Intelligent recommendation method and system for multi-disaster emergency disposal process of gas storage

Abstract

The invention provides a gas storage multi-disaster emergency disposal process intelligent recommendation method and system. The method comprises the following steps: extracting an emergency mechanism and an emergency task; establishing an emergency mechanism-task graph network based on the emergency mechanism and the emergency task; acquiring game elements of the emergency mechanism based on the emergency mechanism-task graph network; building an emergency disposal flow of the emergency mechanism based on the game elements; different emergency event emergency disposal processes are input, an emergency mechanism carries out task strategy gaming, and emergency disposal processes under multiple disaster conditions are recommended; and evaluating the recommended emergency disposal process. According to the method, the existing emergency disposal process is summarized, the emergency disposal process under the condition of multiple disasters is constructed, and the weakness of the emergency capacity of multiple disasters on site and the deficiency of the emergency disposal scheme are made up.

Description

Technical Field

[0001] The present invention belongs to the technical field of emergency response process construction, and particularly relates to an intelligent recommendation method and system for the emergency response process of a gas storage cavern for multiple disasters. Background Art

[0002] As an important energy storage facility, the stable and safe operation of a gas storage cavern plays an important role in meeting energy demands and ensuring social and economic stability. However, during the operation of a gas storage cavern, it may face various sudden disaster situations, such as earthquakes, fires, gas leaks, etc. These disaster accidents may pose serious threats to equipment, personnel safety, and environmental safety. Due to the wide range of fields involved in gas storage caverns, sometimes a gas storage cavern may need to deal with two or more disaster situations simultaneously.

[0003] In the case of multiple disasters, the difficulty of emergency response increases significantly. First, different types of disasters may require different emergency measures. For example, the measures for dealing with a fire may be very different from those for dealing with a gas leak. Second, dealing with multiple disasters simultaneously may require prioritization to determine which disaster needs to be handled more urgently. This requires a quick and accurate assessment of the impacts of various disasters. Third, emergency response for multiple disasters may involve coordinated work among multiple departments, requiring effective organization and coordination. However, in current research, the research and practice of emergency response for multiple disasters are still relatively lacking.

[0004] Existing emergency response methods are usually based on expert judgment and experience, or obtained from a pre-established emergency plan library. These methods may be able to handle single disasters, but their limitations are revealed in the case of multiple disasters. First, expert judgment and experience may be affected by human subjective factors, and the formulation speed is slow, making it difficult to cope with the complexity of multiple disasters. Second, the emergency plans in the plan library may not cover all possible disaster combinations, or the recommended cases may not have a high similarity to the current disaster, thus affecting the quality of the disposal plan. Summary of the Invention

[0005] To solve at least one problem in the background art, the present invention proposes an intelligent recommendation method and system for the emergency response process of a gas storage cavern for multiple disasters.

[0006] To achieve the above object, the present invention adopts the following technical solutions:

[0007] An intelligent recommendation method for the emergency response process of a gas storage cavern for multiple disasters, comprising the following steps:

[0008] Extract emergency agencies and emergency tasks;

[0009] Establish an emergency agency-task graph network based on the emergency agencies and emergency tasks;

[0010] Obtain the game elements of emergency response agencies based on the emergency agency-task graph network;

[0011] Construct the emergency response process of emergency response agencies based on the game elements;

[0012] Input different emergency response processes for emergencies, and the emergency response agencies conduct task strategy games to recommend the emergency response process in the case of multiple disasters;

[0013] Evaluate the recommended emergency response process.

[0014] Preferably, extract emergency response agencies and emergency tasks, including the following steps:

[0015] Collect on-site emergency response cards and emergency plan texts of gas storage facilities to obtain a data set;

[0016] Extract emergency response agencies and emergency tasks from the data set;

[0017] Set each extracted emergency response agency as an independent agency unit, denoted as a∈A, where a is a single emergency response agency and A is the node set of emergency response agencies;

[0018] Set each extracted emergency task as an independent task unit, denoted as r∈R, where r is a single emergency task and R is the node set of emergency tasks;

[0019] Set the connection relationship between each extracted emergency response agency and task, denoted as l∈L, where l is a single connection relationship and L is the set of connection relationships.

[0020] Preferably, establish an emergency agency-task graph network based on emergency response agencies and emergency tasks, including:

[0021]

[0022] In the formula, G AR is the emergency agency-task graph network; A is the node set of emergency response agencies; R is the node set of emergency tasks; L 1 is a set of type I connection relationships; L 2 is a set of type II connection relationships; is the node of emergency response agency A in the emergency agency-task graph network; r j is the node of emergency task R in the emergency agency-task graph network; l ij is the emergency response agency a i and the emergency task r j between the connection relationship; is the connection relationship between emergency response agencies between.

[0023] Preferably, the game elements of the emergency agency are obtained based on the emergency agency-task graph network, including the following steps:

[0024] Construct the strategy space of the emergency agency;

[0025] Construct the utility function of the emergency agency;

[0026] Construct the information structure of the emergency agency.

[0027] Preferably, the strategy space includes:

[0028]

[0029] In the formula, represents the strategy space; where (r j …r j+n ) is a strategy for the emergency agency a i to execute the emergency task; there are n! kinds of strategies for the emergency agency to execute the task, indicating that the size of the strategy space of the emergency agency a i strategy space is n!.

[0030] Preferably, the utility function includes:

[0031]

[0032] In the formula, is the emergency disposal value of the emergency agency a i ; is the disposal value of the emergency task r j ; Δt is the disposal duration of the emergency task r j ; α t is the time correction coefficient, t represents time, and the later the task execution time t is, the smaller the correction coefficient is, and the lower the task disposal value is.

[0033] Preferably, the information structure includes:

[0034]

[0035] where M t is the information structure of the emergency disposal process at time t, is the information structure state at this moment.

[0036] Preferably, the emergency disposal process of the emergency agency is constructed based on the game elements, including:

[0037]

[0038] In the formula, ([A, A], T n-k ) represents at T n-kAt a moment, emergency information is transmitted from an emergency agency to another emergency agency; ([A, R], T n ) indicates that at time T n emergency information is transmitted from an emergency agency to an emergency task; ([R, A], T n+k ) indicates that at time T n+k emergency information is transmitted to an emergency agency through an emergency task; EP represents the emergency response process.

[0039] Preferably, the recommended emergency response process is evaluated, including:

[0040]

[0041] In the formula, is the sum of the times required for emergency agency a i to complete all emergency tasks ; is the emergency response value of emergency agency a i .

[0042] An intelligent recommendation system for multi-hazard emergency response processes in gas storage caverns, comprising:

[0043] An initial unit for establishing emergency agencies and emergency tasks;

[0044] A mapping unit for establishing an emergency agency-task graph network based on emergency agencies and emergency tasks;

[0045] A game unit for obtaining the game elements of emergency agencies based on the emergency agency-task graph network;

[0046] A construction unit for constructing the emergency response process of emergency agencies based on game elements;

[0047] A processing unit for inputting emergency response processes for different emergencies, and emergency agencies conduct task strategy games to recommend emergency response processes in the case of multiple hazards;

[0048] An evaluation unit for evaluating the recommended emergency response process.

[0049] Preferably, the initial unit includes:

[0050] A collection module for collecting on-site emergency response cards and emergency plan texts of gas storage caverns to obtain a data set;

[0051] An extraction module for extracting emergency agencies and emergency tasks from the data set;

[0052] A first setting module for setting each emergency agency as an independent agency unit, denoted as a ∈ A, where a is a single emergency agency and A is the node set of emergency agencies;

[0053] A second setting module, configured to set each emergency task as an independent task unit, denoted as r∈R, where r is a single emergency task and R is the set of nodes of emergency tasks;

[0054] A third setting module, configured to set the connection relationship between each emergency agency and task, denoted as l∈L, where l is a single connection relationship and L is the set of connection relationships.

[0055] Preferably, the game unit includes:

[0056] A first game module, configured to construct a strategy space for emergency agencies;

[0057] A second game module, configured to construct a utility function for emergency agencies;

[0058] A third game module, configured to construct an information structure for emergency agencies.

[0059] Advantages of the present invention:

[0060] 1. The present invention proposes an intelligent recommendation method for the emergency disposal process of multi-disaster types in gas storage caverns. First, collect the on-site emergency disposal cards and emergency plan text data sets of gas storage caverns, perform data cleaning and preprocessing on them, extract the emergency agencies and task sets of gas storage caverns, regard the specific emergency disposal tasks under different accident disasters as nodes, and regard the connection relationship between the agencies involved in the emergency process and the specific emergency tasks as edges to establish a gas storage cavern emergency agency-task graph network. Secondly, according to the emergency disposal processes of different disaster types in the emergency plan and emergency disposal cards, represent them as different matrices, and input the matrices into the emergency agency-task graph network to learn their evolution paths. Through the game rules of agency nodes, form the recommended results of the emergency process, and propose the utility evaluation index of the emergency process. This method constructs the emergency disposal process in the case of multi-disaster types by summarizing the existing emergency disposal processes, making up for the weakness of on-site multi-disaster emergency capabilities and the lack of emergency disposal plans;

[0061] 2. The present invention proposes an intelligent recommendation method and system for the emergency disposal process of multi-disaster types in gas storage caverns. It adopts a new emergency information automatic extraction technology and an emergency agency decision-making game optimization method, which can quickly formulate and recommend the most suitable disposal plan according to the existing disaster information after the disaster occurs, greatly improving the efficiency and quality of emergency disposal in the case of multi-disaster types.

[0062] Other features and advantages of the present invention will be described in the subsequent specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be realized and obtained by the structures pointed out in the specification and the drawings. Description of the Drawings

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0064] Figure 1 Shows the flowchart of an intelligent recommendation method for the multi-disaster emergency response process of a gas storage

[0065] Figure 2 Shows the emergency agency-task diagram network for the leakage of gas field water (left in the figure) and the leakage of natural gas in the southern injection main line (right in the figure);

[0066] Figure 3a Shows the information structure diagram of the emergency response process at time T1;

[0067] Figure 3b Shows the information structure diagram of the emergency response process at time T2;

[0068] Figure 3c Shows the information structure diagram of the emergency response process at time T3;

[0069] Figure 3d Shows the information structure diagram of the emergency response process at time T4;

[0070] Figure 3e Shows the information structure diagram of the emergency response process at time T5;

[0071] Figure 4 Shows the optimal multi-disaster emergency response flowchart;

[0072] Figure 5 Shows the block diagram of an intelligent recommendation system for the multi-disaster emergency response process of a gas storage of the present invention. Specific embodiments

[0073] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0074] Embodiment 1

[0075] An intelligent recommendation method for the multi-disaster emergency response process of a gas storage, as Figure 1As shown in the figure, it includes steps S1 - S6. S1: Establish an emergency agency and emergency tasks; S2: Based on the emergency agency and emergency tasks, establish an emergency agency - task graph network; S3: Based on the emergency agency - task graph network, obtain the game elements of the emergency agency; S4: Based on the game elements, construct the emergency response process of the emergency agency; S5: Input different emergency response processes for emergencies, and the emergency agency conducts task strategy games to recommend the emergency response process in the case of multiple disasters; S6: Evaluate the recommended emergency response process.

[0076] It should be noted that the above steps S1 - S6 are divided into three stages. Among them, S1 and S2 are the first stage, and their main function is to establish the emergency response flowchart network. S3 and S4 are the second stage, and their main function is to conduct the emergency response process game, that is, by importing the adjacency matrix of the emergency response process for different emergencies and the game rules of the emergency agency, to determine the best response strategy of the emergency agency. S5 and S6 are the third stage, and their main function is the recommendation and evaluation of the emergency response process.

[0077] Furthermore, the process in the first stage is as follows: 1) Pre - process the emergency response cards, emergency plan texts, extract emergency agencies and emergency tasks. The pre - processing is to eliminate text noise, and text noise refers to various interference and error information in the text; 2) Extract the connection relationship between the emergency agency and the task, and establish a graph network with the emergency agency and the emergency task as nodes and the connection relationship as edges. The following is a specific description in combination with S1 and S2.

[0078] In S1, it includes the following steps:

[0079] S101: Collect the on - site emergency response cards and emergency plan texts of the gas storage reservoir to obtain a data set; S102: Extract the emergency agency and emergency tasks from the data set; S103: Set each extracted emergency agency as an independent agency unit, denoted as a ∈ A, where a is a single emergency agency and A is the node set of the emergency agency; S104: Set each extracted emergency task as an independent task unit, denoted as r ∈ R, where r is a single emergency task and R is the node set of the emergency task; S105: Set the connection relationship between each extracted emergency agency and task, denoted as l ∈ L, where l is a single connection relationship and L is the connection relationship set.

[0080] It should be noted that in S101 - S105, Chinese word segmentation is achieved by removing stop words, constructing dictionaries, etc., and manual inspection is used to ensure that the emergency agency and emergency tasks after word segmentation are independent and complete, and the connection between the emergency agency and the task is accurate.

[0081] In S2, it is necessary to define the emergency agency-task graph network, as shown in Equation (1). The emergency agencies and emergency tasks are divided into two types of nodes, and the connection relationships between the nodes are divided into two types. One type is the connection relationship between the emergency agency and the emergency task, and the other type is the connection relationship between the emergency agencies.

[0082]

[0083] In the formula, G AR is the emergency agency-task graph network; A is the node set of the emergency agencies; R is the node set of the emergency tasks; L 1 is a set of one type of connection relationship; L 2 is a set of the second type of connection relationship;

[0084] is the node of the emergency agency A in the emergency agency-task graph network; r j is the node of the emergency task R in the emergency agency-task graph network; l ij is the connection relationship between the emergency agency a i and the emergency task r j ; is the connection relationship between the emergency agencies ;

[0085] Furthermore, the process in the second stage is as follows: 1) Establish the game rules of the emergency agencies in the emergency agency-task graph network; 2) Construct the strategy space of the emergency agencies, the utility function of the emergency tasks, and the information structure of the emergency response process; 3) Solve the best response strategy of the emergency agencies. Each iteration selects an agency and selects the best response strategy of this agency when the strategies of other agencies remain unchanged. The following is a detailed description in combination with S3 and S4.

[0086] In S3, it includes the following steps:

[0087] S301: Construct the strategy space of the emergency agencies. Specifically, in the emergency agency-task graph network, the emergency agencies are the execution units, and the emergency tasks are the execution objects. One emergency agency corresponds to one or more emergency tasks, and there are multiple strategies for the emergency agencies to execute tasks, that is, different task execution sequences. Define the strategy space of each emergency agency, that is, all possible sequences for an emergency agency to handle the emergency tasks, as shown in Equation (2).

[0088]

[0089] In the formula, represents the strategy space; where (r j …r j+n ) is the emergency agency a iA strategy for executing emergency tasks; there are n! strategies for the emergency agency to execute tasks, indicating that the size of the strategy space of emergency agency a i Strategy space is n!.

[0090] S302: Construct the utility function of the emergency agency. Specifically, the emergency agency needs to consume human resources and emergency supplies when executing emergency tasks and obtain certain benefits after completing the tasks. Define the utility function as the benefits obtained by the emergency agency when executing tasks, as shown in Equation (3).

[0091]

[0092] In the formula, is the emergency disposal value of emergency agency a i ; is the disposal value of emergency task r j ; Δt is the disposal duration of emergency task r j ; α t is the time correction coefficient, t represents time, and this coefficient is affected by the task execution time t. When the task execution time t is later, the correction coefficient is smaller and the task disposal value is lower.

[0093] S303: Construct the information structure of the emergency agency. Specifically, the information structure refers to the distribution range of emergency disposal information at a certain moment during the emergency disposal process. The information structure defines the information of other emergency agencies known by the emergency agency at this moment. During the emergency process, personnel tend to give priority to executing tasks in the direction they understand. Therefore, define the information structure as a matrix. When an emergency agency knows the emergency information of other agencies, it will give priority to executing this part of the decision during task decision-making, as shown in Equation (4).

[0094]

[0095] where M t is the information structure of the emergency disposal process at time t, is the information structure state at this moment. If agency a i knows the information of other emergency agencies, then otherwise it is 0.

[0096] In S4, set the emergency disposal process activation unit, that is, the emergency agency that performs the initial emergency action divides the construction of the emergency disposal process into two parts: emergency actions and emergency moments, as shown in Equation (5):

[0097]

[0098] In the formula, ([A, A], T n-k ) represents at Tn-k At a moment, emergency information is transmitted from an emergency agency to another emergency agency, that is, the emergency information is transmitted between emergency agencies; emergency information refers to the information generated by emergency agencies during the emergency process and having a promoting effect on the entire emergency process; ([A, R], T n ) indicates that at time T n At a moment, emergency information is transmitted from an emergency agency to an emergency task; ([R, A], T n+k ) indicates that at time T n+k At a moment, emergency information is transmitted to an emergency agency through an emergency task; The information transmission methods of the emergency disposal process include the above three categories. Therefore, the emergency disposal process is represented by the EP formula.

[0099] Furthermore, in the third stage, for the multi-hazard emergency process recommendation, the main steps include: 1) calculating the emergency response speeds of different strategies; 2) calculating the emergency utilities of different strategies; 3) recommending the emergency disposal process in the case of multi-hazards based on the emergency response time and emergency utility. The following will be described in combination with S6 respectively.

[0100] In S6, a verification index for the recommended result of the emergency disposal process is mainly proposed. The traditional method evaluates the time consumption, task completion rate, etc. of the emergency process unilaterally. Since this method realizes the recommendation of the emergency disposal process through the strategic game of emergency agencies and considers the utility index of task execution. Therefore, it is necessary to establish a comprehensive evaluation index through the completion value and emergency time consumption as the emergency process utility verification index, as shown in Equation (6).

[0101]

[0102] In the formula, is the sum of the times required for emergency agency a i to complete all emergency tasks ; is the emergency disposal value of emergency agency a i .

[0103] Example Two

[0104] Example Two evaluates according to the method of Example One for specific data.

[0105] Corresponding to S1, a certain gas storage emergency disposal card, a total of 47 emergency disposal cards for sudden events, is used as the data source.

[0106] Remove the stop words in the emergency response cards, such as "and", "according to", "based on", "to", "carry out", "at any time", "should", "such as", as well as some numbers, symbols, appointments and other words and characters that have no real meaning in the emergency response process. Establish a dictionary of emergency response agencies for gas storage facilities and a dictionary of emergency response tasks for gas storage facilities. Remove duplicates from the segmented Chinese words, and manually screen and confirm the validity of the words. Finally, 43 types of emergency response agencies and 235 emergency response tasks are obtained. Some of the emergency response agencies and emergency response tasks are shown in Table 1 and Table 2.

[0107] Table 1 Emergency Response Agencies for Gas Storage Facilities (Partial)

[0108] Central Control Room Safety Engineering Post General Affairs Post Emergency Leading Group Inspection and Maintenance Team Central Gathering Station Valve Chamber Marketing Department Fire Brigade Medical Rescue Team Production Operation Team Dispatch Room On-site Emergency Command Post Local Government Local Reinforcement Team Compressor Station Expert Group Security and Guard Team Comprehensive Support Team Emergency Rescue Team

[0109] Table 2 Emergency Response Tasks for Gas Storage Facilities (Partial)

[0110] Gas Monitoring Set Up Warning Area Close Inlet Valve Evacuate Crowd Rescue the Wounded Fire Fighting and Rescue Close Injection and Production Wells Close Production Process Close Inlet and Outlet Control Valves Open Bleed Valve Report Leak Point Location and Surrounding Conditions Spray Dilution Remote Shutdown Manually Open Ball Valve Confirm Ball Valve Status Check Pressure Gauge, Fixed Combustible Gas Detector

[0111] Corresponding to S2, select the emergency response cards for environmental pollution caused by gas field water leakage and natural gas leakage in the southern section of the gas injection pipeline. Based on the dictionary of emergency response agencies for gas storage facilities and the dictionary of emergency response tasks for gas storage facilities, extract the emergency response agencies and emergency response tasks involved in the emergency response processes of environmental pollution caused by gas field water leakage and natural gas leakage in the southern section of the gas injection pipeline, and extract the association rules between the emergency response agencies and emergency response tasks to form the emergency response agency-task graph network G AR , such as Figure 2 shown, the emergency response agencies and emergency response tasks are shown in Table 3.

[0112] Table 3 Emergency Response Agencies and Emergency Response Tasks

[0113]

[0114] Corresponding to S3, based on the emergency response agency-task graph network for gas field water leakage and natural gas leakage in the southern section of the gas injection pipeline of the gas storage facility, fuse the two graph networks, and calculate the strategy space and strategy combinations of the emergency response agencies in the fused graph network, as shown in Table 4.

[0115] Table 4 Emergency Response Agency Strategy Space

[0116]

[0117] For the emergency response cards for gas field water leakage and natural gas leakage in the southern section of the gas injection pipeline of the emergency gas storage facility, the emergency response plan for the gas storage facility and the emergency materials, set the emergency response task parameters as shown in Table 5, including task value, time consumption, human resources, emergency materials, and execution prerequisites.

[0118] Table 5 Emergency Response Task Parameters

[0119]

[0120] Part of the information structure of the emergency response process is as follows Figures 3a to 3e as shown, where Figures 3a - 3e are five different moments in sequence. When a 1 knows the state of a 2 at this emergency moment, then a 1 will give priority to a 2 when making a decision game, that is, will give priority to executing the task r 2 connected to a 2 .

[0121] Corresponding to S4, set the central control room as the initial emergency activation node, and calculate the emergency task utility of each emergency agency node under different strategies in the case of multiple disasters. In addition, it is considered that when the emergency agencies are directly connected, the time consumption can be ignored.

[0122] Corresponding to S5, after calculation, the best response strategies in the case of water leakage from the gas storage reservoir and natural gas leakage from the southern injection main pipeline, that is, the optimal multi-disaster emergency response plan is as Figure 4 shown in Table 6. When T = 6, all production processes and relevant valves at sites such as the central control room and compressor station have been closed, the fire brigade has arrived at the scene, set up a warning area, completed the evacuation of the crowd, and prevented the further expansion of the accident; when T = 12, all emergency tasks are completed.

[0123] Table 6 Optimal multi-disaster emergency response plan

[0124]

[0125]

[0126] Corresponding to S6, compare the emergency utility results of implementing the emergency response card for a single disaster with the utility results of each emergency response process under optimized multiple disasters, as shown in Table 7 for details.

[0127] Table 7 Evaluation results

[0128]

[0129] Through comparative analysis and calculation, the emergency response time of the original emergency response card is 16 unit times, and the emergency response process under optimized multiple disasters is 12 unit times, and the overall emergency utility of the emergency agencies has increased by 48.1%.

[0130] Example 3

[0131] As Figure 5As shown, an intelligent recommendation system for the multi-hazard emergency response process of a gas storage reservoir, corresponding to the method of Embodiment 1, includes: an initial unit for establishing emergency organizations and emergency tasks; a mapping unit for establishing an emergency organization-task graph network based on the emergency organizations and emergency tasks; a game unit for obtaining the game elements of the emergency organizations based on the emergency organization-task graph network; a construction unit for constructing the emergency response process of the emergency organizations based on the game elements; a processing unit for inputting different emergency response processes for emergencies, where the emergency organizations conduct task strategy games and recommend the emergency response process in the case of multiple hazards; and an evaluation unit for evaluating the recommended emergency response process.

[0132] Further, the initial unit includes: a collection module for collecting on-site emergency response cards and emergency plan texts of the gas storage reservoir to obtain a data set; an extraction module for extracting emergency organizations and emergency tasks from the data set; a first setting module for setting each emergency organization as an independent organization unit, denoted as a ∈ A, where a is a single emergency organization and A is the node set of the emergency organizations; a second setting module for setting each emergency task as an independent task unit, denoted as r ∈ R, where r is a single emergency task and R is the node set of the emergency tasks; and a third setting module for setting the connection relationship between each emergency organization and task, denoted as l ∈ L, where l is a single connection relationship and L is the connection relationship set.

[0133] Further, the game unit includes: a first game module for constructing the strategy space of the emergency organizations; a second game module for constructing the utility function of the emergency organizations; and a third game module for constructing the information structure of the emergency organizations.

[0134] Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir, characterized in that, it includes the following steps: Extract emergency response agencies and emergency tasks; Build an emergency response agency-task graph network based on the emergency response agencies and emergency tasks; Obtain the game elements of the emergency response agencies based on the emergency response agency-task graph network; Construct the emergency response process of the emergency response agencies based on the game elements; Input the emergency response processes for different emergencies, and the emergency response agencies conduct task strategy games to recommend the emergency response process in the case of multi-hazard types; Evaluate the recommended emergency response process.

2. The intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir according to claim 1, characterized in that, extracting emergency response agencies and emergency tasks includes the following steps: Collect on-site emergency response cards and emergency response plan texts of the gas storage reservoir to obtain a data set; Extract emergency response agencies and emergency tasks from the data set; Set each extracted emergency response agency as an independent agency unit, denoted as a ∈ A, where a is a single emergency response agency and A is the node set of emergency response agencies; Set each extracted emergency task as an independent task unit, denoted as r ∈ R, where r is a single emergency task and R is the node set of emergency tasks; Set the connection relationship between each extracted emergency response agency and task, denoted as l ∈ L, where l is a single connection relationship and L is the connection relationship set.

3. The intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir according to claim 1, characterized in that, building an emergency response agency-task graph network based on the emergency response agencies and emergency tasks includes: Where, G AR is the emergency agency-task graph network; A is the node set of the emergency agency; R is the node set of the emergency task; L 1 is a set of type-I connection relationships; L 2 is a set of type-II connection relationships; is the node of the emergency agency A in the emergency agency-task graph network; r j is the node of the emergency task R in the emergency agency-task graph network; l ij is the connection relationship between the emergency agency a i and the emergency task r j ; is the connection relationship between the emergency agencies .

4. The intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir according to claim 1, characterized in that, obtaining the game elements of the emergency response agencies based on the emergency response agency-task graph network includes the following steps: Construct the strategy space of the emergency response agencies; Construct the utility function of the emergency response agencies; Construct the information structure of the emergency response agencies.

5. The intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir according to claim 4, characterized in that, the strategy space includes: In the formula, represents the strategy space; where (r j …r j+n ) is a strategy for emergency agency a i to perform an emergency task; there are n! kinds of strategies for the emergency agency to perform tasks, indicating that the emergency agency a i strategy space has a size of n!.

6. The intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir according to claim 4, characterized in that, the utility function includes: In the formula, is the emergency response value of the emergency agency a i ; is the response value of the emergency task r j ; Δt is the response duration of the emergency task r j ; α t is the time correction coefficient, t represents time. The later the task execution time t is, the smaller the correction coefficient is, and the lower the task response value is.

7. The intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir according to claim 4, characterized in that, the information structure includes: Among which M t is the information structure of the emergency response process at time t, and is the information structure state at this time.

8. The intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir according to claim 1, characterized in that, constructing the emergency response process of the emergency response agencies based on the game elements includes: In the formula, ([A, A], T n-k represents that at time T n-k , the emergency information is transmitted from one emergency agency to another; ([A, R], T n ) represents that at time T n , the emergency information is transmitted from the emergency agency to the emergency task; ([R, A], T n+k ) represents that at time T n+k , the emergency information is transmitted to the emergency agency through the emergency task; EP represents the emergency response process.

9. The intelligent recommendation method for the emergency response process of multi-hazard types in a gas storage reservoir according to any one of claims 1-8, characterized in that, evaluating the recommended emergency response process includes: Wherein, is the emergency agency a i The sum of the time required to complete all emergency tasks ; is the emergency response value of the emergency agency a i .

10. An intelligent recommendation system for the emergency response process of multi-hazard types in a gas storage reservoir, characterized in that, it includes: An initial unit for establishing emergency response agencies and emergency tasks; A mapping unit for building an emergency response agency-task graph network based on the emergency response agencies and emergency tasks; A game unit for obtaining the game elements of the emergency response agencies based on the emergency response agency-task graph network; A construction unit for constructing the emergency response process of an emergency agency based on game elements; A processing unit for inputting different emergency response processes for emergencies, and the emergency agency conducts task strategy games to recommend the emergency response process in the case of multiple disasters; An evaluation unit for evaluating the recommended emergency response process.

11. A multi-disaster emergency response process intelligent recommendation system for a gas storage reservoir according to claim 10, characterized in that the initial unit includes: A collection module for collecting on-site emergency response cards and emergency plan texts of the gas storage reservoir to obtain a data set; An extraction module for extracting emergency agencies and emergency tasks from the data set; A first setting module for setting each emergency agency as an independent agency unit, denoted as a∈A, where a is a single emergency agency and A is the node set of emergency agencies; A second setting module for setting each emergency task as an independent task unit, denoted as r∈R, where r is a single emergency task and R is the node set of emergency tasks; A third setting module for setting the connection relationship between each emergency agency and task, denoted as l∈L, where l is a single connection relationship and L is the connection relationship set.

12. A multi-disaster emergency response process intelligent recommendation system for a gas storage reservoir according to claim 10 or 11, characterized in that the game unit includes: A first game module for constructing the strategy space of the emergency agency; A second game module for constructing the utility function of the emergency agency; A third game module for constructing the information structure of the emergency agency.

Images