Emergency resource planning method and device based on sparse representation, equipment and medium

Abstract

This invention provides an emergency resource planning method, apparatus, equipment, and medium based on sparse representation. The method includes: acquiring environmental variable information of a target area with disaster relief mission requirements; determining the superimposed and non-superimposed emergency resources required by the target area based on the environmental variable information; establishing a first optimization model based on the disaster relief mission requirements and the attribute information of the non-superimposed emergency resources; establishing a second optimization model based on the environmental variable information, the disaster relief mission requirements, and the attribute information of the superimposed emergency resources, and performing sparse modeling on the second optimization model based on sparse representation; solving the first optimization model and the sparsely modeled second optimization model respectively to obtain an emergency resource planning scheme that meets the disaster relief mission requirements. This solves the problem that existing planning methods cannot guarantee that the obtained solution is the globally optimal solution, and achieves the effect of resource planning with minimal scheduling resources, realizing accurate emergency resource planning and reducing resource waste.

Description

Technical Field

[0001] This invention relates to the field of resource planning technology, and specifically to emergency resource planning methods, devices, equipment and media based on sparse representation. Background Technology

[0002] In the event of disasters such as fires, earthquakes, and floods, timely and precise planning and dispatch of emergency resources are crucial for disaster relief. Current resource planning techniques typically model large-scale resource allocation problems as integer programming problems and employ a random uniform partitioning algorithm to break down the problem into smaller subproblems. These subproblems are then solved in parallel using a distributed framework to improve speed. However, this approach places high demands on computer hardware and performance, and it still struggles to guarantee a globally optimal solution, thus failing to achieve the best resource planning results. Summary of the Invention

[0003] In view of this, embodiments of the present invention provide an emergency resource planning method, apparatus, equipment and medium based on sparse representation, to overcome the problem that the resource planning methods in the prior art are difficult to achieve the best resource planning effect.

[0004] According to a first aspect, embodiments of the present invention provide an emergency resource planning method based on sparse representation, the method comprising:

[0005] Obtain environmental variable information for target areas where disaster relief missions are needed, wherein the environmental variable information is environmental parameters related to the use of emergency resources;

[0006] Based on the environmental variable information, the superimposed and non-superimposed emergency resources required for the target area are determined. The superimposed resources are used to characterize the emergency resources whose ability to meet the disaster relief mission requirements increases with the amount of resources added.

[0007] A first optimization model is established based on the disaster relief task requirements and the attribute information of non-stackable emergency resources. The first optimization model is used to solve the planning decision of non-stackable emergency resources.

[0008] A second optimization model is established based on the environmental variable information, the disaster relief task requirements, and the attribute information of the superimposed emergency resources. The second optimization model is sparsely modeled based on sparse representation. The second optimization model is used to solve the planning decision of the superimposed emergency resources.

[0009] Solve the first optimization model and the second optimization model after sparse modeling respectively to obtain an emergency resource planning scheme that meets the requirements of the disaster relief mission.

[0010] Optionally, establishing the first optimization model based on the disaster relief mission requirements and non-overlapping emergency resources includes:

[0011] Based on the disaster relief mission requirements, objective rule information for the current disaster relief mission is determined, and the objective rule information is used to filter the emergency resources required for the current disaster relief mission.

[0012] A first optimization model is established based on the objective rule information and the attribute information of non-overlapping emergency resources.

[0013] Optionally, the first optimization model is represented by the following formula:

[0014]

[0015] Where A1 represents the first optimization model, This represents the utility function of resource planning. Let y represent the matrix of decision variables corresponding to non-superimposable resources. i Let y represent the decision variable corresponding to the i-th non-superimposable resource in the decision variable matrix, and y i ∈{0,1}, This represents the availability information matrix of resources in non-overlapping resource attribute information. This represents a rule vector composed of objective rule information.

[0016] Optionally, establishing a second optimization model based on the environmental variable information, the disaster relief task requirements, and the attribute information of the superimposed emergency resources includes:

[0017] Based on the environmental variable information and the disaster relief mission requirements, the resource redundancy and the weight values ​​corresponding to the resource cost of the superimposed emergency resources are determined.

[0018] Based on the attribute information of superimposed emergency resources, determine the resource quantity and resource cost of each superimposed emergency resource;

[0019] A second optimization model is established based on the quantity and cost of each superimposed emergency resource, as well as the weight values ​​corresponding to the resource redundancy and cost of the superimposed emergency resources.

[0020] Optionally, the second optimization model is represented by the following formula:

[0021]

[0022] Where A2 represents the second optimization model, Represents the resource redundancy function. Represents the resource cost function. Let x represent the matrix of decision variables corresponding to the superimposed resources. iLet x represent the decision variable corresponding to the i-th superimposed resource in the decision variable matrix, and x i ≥0 indicates that the utility information matrix of resources in the superimposed resource attribute information, C indicates that the resource cost information matrix of superimposed resources, and α1,2 indicate the weight values ​​corresponding to the resource redundancy and resource cost, respectively.

[0023] Optionally, the sparse modeling of the second optimization model based on sparse representation includes:

[0024] The resource redundancy function and resource cost function are modeled using sparse representation.

[0025] Optionally, solving the first optimization model and the second optimization model after sparse modeling to obtain an emergency resource planning scheme that meets the requirements of the disaster relief mission includes:

[0026] Solve the first optimization model to obtain the planning decisions for each non-overlapping resource;

[0027] Solve the second optimization model after sparse modeling to obtain the planning decisions for each superimposed resource;

[0028] An emergency resource planning scheme for the disaster relief task requirements is generated based on the planning decisions for each non-overlapping resource and each overlapping resource.

[0029] According to a second aspect, embodiments of the present invention provide an emergency resource planning device based on sparse representation, the device comprising:

[0030] The acquisition module is used to acquire environmental variable information of target areas where there is a need for disaster relief missions. The environmental variable information is environmental parameters related to the use of emergency resources.

[0031] The first processing module is used to determine the superimposed and non-superimposed emergency resources required by the target area based on the environmental variable information. The superimposed resources are used to characterize the emergency resources whose ability to meet the disaster relief mission requirements increases with the amount of resources added.

[0032] The second processing module is used to establish a first optimization model based on the disaster relief task requirements and the attribute information of non-overlapping emergency resources. The first optimization model is used to solve the planning decision of non-overlapping emergency resources.

[0033] The third processing module is used to establish a second optimization model based on the environmental variable information, the disaster relief task requirements and the attribute information of the superimposed emergency resources, and to perform sparse modeling on the second optimization model based on sparse representation. The second optimization model is used to solve the planning decision of the superimposed emergency resources.

[0034] The fourth processing module is used to solve the first optimization model and the second optimization model after sparse modeling, respectively, to obtain an emergency resource planning scheme that meets the requirements of the disaster relief mission.

[0035] According to a third aspect, embodiments of the present invention provide an emergency resource planning device based on sparse representation, comprising:

[0036] A memory and a processor are communicatively connected, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described in the first aspect and any of its alternative embodiments.

[0037] According to a fourth aspect, embodiments of the present invention provide a computer-readable storage medium storing computer instructions for causing a computer to perform the method described in the first aspect, or any optional embodiment of the first aspect.

[0038] The technical solution of this invention has the following advantages:

[0039] The emergency resource planning method based on sparse representation provided in this invention obtains environmental variable information of a target area with disaster relief needs. These environmental variables are environmental parameters related to the use of emergency resources. Based on the environmental variable information, the method determines the superimposed and non-superimposed emergency resources required by the target area. Superimposed resources characterize emergency resources whose ability to meet disaster relief needs increases with the amount of resources used. A first optimization model is established based on the disaster relief needs and the attribute information of non-superimposed emergency resources. This first optimization model is used to solve the planning decisions for non-superimposed emergency resources. A second optimization model is established based on the environmental variable information, disaster relief needs, and the attribute information of superimposed emergency resources. The second optimization model is then sparsely modeled based on sparse representation and used to solve the planning decisions for superimposed emergency resources. Solving both the first optimization model and the sparsely modeled second optimization model yields an emergency resource planning scheme that meets the disaster relief needs. By dividing emergency resources into stackable and non-stackable resources according to their ability to meet disaster relief needs, and establishing corresponding optimization models for each based on actual disaster relief requirements, specific decision-making schemes are obtained. For non-stackable resources, a conventional non-sparse representation method is used for modeling and solving to obtain feasible solutions. The stackable resource problem is modeled and solved based on the sparse representation method in convex optimization to solve the problem of slow solution speed and inability to guarantee that the solution obtained is the global optimal solution caused by integer programming modeling. It can achieve the effect of resource planning with minimal scheduling resources, realize the precise planning of emergency resources, and reduce resource waste. Attached Figure Description

[0040] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0041] Figure 1 This is a flowchart of an emergency resource planning method based on sparse representation in an embodiment of the present invention;

[0042] Figure 2 This is a structural block diagram of the emergency resource planning system in an embodiment of the present invention;

[0043] Figures 3A to 3C These are emergency resource planning schemes corresponding to three different fire scenarios in this embodiment of the invention;

[0044] Figure 4 This is a schematic diagram of the emergency resource planning device based on sparse representation according to an embodiment of the present invention;

[0045] Figure 5 This is a schematic diagram of the structure of an emergency resource planning device based on sparse representation according to an embodiment of the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] The technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0048] In the event of disasters such as fires, earthquakes, and floods, timely and precise planning and dispatch of emergency resources are crucial for disaster relief. Current resource planning techniques typically model large-scale resource allocation problems as integer programming problems and employ a random uniform partitioning algorithm to break down the problem into smaller subproblems. These subproblems are then solved in parallel using a distributed framework to improve speed. However, this approach places high demands on computer hardware and performance, and it still struggles to guarantee a globally optimal solution, thus failing to achieve the best resource planning results.

[0049] To address the aforementioned problems, embodiments of the present invention provide an emergency resource planning method based on sparse representation, such as... Figure 1 As shown, the method specifically includes the following steps:

[0050] Step S101: Obtain environmental variable information for the target area where disaster relief tasks are needed.

[0051] Among them, environmental variable information refers to environmental parameters related to the use of emergency resources. In this embodiment of the invention, it refers to variables that can quantify the on-site disaster relief needs. Taking a fire scenario as an example, environmental variable information includes: fire area, fire height, percentage of Class A combustibles, percentage of Class B combustibles, percentage of Class C combustibles, percentage of Class D combustibles, percentage of Class E combustibles, etc. Based on environmental variable information, the fire can be quantified. For example, [100,3,0.5,0.5,0,0,0,0] describes a fire with an area of ​​100㎡, a fire height of 3m, and Class A / B combustibles each accounting for half.

[0052] Specifically, such as Figure 2 The diagram illustrates an emergency resource planning system built upon the sparse representation-based emergency resource planning method provided in this embodiment of the invention. Emergency resources are stored in a resource information database according to different disaster relief tasks. In this embodiment, the illustration uses a fire in the target area as an example. In practical applications, disaster relief tasks can also include floods, earthquakes, etc., and this invention is not limited thereto. For example, the emergency resource information related to fire relief tasks stored in the resource information database is shown in Tables 1 and 2, including: fire truck product types, attributes, functions, rescue routes, etc.

[0053] Table 1

[0054]

[0055]

[0056] Table 2

[0057] serial number A11901 A11906 A11908 Path 1 Time 10.2 10.2 / Path 2 Time 11.4 11.4 11.4 Path 3 Time 13.6 13.6 / Path 4 Time 12.8 12.8 12.8

[0058] Step S102: Determine the superimposed and non-superimposed emergency resources required for the target area based on environmental variable information.

[0059] Specifically, resource information is pre-screened through the aforementioned resource information database, and resources are hierarchically divided into non-stackable resources and stackable resources. Stackable resources represent emergency resources whose ability to meet disaster relief needs increases with the quantity of resources. Taking fire extinguishing resources as an example, handheld dry powder fire extinguishers are stackable emergency resources. The dry powder in the extinguisher satisfies the fire extinguishing attribute, and increasing the number of extinguishers leads to an increase in the amount of dry powder, thus increasing the fire extinguishing capacity. Fire ladders, on the other hand, are non-stackable emergency resources because the corresponding business attribute of a ladder is rescue height, and increasing the number of ladders does not increase the rescue height (e.g., it is impossible to use 10 three-meter ladders for a 30-meter-high rescue). For example, non-stackable resources are all resources with the "rescue height" attribute, such as the "high-altitude rescue vehicle" in Table 1. During the resource pre-screening process, a list of all resources with the "rescue height" attribute is created as the non-stackable resource list, as shown in Table 3. The list of resources with attributes other than "rescue height" is the stackable resource list, as shown in Table 4.

[0060] Table 3

[0061] serial number Classification Rescue altitude (m) A11901 01: Water tanker 3 A11906 02: Foam Car 3 A11908 03: Pump truck 3 A11912 04: Dry powder + CO2 vehicle 3 A11915 06: High-altitude rescue vehicle 44 A11918 08: High-pressure jet truck 32

[0062] Table 4

[0063]

[0064] Among them, resources that can be superimposed generally refer to resource attribute information including, but not limited to, resource utility information matrix R such as the usage of n types of planned resources. m×n And resource cost information matrix C p×n (where m represents m different resource utility information corresponding to each specific resource, and p represents p different choices corresponding to each resource). This resource information can be quantified and can be superimposed and changed. Therefore, it is necessary to solve for the resource attribute matrix to achieve the scheduling decision variable, which is set as an n-dimensional vector.

[0065] For example, based on the superimposed resources, i.e., the quantity and cost of resources, such as water tank volume, number of water guns, dry powder capacity, general extinguishing agent capacity, ladder height, rescue time, and rescue cost; where the resource utility information matrix, including resource quantity, is R. m×n The information matrix of resource costs, etc., is C. p×n Here, m represents m different resource utility information, such as water tank volume, number of water guns, dry powder capacity, general extinguishing agent capacity, ladder height, etc.; p represents p different choices corresponding to each resource, such as different paths, etc., and the decision variable for scheduling this type of resource according to specific disaster relief rules is set as an n-dimensional vector. Among them, the resource utility information matrix R m×n Hereinafter referred to as R and the resource cost information matrix C p×n Hereinafter referred to as C, specific examples are as follows:

[0066]

[0067]

[0068] For non-overlapping resources, this refers to the resource availability information in the resource attribute information, denoted as... (Where q represents q specific availability resources), the resource information needs to be solved to determine the decision variable for scheduling resource types. This decision variable is set as an n-dimensional 0-1 indicator vector.

[0069] For example, based on the availability information of non-overlapping resources (such as whether there is a water gun, whether there is foam, etc.), a 0-1 matrix is ​​used. Let represent the n different types of disaster relief vehicle resources, and q represent the q specific resources with fire extinguishing attributes, such as water guns and dry powder; the decision variable for scheduling these resources according to specific disaster relief rules is set as a 0-1 indicator vector. Among them, matrix Abbreviation Specific examples are as follows:

[0070]

[0071] Step S103: Establish the first optimization model based on the disaster relief mission requirements and the attribute information of non-overlapping emergency resources.

[0072] The first optimization model is used to solve the planning and decision-making of non-superimposed emergency resources.

[0073] Step S104: Establish a second optimization model based on environmental variable information, disaster relief task requirements, and attribute information of superimposed emergency resources, and perform sparse modeling on the second optimization model based on sparse representation.

[0074] The second optimization model is used to solve the planning decisions for superimposed emergency resources.

[0075] Step S105: Solve the first optimization model and the second optimization model after sparse modeling respectively to obtain an emergency resource planning scheme that meets the needs of disaster relief mission.

[0076] Specifically, step S105 above obtains the planning decisions for each non-overlapping resource by solving the first optimization model; obtains the planning decisions for each overlapping resource by solving the second optimization model after sparse modeling; and generates an emergency resource planning scheme for disaster relief mission requirements based on the planning decisions for each non-overlapping resource and the planning decisions for each overlapping resource.

[0077] By performing the above steps, the emergency resource planning method based on sparse representation provided in this embodiment of the invention divides emergency resources into superimposed and non-superimposed resources according to their ability to meet the needs of disaster relief tasks. Corresponding optimization models are then established and solved according to the actual disaster relief task requirements to obtain specific decision-making schemes. For non-superimposed resources, a conventional non-sparse representation method is used for modeling and solving to obtain feasible solutions. The superimposed resource problem is modeled and solved using sparse representation in convex optimization to address the problem of slow solution speed and inability to guarantee that the obtained solution is the globally optimal solution caused by integer programming modeling. Furthermore, it achieves the effect of resource planning with minimal scheduling resources, realizing accurate planning of emergency resources and reducing resource waste.

[0078] Specifically, in one embodiment, step S103 above includes the following steps:

[0079] Step S31: Based on the needs of the disaster relief mission, determine the objective rule information of the current disaster relief mission.

[0080] Among them, objective rule information is used to filter the emergency resources required for the current disaster relief mission. Specifically, based on the actual disaster relief mission, objective rule information in specific disaster relief operations is determined to form a rule vector, denoted as vector. For example, in fire rescue operations, it is necessary to cover objective information such as the types of resources that need to be dispatched to cover the problem area.

[0081] Step S32: Establish the first optimization model based on objective rule information and attribute information of non-overlapping emergency resources.

[0082] For example, by analyzing the dialogue content of fire alarms, relevant detailed information about the disaster scene can be inferred, such as fire type, smoke properties, fire area, fire location, and height, and then quantified. Based on professional knowledge and environmental information, basic disaster relief rules are determined, such as disaster relief methods, disaster relief costs, and rescue time for different types of fires. Then, objective rule information regarding non-overlapping resources in the field of fire emergency response is determined, such as fire type and disaster relief methods for different types of fires. A q-dimensional 0-1 indicator vector is used. This indicates the type of disaster relief resources to be used for specific fire types. For example, most flammable and combustible liquid fires can be extinguished with foam; fires involving water-reactive substances should be extinguished with carbon dioxide, dry powder, cement powder, dry sand, etc. It also includes decision-making resource information regarding non-overlapping resources in the field of fire emergency response. Decision variables and objective rules and information for disaster relief Modeling is performed to enable n types of disaster relief scheduling.

[0083] The types of disaster relief resources that can meet the needs of the type of fire on site.

[0084] Specifically, the first optimization model is represented by the following formula (1):

[0085]

[0086] Where A1 represents the first optimization model, This represents the utility function of resource planning. Let y represent the matrix of decision variables corresponding to non-superimposable resources. i Let y represent the decision variable corresponding to the i-th non-superimposable resource in the decision variable matrix, and y i ∈{0,1}, The above represents the availability information matrix of resources in the non-overlayable resource attribute information. This represents a rule vector composed of objective rule information.

[0087] Specifically, in one embodiment, the establishment of a second optimization model based on environmental variable information, disaster relief task requirements, and attribute information of superimposed emergency resources in step S104 includes the following steps:

[0088] Step S41: Based on environmental variable information and disaster relief mission requirements, determine the resource redundancy and the weight value corresponding to the resource cost of superimposed emergency resources.

[0089] Specifically, based on specific disaster situation information, existing fire-fighting resources, and actual business objectives, a mental estimation of decision-makers can be conducted to determine their preference weights for the resource utility function and the resource cost function. If people are trapped at a fire scene, and all efforts are required to complete the disaster relief regardless of cost, the decision weight of the resource cost function term is 0.

[0090] Step S42: Based on the attribute information of the superimposed emergency resources, determine the resource quantity and resource cost of each superimposed emergency resource.

[0091] Specifically, based on the actual resource planning problem, objective quantitative information for the specific disaster relief problem is determined, denoted as an m-dimensional column vector. For example, in fire relief operations, it is necessary to cover objective information such as the quantity of resources to be scheduled in the problem space, and to determine the cost of each resource, such as the price of a single resource.

[0092] Step S43: Based on the resource quantity and resource cost of each superimposed emergency resource, as well as the weight values ​​corresponding to the resource redundancy and resource cost of the superimposed emergency resources, establish a second optimization model.

[0093] Specifically, based on inferred relevant information from the disaster site and basic disaster relief rules, information on superimposed resources in the field of fire emergency response is determined, such as fire area, fire location, height, disaster relief consumption, and rescue time; expressed as an m-dimensional column vector. This indicates the specific rules for resource consumption in a fire scene, such as the amount of resources (water, dry powder) required for firefighting, and the height of the fire; based on a resource utility information matrix R of superimposed resources in the field of fire emergency response. m×n Resource Cost Information Matrix C p×n Decision variables Rules for the consumption of disaster relief resources in practical situations Modeling is performed to ensure that the n types of disaster relief resources scheduled can meet the disaster relief needs of the fire on site, while keeping the disaster relief cost as low as possible.

[0094] Furthermore, through the aforementioned decision-making resource information and variables regarding superimposed resources... and decision weight value vector Modeling the disaster relief problem involves optimizing the objective function, which can be divided into two parts: one is the resource redundancy function related to resource planning, denoted as... This enables decision-making resources to meet the utility information needs of specific disaster relief issues as much as possible. One term is the resource cost function. To minimize the cost of each resource; based on the actual resource planning problem, in order to achieve the goal of resource optimization, it is necessary to minimize the objective function. The established second optimization model is expressed by the following formula (2):

[0095]

[0096] Where A2 represents the second optimization model, Represents the resource redundancy function. Represents the resource cost function. Let x represent the matrix of decision variables corresponding to the superimposed resources. i Let x represent the decision variable corresponding to the i-th superimposed resource in the decision variable matrix, and x i ≥0 indicates that the utility information matrix of the resource in the superimposed resource attribute information, i.e., the aforementioned R. m×n, C represents the resource cost information matrix of the stackable resources, that is, the above-mentioned C p×n , α1 and α2 respectively represent the weight values corresponding to the resource redundancy and the resource cost. If the resource planning is completed without considering the cost according to the actual disaster relief problem, the decision weight value α1 of the resource cost function term is 0.

[0097] In practical applications, since in most cases it should be a discrete non-negative integer vector, the actual model should be a mixed integer programming model. However, its solution speed is slow and it cannot guarantee that the obtained solution is the global optimal solution. Therefore, in the modeling process of the present invention, is relaxed to a continuous non-negative real number vector, that is, take so as to introduce the modeling and solution algorithms in continuous optimization to improve the solution speed and ensure the global optimal solution is obtained.

[0098] Specifically, the resource redundancy function and the resource cost function are modeled by using sparse representation.

[0099] In practical applications, the specific resource redundancy function and the resource cost function are modeled by using sparse representation to achieve the purpose of resource planning with as few scheduling resources as possible. The forms of sparse modeling include but are not limited to L0 norm modeling, L1 norm modeling, L p (0 < p < 1) norm modeling, etc. In the embodiment of the present invention, the L1 norm in convex optimization is used for sparse modeling. Among them, the resource cost function can be taken as resource redundancy function

[0100] Based on the different optimization problems corresponding to superimposed and non-superimposed resources, corresponding problem-solving modules are used to solve them, resulting in resource planning decision schemes for specific disaster relief tasks. Specifically, the solution for the first optimization model of non-superimposed resources includes, but is not limited to, exhaustive methods, rule-based judgment methods, and branch-and-bound methods for small-scale combinatorial optimization problems; convex relaxation algorithms and heuristic algorithms for large-scale combinatorial optimization problems; the solution for the second optimization model of sparse representation of superimposed resources includes, but is not limited to, integrated software packages such as the convex optimization solution package CVX in MATLAB and the convex optimization solution package CVXOPT in Python; transforming it into linear programming or quadratic programming, and solving it using the simplex method, interior-point method, or classic Newton's method; for large-scale sparse optimization problems, efficient first-order algorithms are used, such as the primal-dual interior-point method, projected gradient descent method, proximal gradient method, homotopy algorithm, convexity relaxation method, iterative thresholding method, fast iterative threshold reduction algorithm (FISTA), Bregman separation method, augmented Lagrange multiplier method, and alternating direction multiplier method (ADMM). The above-described solution algorithm is used to solve the corresponding model, yielding the decision variables regarding non-superimposable resources. Feasible solutions and superimposed resource decision variables The optimal solution can be obtained by combining the information from the solution to the problem, and then a decision-making scheme for the specific resource planning problem can be obtained.

[0101] For example, the exhaustive method is used to solve formula (1) to obtain a feasible solution, i.e., the specific types of disaster relief resources. For formula (2), an integrated software package is used to solve for the optimal solution, i.e., the specific number and path of disaster relief resources. Thus, based on disaster situation information and fire resource information, specific decision-making schemes can be given regarding which disaster relief resources to call, how many to call, and the specific dispatch routes, to achieve refined scheduling management for cost reduction and efficiency improvement in fire emergency resource planning. Taking different fire scenarios as examples, the best resource planning scheme is finally selected as follows: Figures 3A to 3C The option with the highest overall score.

[0102] Furthermore, in practical applications, the first optimization model can be solved first to obtain the solution results for non-stackable resources. Then, it can be determined whether the solution results for non-stackable resources can meet the needs of the current disaster relief task. If they cannot meet the needs of the current disaster relief task, the second optimization model can be solved for the disaster relief task that cannot be met. This way, the solution results for both stackable and non-stackable resources can be used to meet the needs of the current disaster relief task, making the final emergency resource planning scheme more in line with the actual disaster relief scenario.

[0103] The emergency resource planning scheme based on sparse representation provided in this invention hierarchically divides emergency resources into stackable and non-stackable resources according to their attributes. Based on the actual resource planning problem, corresponding optimization models are established and solved to obtain specific decision-making schemes. The main invention is based on sparse representation in convex optimization to model and solve the problem of slow solution speed and inability to guarantee a globally optimal solution in integer programming modeling. It also achieves resource planning with minimal scheduling resources. Furthermore, it employs mental estimation for decision-makers to better determine the weight values ​​of resource functions in the model. For non-stackable resources, conventional non-sparse representation methods are used for modeling and solving to obtain feasible solutions. Combining these with the optimal solution for the stackable resource planning problem yields a complete resource planning decision scheme.

[0104] This invention also provides an emergency resource planning device based on sparse representation, such as... Figure 4 As shown, the emergency resource planning device based on sparse representation includes:

[0105] The acquisition module 101 is used to acquire environmental variable information of the target area where there is a need for disaster relief missions. The environmental variable information consists of environmental parameters related to the use of emergency resources. For details, please refer to the relevant description of step S101 in the above method embodiment, which will not be repeated here.

[0106] The first processing module 102 is used to determine the stackable and non-stackable emergency resources required for the target area based on environmental variable information. Stackable resources are used to characterize emergency resources whose ability to meet disaster relief mission requirements increases with the amount of resources added. For details, please refer to the relevant description of step S102 in the above method embodiments, which will not be repeated here.

[0107] The second processing module 103 is used to establish a first optimization model based on the disaster relief mission requirements and the attribute information of non-overlapping emergency resources. The first optimization model is used to solve the planning decision for non-overlapping emergency resources. For details, please refer to the relevant description of step S103 in the above method embodiment, which will not be repeated here.

[0108] The third processing module 104 is used to establish a second optimization model based on environmental variable information, disaster relief task requirements, and attribute information of superimposed emergency resources. It then performs sparse modeling on the second optimization model based on sparse representation. The second optimization model is used to solve the planning decisions for superimposed emergency resources. For details, please refer to the relevant description of step S104 in the above method embodiments, which will not be repeated here.

[0109] The fourth processing module 105 is used to solve the first optimization model and the second optimization model after sparse modeling, respectively, to obtain an emergency resource planning scheme that meets the needs of disaster relief tasks. For details, please refer to the relevant description of step S105 in the above method embodiments, which will not be repeated here.

[0110] The emergency resource planning device based on sparse representation provided in this embodiment of the invention is used to execute the emergency resource planning method based on sparse representation provided in the above embodiment. Its implementation method and principle are the same. For details, please refer to the relevant description of the above method embodiment, which will not be repeated here.

[0111] Through the collaborative efforts of the aforementioned components, the emergency resource planning device based on sparse representation provided in this embodiment of the invention divides emergency resources into superimposed and non-superimposed resources according to their varying capabilities to meet disaster relief mission requirements. It then establishes corresponding optimization models for each resource based on actual disaster relief mission needs, obtaining specific decision-making schemes. For non-superimposed resources, a conventional non-sparse representation method is used for modeling and solving to obtain feasible solutions. The superimposed resource problem is modeled and solved using sparse representation in convex optimization, addressing the issues of slow solution speed and inability to guarantee a globally optimal solution arising from integer programming modeling. Furthermore, it achieves resource planning with minimal scheduling resources, realizing precise emergency resource planning and reducing resource waste.

[0112] This invention also provides an emergency resource planning device based on sparse representation, such as... Figure 5 As shown, the emergency resource planning device based on sparse representation includes a processor 901 and a memory 902, wherein the processor 901 and the memory 902 can be connected via a bus or other means. Figure 5 Taking the example of a connection between China and Israel via a bus.

[0113] Processor 901 can be a Central Processing Unit (CPU). Processor 901 can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations of the above types of chips.

[0114] The memory 902, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the methods in the above method embodiments. The processor 901 executes various functional applications and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory 902, thereby implementing the methods in the above method embodiments.

[0115] The memory 902 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor 901, etc. Furthermore, the memory 902 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 902 may optionally include memory remotely located relative to the processor 901, and these remote memories may be connected to the processor 901 via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0116] One or more modules are stored in memory 902 and, when executed by processor 901, perform the methods described in the above method embodiments.

[0117] The specific details of the emergency resource planning equipment based on sparse representation can be understood by referring to the relevant descriptions and effects in the above embodiments, and will not be repeated here.

[0118] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The implemented program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0119] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. An emergency resource planning method based on sparse representation, characterized in that, The method includes: Obtain environmental variable information for target areas where disaster relief missions are needed, wherein the environmental variable information is environmental parameters related to the use of emergency resources; Based on the environmental variable information, the superimposed and non-superimposed emergency resources required for the target area are determined. The superimposed emergency resources are used to characterize the emergency resources whose ability to meet the disaster relief mission requirements increases with the amount of resources added. A first optimization model is established based on the disaster relief task requirements and the attribute information of non-stackable emergency resources. The first optimization model is used to solve the planning decision of non-stackable emergency resources. A second optimization model is established based on the environmental variable information, the disaster relief task requirements, and the attribute information of the superimposed emergency resources. The second optimization model is sparsely modeled based on sparse representation. The second optimization model is used to solve the planning decision of the superimposed emergency resources. Solve the first optimization model and the second optimization model after sparse modeling respectively to obtain an emergency resource planning scheme that meets the requirements of the disaster relief mission; The establishment of the first optimization model based on the disaster relief mission requirements and non-overlapping emergency resources includes: Based on the disaster relief mission requirements, objective rule information for the current disaster relief mission is determined, and the objective rule information is used to filter the emergency resources required for the current disaster relief mission. A first optimization model is established based on the objective rule information and the attribute information of non-overlapping emergency resources; The establishment of the second optimization model based on the environmental variable information, the disaster relief task requirements, and the attribute information of the superimposed emergency resources includes: Based on the environmental variable information and the disaster relief mission requirements, the resource redundancy and the weight values ​​corresponding to the resource cost of the superimposed emergency resources are determined. Based on the attribute information of superimposed emergency resources, determine the resource quantity and resource cost of each superimposed emergency resource; A second optimization model is established based on the resource quantity and resource cost of each superimposed emergency resource, as well as the weight values ​​corresponding to the resource redundancy and resource cost of the superimposed emergency resources. The sparse modeling of the second optimization model based on sparse representation includes: The resource redundancy function and resource cost function are modeled using sparse representation.

2. The method according to claim 1, characterized in that, The first optimization model is represented by the following formula: ， in, This represents the first optimization model. This represents the utility function of resource planning. This represents the matrix of decision variables corresponding to non-superimposable emergency resources. Represents the first in the decision variable matrix Decision variables corresponding to non-additive emergency resources, and , This represents a matrix of resource availability information within non-overlapping emergency resource attribute information. This represents a rule vector composed of objective rule information.

3. The method according to claim 1, characterized in that, The second optimization model is expressed by the following formula: A2= ， in, This represents the second optimization model. Represents the resource redundancy function. Represents the resource cost function. This represents the matrix of decision variables corresponding to superimposed emergency resources. Represents the first in the decision variable matrix There are several decision variables corresponding to superimposed emergency resources, and This represents a matrix of resource utility information that can be overlaid with emergency resource attribute information. This represents a resource cost information matrix that can be overlaid with emergency resources. These represent the weight values ​​corresponding to resource redundancy and resource cost, respectively.

4. The method according to claim 3, characterized in that, The step of solving the first optimization model and the second optimization model after sparse modeling respectively to obtain an emergency resource planning scheme that meets the requirements of the disaster relief mission includes: Solve the first optimization model to obtain the planning decisions for each non-overlapping emergency resource; Solve the second optimization model after sparse modeling to obtain the planning decisions for each superimposed emergency resource; An emergency resource planning scheme for the disaster relief task requirements is generated based on the planning decisions for each non-overlapping emergency resource and each overlapping emergency resource.

5. An emergency resource planning device based on sparse representation, applied to the emergency resource planning method based on sparse representation as described in any one of claims 1-4, characterized in that, The device includes: The acquisition module is used to acquire environmental variable information of target areas where there is a need for disaster relief missions. The environmental variable information is environmental parameters related to the use of emergency resources. The first processing module is used to determine the superimposed and non-superimposed emergency resources required by the target area based on the environmental variable information. The superimposed emergency resources are used to characterize the emergency resources whose ability to meet the disaster relief mission requirements increases with the amount of resources added. The second processing module is used to establish a first optimization model based on the disaster relief task requirements and the attribute information of non-overlapping emergency resources. The first optimization model is used to solve the planning decision of non-overlapping emergency resources. The third processing module is used to establish a second optimization model based on the environmental variable information, the disaster relief task requirements and the attribute information of the superimposed emergency resources, and to perform sparse modeling on the second optimization model based on sparse representation. The second optimization model is used to solve the planning decision of the superimposed emergency resources. The fourth processing module is used to solve the first optimization model and the second optimization model after sparse modeling, respectively, to obtain an emergency resource planning scheme that meets the requirements of the disaster relief mission.

6. An emergency resource planning device based on sparse representation, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the method as described in any one of claims 1-4.

Images