Underwater manifold leakage emergency response decision support system based on heterogeneous information

Abstract

The invention belongs to the technical field of underwater production system information processing and intelligent operation and maintenance, and discloses an underwater manifold leakage emergency response decision support system based on heterogeneous information, which comprises a data acquisition and integration module, a leakage consequence evaluation module and an emergency response scheme generation module. Firstly, a data acquisition and integration module acquires multi-source heterogeneous data in real time, including underwater sensor data, environmental data and historical leakage accident data, and cleans, formats and stores the data; secondly, the leakage consequence evaluation module predicts the leakage diffusion range, evaluates the accident influence degree and quantifies the accident level based on computational fluid mechanics leakage diffusion simulation and in combination with an SVM model; and finally, an emergency response scheme generation module calculates an index weight by adopting an entropy weight method based on a leakage evaluation result and historical emergency plan data, performs priority ranking on emergency schemes in combination with CoCoSo, selects an optimal emergency response scheme, and supports real-time adjustment and optimization.

Description

Technical Field

[0001] The present invention belongs to the technical field of information processing and intelligent operation and maintenance of underwater production systems, and relates to an underwater manifold leakage emergency response decision support system based on heterogeneous information. Background Art

[0002] The underwater manifold is one of the key equipment of the underwater production system, mainly used to collect and distribute fluids from multiple wellheads and transport them to the underwater Christmas tree or subsea pipeline. However, due to its long-term exposure to the deep-sea environment, facing complex factors such as high pressure, low temperature, highly corrosive media, and marine dynamic environment, the underwater manifold is extremely vulnerable to factors such as structural fatigue, corrosion, erosion, and connection seal failure, which may lead to leakage accidents.

[0003] The leakage of the underwater manifold will not only cause waste of oil and gas resources, but may also lead to serious marine environmental pollution and even affect the stability of the entire underwater production system. Therefore, a fast, accurate, and scientific emergency response decision support system is crucial for reducing the losses of leakage accidents and improving the safety of underwater production systems.

[0004] Currently, the emergency response to underwater manifold leakage accidents mainly relies on manual experience and historical cases, but the existing technologies still have the following deficiencies:

[0005] (1) Insufficient information integration ability

[0006] The existing emergency response decision-making system relies on a single data source (such as monitoring sensor data, expert experience, or historical cases), and cannot fully integrate and utilize multi-source heterogeneous information (including real-time sensor data, environmental data, historical accident data, etc.), resulting in limited decision support capabilities.

[0007] (2) Low accuracy of leakage impact assessment

[0008] Most traditional leakage impact assessment methods use empirical formulas or simple numerical simulations, without fully considering the influence of the marine environment (such as ocean currents, temperature, depth) on leakage diffusion, and lack accurate leakage diffusion simulation and dynamic prediction capabilities.

[0009] (3) Insufficient optimization ability of emergency response plans

[0010] The existing emergency response plans are usually preset static plans, and fail to optimize and sort the plans in combination with the leakage assessment results, resulting in the lack of pertinence and adaptability of the emergency response plans during actual implementation.

[0011] (4) Lack of scientificity and timeliness in the decision-making process

[0012] The existing emergency response decision-making system relies more on manual analysis and fails to effectively utilize intelligent methods such as machine learning and optimization algorithms to improve decision-making efficiency, resulting in a slow emergency response speed and difficulty in meeting the needs of quickly responding to leakage accidents.

[0013] To address the above problems, the present invention proposes an underwater pipeline leak emergency response decision support system based on heterogeneous information, which uses computational fluid dynamics (CFD) leak diffusion simulation, support vector machine (SVM) accident impact assessment, and optimization algorithm-driven emergency response plan generation to achieve accurate prediction, intelligent decision-making, and dynamic optimization of underwater pipeline leak accidents. Summary of the Invention

[0014] The purpose of the present invention is to provide an underwater pipeline leak emergency response decision support system based on heterogeneous information, which solves the problems of insufficient information integration ability, low decision support accuracy, and weak response timeliness in the prior art. By integrating and analyzing multi-source heterogeneous information, it provides scientific, comprehensive, and efficient decision support for emergency response.

[0015] The technical solution of the present invention:

[0016] An underwater pipeline leak emergency response decision support system based on heterogeneous information includes the following modules:

[0017] (1) Data acquisition and integration module;

[0018] (1.1) Collect multi-source data information, including:

[0019] Real-time sensor data: Monitor the operating parameters of the underwater pipeline, including pressure, temperature, and flow rate;

[0020] Environmental data: Obtain ocean current, temperature, depth, and sea condition information in the leakage area;

[0021] Historical data: Call the fault records, emergency response records, and expert experience of past leakage accidents;

[0022] (1.2) Data storage; Integrate multi-source data information and unify it into structured, semi-structured, and unstructured forms, and store it in a distributed database;

[0023] (2) Leakage consequence assessment module;

[0024] (2.1) According to the real-time sensor data and environmental data, construct a CFD-based leak diffusion simulation model to predict the pollution diffusion range and diffusion rate in the leakage area in real time, and obtain the leak diffusion simulation results;

[0025] (2.2) Based on the leak diffusion simulation results and multi-source data information, use support vector machine to evaluate the consequences and impacts of the leakage accident to obtain the leakage assessment results;

[0026] (3) Emergency response plan generation module;

[0027] (3.1) Based on the leakage assessment results, combined with historical data and environmental data, using the combined compromise sorting method (CoCoSo method), comprehensively considering factors such as time, cost, and effect, prioritize multiple preset leakage emergency plans, and provide a comprehensive score of the plans for decision-makers' reference;

[0028] (3.2) Support users to adjust the response plan in real time and dynamically update the evaluation and sorting results.

[0029] The present invention discloses an underwater pipeline leakage emergency response decision support system based on heterogeneous information, aiming to solve problems such as insufficient information integration ability, low accuracy of leakage impact assessment, and insufficient emergency response plan optimization ability in the prior art. The system consists of a data acquisition and integration module, a leakage consequence assessment module, and an emergency response plan generation module. First, the data acquisition and integration module obtains multi-source heterogeneous data in real time, including underwater sensor data (pressure, temperature, flow rate), environmental data (ocean current, depth, temperature), and historical leakage accident data, and cleans, formats, and stores the data. Secondly, the leakage consequence assessment module predicts the leakage diffusion range, evaluates the accident impact degree, and quantifies the accident level based on the computational fluid dynamics leakage diffusion simulation and in combination with the SVM model. Finally, the emergency response plan generation module calculates the index weights using the entropy weight method based on the leakage assessment results and historical emergency plan data, and combines CoCoSo to prioritize the emergency plans, selects the optimal emergency response plan, and supports real-time adjustment and optimization.

[0030] The beneficial effects of the present invention include:

[0031] (1) Multi-source heterogeneous information fusion to improve the decision-making data support ability

[0032] Adopt the data acquisition and integration module to obtain underwater sensor data (such as pressure, temperature, flow rate), environmental data (such as ocean current, temperature, depth), and historical accident cases in real time, and realize the standardized storage and processing of multi-source data.

[0033] Through a data-driven approach, avoid information loss caused by a single data source, and improve the scientificity and comprehensiveness of decision-making.

[0034] (2) Precise leakage diffusion simulation to improve the accuracy of accident impact assessment

[0035] Adopt the Navier-Stokes equation and the VOF multiphase flow model to construct a leakage diffusion prediction model based on CFD, which can accurately simulate the diffusion path, diffusion range, and impact area of the leaked substance.

[0036] Combined with dynamic environmental factors such as ocean currents, temperature, and pressure, it provides more accurate prediction results of leakage impacts and improves the evaluation accuracy.

[0037] (3) Optimized emergency response plan to improve the efficiency of emergency response

[0038] Combined with the accident assessment results and historical emergency plan data, using the combined compromise ranking method, optimize the ranking of different emergency plans to ensure the selection of the optimal emergency response plan.

[0039] Calculate the weights of each plan through the entropy weight method, avoid the deviation impact of a single index, and ensure the scientificity and rationality of decision-making. Description of the Drawings

[0040] Figure 1 It is a schematic diagram of the underwater pipeline leakage emergency response decision support system based on heterogeneous information provided by the present invention. Detailed Implementation Manner

[0041] The following details the specific structure and implementation process of this solution through specific embodiments and drawings.

[0042] As Figure 1 shown, in an embodiment of the present invention, an underwater pipeline leakage emergency response decision support system based on heterogeneous information is disclosed. Through core modules such as data collection and integration, leakage consequence assessment, and emergency response plan generation, it realizes a rapid and accurate response to underwater pipeline leakage incidents.

[0043] The three core modules of the present invention are constituted as follows:

[0044] (1) Data collection and integration module: Responsible for the collection, preprocessing, storage, and formatting of multi-source data, providing basic data for subsequent analysis.

[0045] (2) Leakage consequence assessment module: Based on CFD leakage simulation and SVM model, calculate the leakage diffusion range, influence degree, and quantify the severity of the leakage accident.

[0046] (3) Emergency response plan generation module: Based on the leakage assessment results and historical emergency plan data, use the CoCoSo method to evaluate, rank different emergency response plans, and dynamically optimize the emergency plan.

[0047] The present invention adopts a data bus architecture to ensure the logical smoothness of the data flow from data collection to leakage assessment and then to decision support, and realizes efficient information interaction between different modules.

[0048] The following details the specific implementation manners of each module of the present invention:

[0049] The data acquisition and integration module is as follows:

[0050] (1.1) Collect multi-source data information;

[0051] (1.1.1) Real-time sensor data, structured data:

[0052] Through the pressure sensors, temperature sensors, and flow meters installed on the underwater manifold, collect the pressure change, temperature change, and flow anomaly in real time;

[0053] Format of real-time sensor data:

[0054] S t ={P t ,T t ,F t ,V t}

[0055] Among them, P t is the leakage pressure, Pa; T t is the temperature, °C; F t is the flow rate, m 3 / s; V t is the leakage velocity, m / s;

[0056] (1.1.2) Environmental data, semi-structured data:

[0057] Obtain ocean currents, temperature, depth, and sea conditions through the ocean monitoring system, which affect the leakage diffusion behavior;

[0058] Format of environmental data:

[0059] E t ={C t ,G t ,H t}

[0060] Among them, C t is the ocean current rate, m / s; G t is the depth of the leakage point, m; H t is the environmental temperature, °C;

[0061] (1.1.3) Historical data, unstructured data:

[0062] Use natural language processing (NLP) to extract key parameters and format them into a historical case library:

[0063] B ={(L i ,D i ,Y i ,R i ,C i ,H i )|i = 1,2,...,N}

[0064] Among them, N is the number of cases of underwater pipeline leak accidents in the historical case base, L i is the leakage location in the i-th historical case, D i is the leakage range, Y i is the leakage impact degree level, R i is the relevant response measure, C i is the ocean current speed at the time of the accident, H i is the environmental temperature at the time of the accident;

[0065] (1.2) Data storage;

[0066] The real-time sensor data is stored in a time-series database (Time-Series DB);

[0067] The environmental data is stored in a relational database (MySQL);

[0068] The historical case base data is stored in a NoSQL database (MongoDB).

[0069] The leakage consequence assessment module is as follows:

[0070] The specific implementation process of step (2.1) is as follows:

[0071] (2.1.1) Data input;

[0072] The leakage consequence assessment module receives the standardized data from the data acquisition and integration module, including:

[0073] Real-time sensor data S t 、Environmental data E t 、Historical case base B;

[0074] (2.1.2) Calculation method;

[0075] Leakage diffusion simulation

[0076] The Navier-Stokes equation is used to describe the motion of the leaked fluid:

[0077]

[0078] Among them, u is the fluid velocity vector, m / s; t is the time, s; P is the pressure, Pa; ρ is the fluid density, kg / m 3 ; ν is the dynamic viscosity, m 2 / s; is the gradient operator, representing the partial derivative with respect to the spatial coordinates;

[0079] The VOF multiphase flow model is used to simulate the oil and gas leakage process and calculate the pollution diffusion range D of the leakage areat :

[0080] D t = f(P t , V t , C t , H t )

[0081] where D t is the pollution diffusion range of the leakage area, referring to the affected area of the leaked fluid underwater, m 3 ; P t is the leakage pressure, i.e., the pressure difference between the inside and outside of the manifold, which affects the leakage speed, Pa; V t is the leakage speed, which determines the initial motion state of the leaked substance, m / s; C t is the ocean current rate, which affects the diffusion direction and range of the leaked substance in seawater, m / s; H t is the environmental temperature, which affects the evaporation, dissolution, and diffusion behavior of the leaked gas or liquid, °C;

[0082] (2.2) Obtain the leakage assessment result;

[0083] (2.2.1) Accident impact assessment

[0084] Based on the support vector machine, classify the accident impact level Y using the diffusion range D, ocean current rate C, and environmental temperature H, divided into levels 0 - 5, with level 0 to level 5 evolving from no impact to extreme impact;

[0085] Extract the data sets X and Y from the historical case library B, each X i = (D i , C i , H i ) represents the key features of a leakage accident; train the support vector machine with labeled data, X is the input feature, and Y is the classification label; use the RBF kernel function to improve the classification accuracy:

[0086] K(X i , X j ) = exp(-γ||X i - X j || 2 )

[0087] where γ is the kernel function parameter; X i , X j are the input features; ||X i - X j || 2 represents the Euclidean distance, measuring the feature X i and X jDistance in the feature space; the goal is to find the optimal hyperplane f(x):

[0088]

[0089] where X t is the input sample to be predicted, and α i is the Lagrange multiplier used to optimize the hyperplane of the support vector machine, enabling correct classification of points of different classes and maximizing the margin; b is the bias term used to adjust the decision hyperplane to better adapt to the data distribution;

[0090] Input the data X t =(D t , C t , H t ) of the new leakage event, calculate the classification based on the SVM model, and finally output the evaluation result Y t of the accident impact level, with a value range of {0, 1, 2, 3, 4, 5}, as follows:

[0091] Y t = argmax f(X t )

[0092] where argmax represents the category corresponding to the maximum value, ensuring that the classification result corresponds to the correct accident impact level;

[0093] (2.2.2) Leakage assessment result output;

[0094] (2.2.2.1) Leakage diffusion prediction map;

[0095] (2.2.2.2) Leakage accident impact level.

[0096] The emergency response plan generation module is as follows:

[0097] (3.1) Data input;

[0098] The leakage assessment result obtained by the leakage consequence assessment module: R = {D t , Y t , C t , H t};

[0099] The data of the historical case library B obtained by the data collection and integration module;

[0100] (3.2) Pre - plan priority calculation;

[0101] (3.2.1) Determination of alternative emergency response plans:

[0102] According to the leakage assessment result and historical emergency response plan data, determine m alternative emergency response plans P:

[0103] P = {P1, P2,..., P m}

[0104] (3.2.2) Construct evaluation indicators:

[0105] Time dimension: response time, plugging time;

[0106] Cost dimension: equipment cost, labor cost;

[0107] Effect dimension: plugging success rate, leakage control effect;

[0108] (3.2.3) Use the entropy weight method to calculate the weights of each indicator:

[0109]

[0110] Among them, p ij is the standardized score of the i-th emergency plan on the j-th indicator, and the calculation method is:

[0111]

[0112] Among them, x ij is the initial score data of the i-th emergency plan on the j-th indicator in the historical case library B, given by the decision maker;

[0113] (3.2.4) Use the CoCoSo algorithm to calculate the scores of the emergency plans:

[0114] Calculate the comprehensive scores of each alternative emergency plan, and select the plan with the highest score as the optimal plan;

[0115] The first step is to calculate the weighted sum, addition model:

[0116]

[0117] Among them, S i is the weighted score of the i-th alternative emergency plan; w j is the weight of the j-th indicator, calculated by the entropy weight method;

[0118] The second step is to calculate the geometric mean, multiplication model:

[0119]

[0120] Among them, G i is the geometric mean score of the i-th alternative emergency plan;

[0121] The third step is to calculate the final score, weighted combination:

[0122]

[0123] Among them, Q i is the final score of the i-th alternative emergency plan;

[0124] (3.3) Result output

[0125] (3.3.1) Optimal emergency response plan;

[0126] (3.3.2) Users can adjust the index weights, and the system will update the sorting results in real time; after the sensor data is updated, the system will automatically adjust the evaluation results and optimize the sorting of the emergency plans.

[0127] The embodiments of the present invention have the following beneficial effects compared with the prior art:

[0128]

[0129] The present invention significantly improves the emergency response ability of underwater pipeline leakage accidents through the data-driven + intelligent optimization method. Its main innovation points include:

[0130] (1) Multi-source heterogeneous information fusion to construct a data-driven emergency decision support system.

[0131] (2) The combination of CFD simulation and SVM to achieve accurate leakage assessment and accident impact prediction.

[0132] (3) Driven by an optimization algorithm to ensure the rationality and optimality of the emergency response plan.

[0133] In summary, the present invention breaks through the limitations of traditional emergency response methods, can significantly improve the response speed, decision-making accuracy and emergency treatment effect of underwater pipeline leakage accidents, is applicable to multiple fields such as offshore oil and gas production, subsea equipment operation and maintenance, and marine environment emergency management, and has broad application prospects and technical value.

[0134] At this point, those skilled in the art should recognize that although many exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications that conform to the principles of the present invention can still be directly determined or derived from the content disclosed in the present invention without departing from the spirit and scope of the present invention. Therefore, the scope of the present invention should be understood and recognized to cover all these other variations or modifications.

Claims

1. An underwater manifold leakage emergency response decision support system based on heterogeneous information, characterized in that: Contains the following modules: (1) Data collection and integration module; (1.1) Collect multi-source data information, including: Real-time sensor data: monitors the operating parameters of the underwater manifold, including pressure, temperature, and flow; Environmental data: Obtain information on ocean currents, temperature, depth, and sea conditions in the spill area; Historical data: call on failure records, emergency response records and expert experience of past leakage accidents; (1.2) Data storage: Integrate multi-source data information into structured, semi-structured, and unstructured forms and store them in distributed databases; (2) Leakage consequence assessment module; (2.1) Based on real-time sensor data and environmental data, a CFD-based leakage diffusion simulation model is constructed to predict the pollution diffusion range and diffusion rate in the leakage area in real time, which is the leakage diffusion simulation result; (2.2) Based on the leakage diffusion simulation results and multi-source data information, the consequences and impacts of the leakage accident are evaluated using support vector machines to obtain leakage assessment results; (3) Emergency response plan generation module; (3.1) Based on the leakage assessment results, combined with historical data and environmental data, a combined compromise ranking method is used to comprehensively consider time, cost, and effect factors, prioritize the various preset leakage emergency plans, and provide a comprehensive score for the plans for reference by decision makers; (3.2) Support users to adjust response plans in real time and dynamically update evaluation and ranking results.

2. The underwater manifold leakage emergency response decision support system according to claim 1 is characterized in that: The data collection and integration modules are as follows: (1.1) Collect multi-source data information; (1.1.1) Real-time sensor data, structured data: The pressure sensors, temperature sensors and flow meters installed in the underwater manifold can collect pressure changes, temperature changes and flow anomalies in real time; The format of real-time sensor data is: S t ={P t ,T t ,F t ,V t } Among them, P t is the leakage pressure, Pa; T t is the temperature, ℃; F t is the flow rate, m 3 / s; V t Leakage velocity, m / s; (1.1.2) Environmental data, semi-structured data: Obtain ocean currents, temperature, depth, and sea conditions through ocean monitoring systems to influence leak diffusion behavior; The format of the environment data is: E t ={C t ,G t ,H t } Among them, C t is the ocean current velocity, m / s; G t is the depth of the leak point, m; H t is the ambient temperature, °C; (1.1.3) Historical data, unstructured data: Natural language processing is used to extract key parameters and format them into a historical case library: B={(L i ,D i ,Y i ,R i ,C i ,H i )∣i=1,2,...,N} Where N is the number of underwater manifold leakage accidents in the historical case database, L i is the leakage location in the i-th historical case, D i is the leakage range, Y i is the leakage impact level, R i For relevant response measures, C i is the ocean current speed when the accident occurred, H i is the ambient temperature when the accident occurred; (1.2) Data storage; Real-time sensor data is stored in a time series database; Environmental data is stored in a relational database; The historical case library data is stored in the NoSQL database.

3. The underwater manifold leakage emergency response decision support system according to claim 2 is characterized in that: The leakage consequence assessment module is as follows: The specific implementation process of step (2.1) is as follows: (2.1.1) Data input; The spill consequence assessment module receives standardized data from the data collection and integration module, including: Real-time sensor data t 、Environmental data E t , Historical Case Library B; (2.1.2) Calculation method; Leakage Diffusion Simulation The Navier-Stokes equation is used to describe the motion of the leaking fluid: Where u is the fluid velocity vector, m / s; t is time, s; P is pressure, Pa; ρ is the fluid density, kg / m 3 ; ν is the dynamic viscosity, m 2 / s; is the gradient operator, which represents the partial derivative with respect to the spatial coordinates; The VOF multiphase flow model is used to simulate the oil and gas leakage process and calculate the pollution diffusion range D of the leakage area. t : D t =f(P t ,V t ,C t ,H t ) Among them, D t is the pollution diffusion range of the leakage area, which refers to the underwater impact area of ​​the leakage fluid, m 3 ;P t is the leakage pressure, i.e. the pressure difference between the inside and outside of the manifold, which affects the leakage rate, Pa; V t is the leakage velocity, which determines the initial motion state of the leaked material, m / s; C t is the ocean current velocity, which affects the diffusion direction and range of the leaked substances in the seawater, m / s; H t is the ambient temperature, which affects the evaporation, dissolution and diffusion behavior of the leaked gas or liquid, ℃; (2.2) Obtain leakage assessment results; (2.2.1) Accident impact assessment Based on the support vector machine, the diffusion range D, ocean current velocity C and ambient temperature H are used to classify the accident impact level Y into 0-5 levels, where 0 to 5 evolve from no impact to extreme impact; Extract data sets X and Y from historical case library B, Each X i =(D i ,C i ,H i ) represents the key features of a leakage accident; label data is used to train the support vector machine, X is the input feature, and Y is the classification label; the RBF kernel function is used to improve the classification accuracy: K(X i ,X j )=exp(-γ||X i -X j || 2 ) Among them, γ is the kernel function parameter; X i , X j is the input feature; ||X i -X j || 2 Represents the Euclidean distance, which measures the feature X i and X j Distance in feature space; the goal is to find the optimal hyperplane f(x): Among them, X t is the input sample to be predicted, α i is the Lagrange multiplier, which is used to optimize the hyperplane of the support vector machine so that points of different categories are correctly classified and the interval is maximized; b is the bias term, which is used to adjust the decision hyperplane to make it more suitable for the data distribution; Enter data X for new leak event t =(D t ,C t ,H t ), calculate the classification based on the SVM model, and finally output the assessment result Y of the accident impact level t , the value range is {0,1,2,3,4,5}, as shown below: Y t =argmaxf(X t ) Among them, argmax represents the category corresponding to the maximum value, ensuring that the classification result corresponds to the correct accident impact level; (2.2.2) Leakage assessment result output; (2.2.2.1) Leakage diffusion prediction diagram; (2.2.2.2) Impact level of leakage accident.

4. The underwater manifold leakage emergency response decision support system according to claim 1, characterized in that: The emergency response plan generation module is as follows: (3.1) Data input; The leakage consequence assessment module obtains the leakage assessment result: R = {D t ,Y t ,C t ,H t }; The data of historical case library B obtained by the data collection and integration module; (3.2) Calculation of plan priority; (3.2.1) Determination of alternative emergency plans: According to the leakage assessment results and historical emergency plan data, m alternative emergency plans P are determined: P={P1,P2,...,P m } (3.2.2) Construct evaluation indicators: Time dimension: response time, blocking time; Cost dimension: equipment cost, labor cost; Effect dimension: plugging success rate, leakage control effect; (3.2.3) The entropy weight method is used to calculate the weight of each indicator: Among them, p ij is the standardized score of the i-th emergency plan on the j-th indicator, calculated as: Among them, x ij is the initial score data of the i-th emergency plan on the j-th indicator in the historical case library B, given by the decision maker; (3.2.4) CoCoSo algorithm calculates the plan score: Calculate the comprehensive scores of each alternative emergency plan and take the plan with the highest score as the optimal plan; The first step is to calculate the weighted sum, the additive model: Among them, S i is the weighted score of the i-th alternative emergency plan; w j is the weight of the jth indicator, calculated by the entropy weight method; The second step is to calculate the geometric mean and multiplicative model: Among them, G i is the geometric mean score of the i-th alternative emergency plan; The third step is to calculate the final score and weighted combination: Among them, Q i is the final score of the i-th alternative emergency plan; (3.3) Result output (3.3.1) Optimal emergency response plan; (3.3.2) Users can adjust the indicator weights, and the system will update the ranking results in real time. After the sensor data is updated, the system will automatically adjust the evaluation results and optimize the ranking of emergency plans.

Images