matrix, which is not applicable to the discrete integer domain. Therefore, when solving the emergency resource scheduling problem in chemical industrial parks, it is necessary to convert the real positions of the gray wolves into integers to obtain the actual scheduling scheme.

[0132] To address the emergency resource allocation problem, the following encoding transformation method can be used to transform the elements x of all individual gray wolves in the continuous space. i All are mapped to integers y i ,Right now:

[0133]

[0134] In the formula: [x] is the floor function, which is the largest integer not greater than the real number x; L and U are respectively the values ​​of y and y. i The lower and upper bounds of y i ∈[L,U]∪Z + .

[0135] b) Improved Gray Wolf Algorithm TSGWO-GA

[0136] The GWO algorithm iterates by updating the positions of the remaining gray wolves based on the position of the leader gray wolf. It selects the three individuals with the best fitness as the leader gray wolves (α, β, δ), and then iterates using these three individuals until the iteration terminates. The algorithm converges quickly in the early stages, but if the leader wolf gets trapped in a local optimum in the later stages of iteration, the remaining individuals will tend to revolve around it when updating their positions. Therefore, the algorithm is prone to getting trapped in local optima in the later stages of iteration, resulting in slower convergence and lower optimization accuracy. Therefore, this invention uses the following two methods to enhance the global search optimization capability: first, it introduces the crossover operator from the genetic algorithm, where crossover generates new individuals by exchanging genes between chromosomes in the population; second, it introduces a tabu list strategy, which typically records the locally optimal solutions selected in the previous few iterations, avoiding them in the next iteration to escape local optima.

[0137] Cross-operation extracts partial information from ordinary gray wolves and exchanges it with the corresponding position information of the alpha wolf. If the fitness value of the new gray wolf individual generated after the exchange is better, the exchanged gray wolf individual is retained and the original gray wolf individual is eliminated; otherwise, the original individual is retained. The information exchanged here is Equation (20)φ l The overall information of the two-dimensional array is given by equation (21). The information in remains unchanged. For example... Figure 3 As shown, the specific operation process is as follows: Select the ordinary gray wolf individual as φ l ′, φ l For the alpha wolf individual, randomly extract and exchange information and merge it into φ l New gray wolf individuals φ are formed in the process. l ″.

[0138] like Figure 2 As shown, the steps of the improved TSGWO-GA algorithm are as follows:

[0139] Step 1: Initialize population parameters.

[0140] Step 2: Calculate the fitness values ​​of all gray wolf individuals in the population, and select the three best individuals as the first generation α, β, and δ.

[0141] Step 3: Calculate parameters a, A, and C, and update the positions of the remaining gray wolves using equations (16) to (18).

[0142] Step 4: Use crossover operations to generate new gray wolf individuals, and select the best ones to keep.

[0143] Step 5: Recalculate the fitness value of each gray wolf and select the top three individuals as the new generation of α, β, and δ.

[0144] Step 6: Determine if the selected gray wolf is in the taboo list. If it exists, exclude the gray wolf from the candidate solutions and jump to Step 5. If it does not exist, proceed to the next step.

[0145] Step 7: Update the tabu list by adding the selected solution to the tabu list and replacing the earliest solution to enter the tabu list.

[0146] Step 8: Determine if the number of iterations has reached the maximum. If it has, proceed to Step 9; otherwise, proceed to Step 3 to begin the next iteration.

[0147] Step 9: Output the optimal resource scheduling strategy.

[0148] The resource scheduling method in this application, based on the characteristics of emergency resource scheduling in chemical industrial parks, can intelligently screen, match and schedule resources within the park, thereby solving the technical problems of inaccurate resource scheduling, high cost and high error rate in existing chemical industrial parks when emergencies occur. It can provide decision-makers with a scientific and reasonable emergency resource scheduling strategy, thereby improving emergency response and rescue capabilities.

[0149] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for establishing an emergency resource dispatch model in a chemical industrial park, wherein the applied chemical industrial park has several reserve points, each reserve point has several emergency resources, and the emergency resources can be transported to various accident sites by transport vehicles, characterized in that... The method includes: Based on information from the chemical industrial park reflecting the status of reserve points, emergency resources, transport vehicles, and accident sites, the primary and secondary objective functions for emergency resource allocation are determined. The primary objective function aims to minimize the time required for emergency resource allocation to accident sites, while the secondary objective function aims to minimize the cost of emergency resource allocation to accident sites. In the chemical industrial parks where this method is applied, there are m One reserve point , n accident point , u Emergency resources ; The expression for the main objective function is as follows: ； In this expression, Preparation time t i This refers to the preparation time required for each transport vehicle to load resources at each reserve point. H ij Reserve points A i To the accident site B j The number of transport vehicles dispatched H ij The calculation expression is as follows: ； In this expression, Representing the k The space coefficient of each resource in each transport vehicle X ijk Indicates reserve points A i To the accident site B j Scheduling k The quantity of various emergency resources; In the expression of the main objective function, Refers to travel time. G ij This refers to emergency resources from reserve points A i to the accident site B j The scheduling status is determined by a value of 1 for dispatching and 0 otherwise. G ij The calculation expression is as follows: ； T ij The calculation expression is as follows: ; d ij Indicates resources from the reserve point A i to the accident site B j The transportation distance v Indicates the speed of the transport vehicle; The method also includes: setting the following constraints; Failure time constraint: This requires that all necessary emergency resources arrive at each accident site before the domino effect occurs. The expression for this constraint is as follows: ； in, ttf j This refers to the latest time to prevent a domino effect from occurring in storage tanks near the accident site; The ideal point method is used to transform the main objective function and the secondary objective function into a single-objective emergency resource scheduling model, in which the emergency resource scheduling model aims to simultaneously minimize the emergency resource scheduling time and cost. Normalization is performed during the conversion process to eliminate the effects of dimensional inconsistencies. The expression for the emergency resource scheduling model is as follows: ； in, f 1，min and f 1，max The main objective functions are respectively f The optimal and worst solutions for 1. f 2，min and f 2，max These represent the optimal and worst solutions of the secondary objective function, respectively, and their weighting coefficients. ρ 1 , ρ 2 As f 1 , f 2 The expected weights, where, ρ 1 + ρ 2 =1 and 0≤ ρ 1 ≤1, 0≤ ρ 2 ≤1.

2. The method for establishing an emergency resource scheduling model in a chemical industrial park according to claim 1, characterized in that, The chemical industrial parks in which this method is applied meet the following basic assumptions: The storage capacity of various emergency resources at each reserve point is known; The demand for various emergency resources at each accident site is known; The transportation distances from each reserve point to each accident site are known. The transport vehicles are identical and plentiful, and the vehicle speed and unit distance transport cost are known. There was no traffic congestion or road damage. Each reserve point is independent of the others and cannot allocate emergency resources to each other.

3. The method for establishing an emergency resource scheduling model in a chemical industrial park according to claim 2, characterized in that, The expression for the secondary objective function is as follows: ； In this expression, s refers to the transportation cost per unit distance for the vehicle. d ij This refers to emergency resources from reserve points A i to the accident site B j The transportation distance H ij Reserve points A i To the accident site B j The number of transport vehicles dispatched.

4. The method for establishing an emergency resource scheduling model in a chemical industrial park according to claim 3, characterized in that, The method also includes the following constraints to limit the scope of the emergency resource scheduling model: Storage constraints at reserve points: Reserve points A i Dispatch the first one to each accident point k The total amount of the seed resources is less than or equal to the reserve points. A i For the k The total reserve quantity of various resources is expressed as follows: ； in, P ik Reserve points A i For the k Reserves of various resources; Accident point demand constraint: All reserve points are accident points. B j Scheduling k The quantity of resources equals the accident point B j For the k The demand for this resource can be expressed as follows: ； in, Q jk Point to the accident point B j For the k The demand for such resources; Non-negative integer constraint on variables: representing reserve points A i To the accident site B j The number of resources to be scheduled can be non-negative integers, and its expression is as follows: X ijk And it is an integer. .

5. A method for dispatching emergency resources in a chemical industrial park, characterized in that, This scheduling method solves the emergency resource scheduling model established by the method described in any one of claims 1 to 4 for establishing an emergency resource scheduling model in a chemical industrial park, based on an improved gray wolf optimization algorithm. The scheduling method includes: S1: Initialize population parameters; S2: Calculate the fitness values ​​of all gray wolves in the population, and select the three best gray wolves as the first generation. α Wolf, β Wolf, δ Wolf; S3: Calculate the convergence factor of the parameters. a sum coefficient vector A , C Update the positions of the remaining gray wolves; S4: Use crossover operations to generate new gray wolf individuals and select the best ones to keep; S5: Recalculate the fitness value of each gray wolf and select the top three individuals as the new generation. α Wolf, β Wolf, δ Wolf; S6: Determine whether the selected gray wolf is in the taboo list. If it exists, exclude the gray wolf from the candidate solution and jump to S5. If it does not exist, proceed to the next step. S7: Update the tabu list by adding the selected solution to the tabu list and replacing the earliest solution to enter the tabu list; S8: Determine if the number of iterations has reached the maximum. If it has, jump to S9; otherwise, jump to S3 to start the next iteration. S9: Output the optimal resource scheduling strategy.

6. The method for dispatching emergency resources in a chemical industrial park according to claim 5, characterized in that, In this method, the following is adopted: The position of an individual gray wolf is represented by a three-dimensional matrix, as shown in the following formula: ； in, Φ l Indicates the first l The three-dimensional matrix position of each individual gray wolf u Indicates the number of resource types. φ k Indicates the first k A two-dimensional array corresponding to the class resource. Indicates the number of reserve points. Indicates the number of accident sites.

7. A method for dispatching emergency resources in a chemical industrial park according to claim 6, characterized in that, The method also includes: Convert the positions of the gray wolves from real numbers to integers using the following encoding transformation method, which encodes the elements of all individual gray wolves in the continuous space. All are mapped to integers ,Right now: in, The floor function is a function that rounds down to the nearest integer, and the integer is not greater than a real number. The largest integer; and They are respectively The lower and upper bounds, i.e. .

8. A method for dispatching emergency resources in a chemical industrial park according to claim 5, characterized in that, ； Step S4 specifically includes: The crossover operation extracts some information from ordinary gray wolves and exchanges it with the corresponding location information of the alpha wolf. If the fitness value of the new gray wolf individual generated after the exchange is better, the gray wolf individual after the exchange is retained and the original gray wolf individual is eliminated; otherwise, the original individual is retained.

9. A method for dispatching emergency resources in a chemical industrial park according to claim 5, characterized in that, In steps S6 and S7, the tabu list is used to record the local optima selected in the previous several iterations, so as to avoid them in the next iteration and achieve the purpose of escaping local optima.

Images