A digital emergency plan management method and system

Abstract

The application provides a kind of digital emergency plan management method and system, it is related to emergency plan technical field, carries out monitoring scene information collection, constructs training scene set, sets safety event type information, carries out scene mapping to training scene set and obtains mapping result, constructs additional feature set, mapping result and additional feature set are used as matching feature for plan matching, executes the plan simulation under corresponding scene, outputs plan simulation result, and generates plan compensation data, the plan matching result is revised, and the revised plan is associated with corresponding mapping result and additional feature and is stored.The application solves the technical problem that the existing technology cannot effectively optimize and update the emergency plan, resulting in poor emergency plan management effect, realizes accurate control of event scene and event type in historical emergencies, and simulates and revises the plan according to historical events, to improve the technical effect of emergency plan management.

Description

Technical Field

[0001] This invention relates to the field of emergency response plan technology, specifically to a digital emergency response plan management method and system. Background Technology

[0002] Rapid urbanization has led to a rapid concentration of population and economy in cities. As the political, economic, cultural, and technological centers of a region, cities are characterized by concentrated populations, industries, wealth, buildings and structures, and various disasters. Once an accident or disaster occurs, it can cause enormous economic losses and casualties. In this context, the threat of public safety incidents to the lives of the people and the socio-economic situation becomes increasingly prominent. If, when an accident or disaster occurs, appropriate emergency plans are promptly activated and response procedures are followed based on the magnitude and development of the incident or disaster, the losses caused by the accident or disaster can be greatly reduced.

[0003] However, existing technologies fail to effectively optimize and update emergency response plans, resulting in poor emergency response plan management. Summary of the Invention

[0004] This application provides a digital emergency response plan management method and system to address the technical problem in the prior art that emergency response plans cannot be effectively optimized and updated, resulting in poor emergency response plan management.

[0005] In view of the above problems, this application provides a digital emergency plan management method and system.

[0006] In a first aspect, embodiments of this application provide a digital emergency response plan management method, the method comprising: collecting monitoring scenario information through big data to construct a training scenario set, wherein the training scenario set has a granular clustering identifier; setting safety event type information, mapping the training scenario set to scenarios using the safety event type information to obtain mapping results; constructing an additional feature set, using the mapping results and the additional feature set as matching features to perform emergency response plan matching; obtaining the emergency response plan matching result, performing emergency response plan simulation under the corresponding scenario using the emergency response plan matching result; outputting the emergency response plan simulation result and generating emergency response plan compensation data; correcting the emergency response plan matching result using the emergency response plan compensation data, and associating and storing the corrected emergency response plan with the corresponding mapping result and additional features.

[0007] Secondly, embodiments of this application provide a digital emergency response plan management system, the system comprising: a training scenario set construction module, which is used to collect monitoring scenario information through big data and construct a training scenario set, wherein the training scenario set has a granular clustering identifier; a scenario mapping module, which is used to set safety event type information and perform scenario mapping on the training scenario set according to the safety event type information to obtain mapping results; a plan matching module, which is used to construct an additional feature set and use the mapping results and the additional feature set as matching features to perform plan matching; a plan simulation module, which is used to obtain plan matching results and execute plan simulation under the corresponding scenario according to the plan matching results; a plan compensation data generation module, which is used to output plan simulation results and generate plan compensation data; and a plan matching result correction module, which is used to correct the plan matching results according to the plan compensation data and associate and store the corrected plan with the corresponding mapping results and additional features.

[0008] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0009] This application provides a digital emergency response plan management method, relating to the field of emergency response plan technology. The method involves collecting monitoring scenario information, constructing a training scenario set, defining safety event type information, mapping the training scenario set to obtain mapping results, constructing an additional feature set, using the mapping results and the additional feature set as matching features for emergency response plan matching, executing emergency response plan simulations under corresponding scenarios, outputting simulation results, generating emergency response plan compensation data, correcting the matching results, and associating and storing the corrected emergency response plan with the corresponding mapping results and additional features. This method solves the technical problem in existing technologies where emergency response plans cannot be effectively optimized and updated, resulting in poor emergency response plan management. It achieves precise control over event scenarios and types in historical emergencies, and simulates and corrects emergency response plans based on historical events, thereby improving the technical effect of emergency response plan management.

[0010] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0011] Figure 1 This application provides a schematic diagram of a digital emergency response plan management method.

[0012] Figure 2 This application provides a schematic diagram illustrating the process of constructing a training scenario set in a digital emergency response plan management method.

[0013] Figure 3 This application provides a schematic diagram of the emergency plan matching process in a digital emergency plan management method.

[0014] Figure 4 This application provides a schematic diagram of the structure of a digital emergency response plan management system.

[0015] Explanation of reference numerals in the attached diagram: Training scenario set construction module 10, scenario mapping module 20, pre-plan matching module 30, pre-plan simulation module 40, pre-plan compensation data generation module 50, and pre-plan matching result correction module 60. Detailed Implementation

[0016] This application provides a digital emergency plan management method to address the technical problem in the prior art where emergency plans cannot be effectively optimized and updated, resulting in poor emergency plan management.

[0017] Example 1

[0018] like Figure 1 As shown in the figure, this application provides a digital emergency response plan management method, the method including:

[0019] Step S100: Collect monitoring scene information through big data and construct a training scene set, wherein the training scene set has a granular clustering identifier;

[0020] Specifically, the digital emergency response plan management method provided in this application is applied to a digital emergency response plan management system. An emergency response plan refers to emergency management, command, and rescue plans for sudden events such as natural disasters, major accidents, environmental pollution, and man-made damage, and is generally based on comprehensive disaster prevention planning.

[0021] Traditional surveillance systems include front-end cameras, transmission cables, and video surveillance platforms. They utilize video surveillance technology to detect and monitor protected areas, display and record real-time images, and retrieve and display historical images. By connecting the video surveillance platform to big data, information on monitoring scenarios in areas prone to emergencies can be obtained, such as scenarios arising from floods or mudslides during natural disasters, or scenarios arising from acute industrial poisoning accidents during major incidents. The requirements information in the contingency plan database is analyzed to determine the level of detail and comprehensiveness of the data. Higher detail corresponds to finer granularity, and lower detail corresponds to coarser granularity. This generates granularity constraint data, which is then used to cluster monitoring scenario information, constructing a training scenario set. The granularity constraint data serves as the identifier for the granular clustering.

[0022] Step S200: Set security event type information, and perform scene mapping on the training scene set according to the security event type information to obtain the mapping result;

[0023] Specifically, a security incident refers to an event that negatively impacts society due to natural disasters, human-caused events, or other factors. Security incidents can be broadly categorized into four types: natural disasters, accidents, public health incidents, and social security incidents. More specifically, natural disasters include floods and droughts, meteorological disasters, earthquakes, and geological disasters; accidents include transportation accidents, public facility and equipment accidents, and production safety accidents; public health incidents include food safety and occupational hazards; and social security incidents include terrorist attacks and economic security incidents. These refined security incident types are used as security incident type information. Since a single scenario may generate multiple security incident types, a mapping relationship is constructed between the training scenario and multiple security incident types for any training scenario in the training scenario set. Based on the security incident type information, security incident simulations are performed on the training scenarios in the training scenario set. For example, based on the training scenario set, digital twin technology is used to perform simulation modeling using scenario simulation tools. This fully utilizes physical models, sensors, and monitoring scene information to complete the mapping in virtual space, thereby simulating the entire process of a security incident occurring in the training scenario and obtaining the mapping results.

[0024] Step S300: Construct an additional feature set, and use the mapping result and the additional feature set as matching features to perform pre-plan matching;

[0025] Specifically, emergencies are often accompanied by multiple disasters. Many natural disasters, especially high-level and intense ones, often trigger a chain of other disasters, a phenomenon known as a disaster chain. The process of a disaster is often very complex; sometimes a single disaster can be caused by several factors, or one factor can simultaneously cause several different disasters. For example, in a fire, if a container ruptures and flammable or explosive media leaks out and mixes with air, it can cause a secondary explosion if it comes into contact with an open flame. The instantaneous high temperature and strong shock wave generated by the secondary explosion can cause burns, serious injury, or death to people, and cause severe damage to equipment and buildings. Situations that could cause secondary or even multiple disasters are considered as additional disasters. Features of these additional disasters are extracted to construct an additional feature set. This additional feature set is then added to the mapping results and used together as matching features. For example, if a fire occurs in a production workshop and there are pressure vessels containing flammable and explosive media, such as compressed natural gas or liquefied petroleum gas, when matching contingency plans, the production workshop, the fire, and the pressure vessels containing flammable and explosive media should all be used as matching features. The matched contingency plan should not only extinguish the fire but also treat the pressure vessels to prevent them from bursting due to overpressure or overloading, which could lead to a secondary explosion.

[0026] Step S400: Obtain the pre-plan matching result, and execute the pre-plan simulation in the corresponding scenario based on the pre-plan matching result;

[0027] Specifically, the contingency plan matching results include the matching of emergency organization management and command systems, emergency engineering rescue and support systems, mutual support systems, supply guarantee systems, and emergency teams that need to be dispatched under the triggering scenario of a safety incident. The corresponding scenarios are simulated in a computer simulation model, and the contingency plan matching results are added to the simulated scenarios. Through contingency plan simulation, the entire process of changes in various variables in the model is observed, and the model is used to reproduce the essential impact of the actual scenario contingency plan matching results on the safety incident scenario.

[0028] Step S500: Output the simulation results of the contingency plan and generate the contingency plan compensation data;

[0029] Specifically, the simulation results of the contingency plan represent the impact of the contingency plan on the safety incident scenario in the simulation model. To achieve better emergency response, reduce personal, property, and environmental losses caused by accidents, and control the development of accidents, contingency plan compensation is applied to multiple aspects of the contingency plan based on the simulation results. Although the handling of safety incidents can be guaranteed when the contingency plan is matched, the safety incident scenario is a complex environment with too many unknowns. To prevent personal, property, and environmental losses caused by emergencies, it is necessary to consider the potential hazards in each aspect. When the contingency plan fails, action compensation measures are implemented, such as increasing personnel deployment for insufficient human resources, increasing rescue supplies for insufficient material resources, refining action steps, and optimizing action equipment to ensure the complete resolution of the accident. The adjustments and supplements to the contingency plan simulation results are used as contingency plan compensation data.

[0030] Step S600: Correct the matching result of the contingency plan using the contingency plan compensation data, and associate and store the corrected contingency plan with the corresponding mapping result and additional features.

[0031] Specifically, the compensation data from the emergency response plan is used as an alternative solution for each stage of the plan matching result, thus correcting the matching result and making the emergency response plan more complete and reliable. The corrected emergency response plan, along with the safety event triggering scenario and additional features, are stored as correlated information, forming a data organization based on the relationship between safety event, additional features, and revised plan. Since the plan database contains a large number of emergency response plans and corresponding multiple safety event triggering scenarios, this correlated storage allows for rapid matching of event types and accident scenarios, thereby improving the matching speed of emergency response plans.

[0032] Furthermore, such as Figure 2 As shown, step S100 of this application further includes:

[0033] Step S110: Obtain the requirement information from the contingency plan database;

[0034] Step S120: Perform requirement parsing on the requirement information to generate granular constraint data;

[0035] Step S130: Cluster the monitoring scene information collection results using the granularity constraint data, construct a training scene set based on the clustering results, and use the granularity constraint data as the granularity clustering identifier.

[0036] Specifically, the contingency plan database is designed to develop specific emergency plans and on-site response schemes for all possible accidents and hazards, clearly defining the responsibilities of relevant departments and personnel at each stage: before, during, and after an incident. The database is designed to comprehensively reflect the information from the monitored scenario and the relationships between these information points, and to facilitate various data retrieval and processing operations. Data processing requirements analysis is conducted on the contingency plan database to identify the various data items and structures corresponding to the emergency plans from the perspective of data organization and storage. The required data access operations for each data item are defined, forming a data dictionary to generate granular constraint data.

[0037] Granularity represents the scope and level of detail in data analysis. For example, monitoring scene information can be simply categorized into four types: natural disasters, major accidents, environmental pollution, and man-made damage. Further refinement can be achieved by subdividing each category, resulting in more than ten different scenarios. Since the purpose of establishing a contingency plan database is to identify the disasters and response methods corresponding to specific scenarios, a granular and detailed classification method is used to constrain the monitoring scene information collection results. Based on the granularity-constrained data, the monitoring scene information collection results are divided into different clusters. This maximizes the similarity of monitoring scenes within the same cluster while maximizing the differences between monitoring scenes outside the same cluster. In other words, it clusters similar monitoring scenes together as much as possible and separates different types of monitoring scenes as much as possible. A training scene set is then constructed based on the clustering results.

[0038] Furthermore, such as Figure 3 As shown, step S300 of this application further includes:

[0039] Step S310: Collect and obtain accident site information, and evaluate the information quality based on the accident site information to obtain a quality label;

[0040] Step S320: Perform accident analysis on the accident location information to obtain accident type information, and perform feature extraction using the accident type information and the accident location information to obtain initial feature extraction results, wherein the initial feature extraction results are labeled with feature values;

[0041] Step S330: Correct the feature value identifier using the quality identifier, and sort the features of the initial feature extraction results according to the corrected feature value identifier to obtain the feature sorting result;

[0042] Step S340: Perform emergency response plan matching based on the feature sorting results, the accident type information, and the associated stored data.

[0043] Specifically, accident site information refers to the location of a disaster or accident, including geographical location and related geographical information. For example, for an earthquake, geological information is obtained, while for a factory explosion, temperature and climate information are acquired. Information quality is evaluated based on the correlation between the accident site information and the accident itself; a higher correlation results in a higher information quality evaluation. Accident analysis is then performed based on the basic causes, occurrence process, exposed problems, and resulting hazards identified in the accident site information. For instance, the cause of the accident is used to preliminarily determine whether it is a natural disaster or a man-made accident, and the severity of the accident is assessed based on the resulting hazards to obtain accident type information.

[0044] The total number of documents N in the accident type and accident location information is statistically analyzed. The frequency of each word appearing in positive documents, negative documents, no positive documents, and no negative documents is calculated. The chi-square value of each word is calculated, and each word is sorted from largest to smallest chi-square value. The top k words are selected as the initial feature extraction results, where k is the feature dimension. The chi-square value is a statistic in nonparametric tests, primarily used in nonparametric statistical analysis. It is a key indicator in the chi-square test, a widely used hypothesis testing method for count data. It falls under the category of nonparametric tests and mainly compares the correlation between two or more sample rates and two categorical variables. Simply put, it compares the degree of agreement or goodness of fit between theoretical and actual frequencies. The percentage of information quality is obtained based on quality indicators and used as a ranking coefficient. The product of the ranking coefficient and the chi-square value is calculated, and the results are used to re-rank the features to obtain the final feature ranking.

[0045] Furthermore, step S320 of this application also includes:

[0046] Step S321: Obtain an initial feature association value mapping set through the accident type information;

[0047] Step S322: Extract accident-related features from the accident site information and generate basic accident feature values;

[0048] Step S323: Perform association value mapping matching based on the accident association feature extraction results and the initial feature association value mapping set, and generate the feature value identifier based on the association value mapping matching results and the accident feature base values.

[0049] Specifically, based on accident type information, a mapping relationship between the accident location and one or more corresponding accident types is obtained, thus acquiring an initial feature association value mapping set. Words associated with the accident are extracted from the accident location information, and the frequency of each word is calculated and sorted from high to low. The top j words are extracted as accident association features, and the accident location information corresponding to these j words is used as the basic accident feature values. The extracted j high-frequency words are then matched with the data in the initial feature association value mapping set to obtain the association value mapping matching results, thereby obtaining feature value identifiers.

[0050] Furthermore, step S600 of this application also includes:

[0051] Step S610: Obtain road segment distance information based on the accident site information and warehouse location information;

[0052] Step S620: Read real-time traffic data, and perform time fitting based on the real-time traffic data and the road segment distance information to obtain the time fitting result;

[0053] Step S630: Perform emergency plan matching and correction based on the time fitting results.

[0054] Specifically, the geographical location of the accident site is obtained based on the accident site information. The warehouse, being a supply depot, is identified using its location information. Navigation data is used to obtain the distance, route, and road conditions from the warehouse to the accident site, serving as road segment distance information. The normal travel time is calculated based on this road segment distance information. Real-time traffic data, such as real-time traffic congestion and road maintenance information, is obtained to calculate the opening time or detour time. The normal travel time is then added to the opening time or detour time to obtain a fitted time from the warehouse to the accident site. Emergency plans are then matched and adjusted. For example, if the distance is short and the travel time is short, it indicates that supplies can be replenished at any time, and lightweight equipment can be prioritized to improve rescue speed. If the distance is long and the travel time is long, it indicates that supplies are difficult to replenish, requiring sufficient supplies to be carried and a continuous supply of subsequent rescue materials to achieve the best rescue effect.

[0055] Furthermore, step S630 of this application includes:

[0056] Step S631: Obtain the route information, perform a traffic stability evaluation based on the historical traffic data of the route, and obtain the traffic stability evaluation result;

[0057] Step S632: Construct a deviation time window based on the traffic stability evaluation results and the road segment distance information;

[0058] Step S633: Adjust the time fitting result using the time deviation window, and use the time adjustment result to match and correct the emergency plan.

[0059] Specifically, the traffic route information mainly represents the road capacity between the warehouse and the accident site. This refers to the maximum number of vehicles that can pass through a road facility at a specific point or cross-section per unit time under normal road, traffic, control, and operational quality requirements. Road capacity is a road performance indicator, measuring its ability to manage traffic. It reflects both the maximum capacity of a road to handle traffic and the limit of vehicle traffic it can bear under specified characteristics. If the road capacity is strong, there will be virtually no traffic congestion and smooth traffic flow, indicating strong stability. Conversely, if the road capacity is weak, there will be frequent traffic jams and roadworks, resulting in unstable travel times, indicating poor stability. This is how the traffic stability evaluation result is obtained.

[0060] The time required for traffic jams and road repairs is calculated based on the traffic stability evaluation results. This time is used as a deviation window, and the time required for normal driving, traffic opening time or detour time, and the time required for traffic jams or road repairs are added together to obtain the time adjustment results. The emergency plan is then revised again based on the time adjustment results to make the time allocation of the emergency plan more accurate.

[0061] Furthermore, step S330 of this application includes:

[0062] Step S331: Set the feature value filtering threshold;

[0063] Step S332: Perform feature filtering on the modified feature value identifier using the feature value filtering threshold to extract the initial feature extraction result.

[0064] Step S333: The initial feature extraction results of the application are sorted according to the order of the corrected feature value identifiers to obtain the feature sorting results.

[0065] Specifically, as the number of features increases, many data mining algorithms require more time and resources. Therefore, setting a feature value filtering threshold reduces the number of features, thereby improving algorithm speed and reducing resource usage. By selecting an appropriate threshold, features greater than the threshold are retained, while those less than the threshold are deleted. This removes features most likely independent of the label and irrelevant to the classification objective. The filtered features are then used as the initial feature extraction results. The initial feature extraction results are then reordered according to the corrected feature value labels to obtain the feature ranking results.

[0066] Example 2

[0067] Based on the same inventive concept as the digital emergency response plan management method in the foregoing embodiments, such as Figure 4 As shown, this application provides a digital emergency response plan management system, the system comprising:

[0068] Training scenario set construction module 10 is used to collect monitoring scenario information through big data and construct a training scenario set, wherein the training scenario set has a granular clustering identifier.

[0069] Scene mapping module 20 is used to set security event type information, and to perform scene mapping on the training scene set according to the security event type information to obtain mapping results;

[0070] The pre-plan matching module 30 is used to construct an additional feature set, and use the mapping result and the additional feature set as matching features to perform pre-plan matching.

[0071] The contingency plan simulation module 40 is used to obtain contingency plan matching results and execute contingency plan simulation under the corresponding scenario based on the contingency plan matching results.

[0072] The contingency plan compensation data generation module 50 is used to output the contingency plan simulation results and generate contingency plan compensation data.

[0073] The contingency plan matching result correction module 60 is used to correct the contingency plan matching result through the contingency plan compensation data, and to associate and store the corrected contingency plan with the corresponding mapping result and additional features.

[0074] Furthermore, the system also includes:

[0075] The requirement information acquisition module is used to obtain requirement information from the contingency plan database;

[0076] The requirement parsing module is used to parse the requirement information and generate granular constraint data;

[0077] The clustering constraint module is used to perform clustering constraints on the monitoring scene information collection results through the granular constraint data, construct a training scene set based on the clustering results, and use the granular constraint data as the granular cluster identifier.

[0078] Furthermore, the system also includes:

[0079] The information quality evaluation module is used to collect accident site information and evaluate the information quality based on the accident site information to obtain a quality label;

[0080] The accident analysis module is used to analyze the accident location information to obtain accident type information, and to extract features using the accident type information and the accident location information to obtain an initial feature extraction result, wherein the initial feature extraction result is labeled with feature values.

[0081] The feature sorting module is used to correct the feature value identifier using the quality identifier, sort the features of the initial feature extraction results according to the corrected feature value identifier, and obtain the feature sorting result.

[0082] The emergency plan matching module is used to match emergency plans based on the feature sorting results, the accident type information, and associated stored data.

[0083] Furthermore, the system also includes:

[0084] The mapping set acquisition module is used to obtain an initial feature association value mapping set through the accident type information;

[0085] The accident-related feature extraction module is used to extract accident-related features from the accident location information and generate basic accident feature values.

[0086] The association value mapping matching module is used to perform association value mapping matching based on the accident association feature extraction results and the initial feature association value mapping set, and generate the feature value identifier based on the association value mapping matching results and the accident feature base values.

[0087] Furthermore, the system also includes:

[0088] The road segment distance information acquisition module is used to obtain road segment distance information based on the accident site information and warehouse location information;

[0089] The time fitting module is used to read real-time traffic data, perform time fitting based on the real-time traffic data and the road segment distance information, and obtain the time fitting result.

[0090] The emergency plan matching and correction module is used to perform emergency plan matching and correction based on the time fitting results.

[0091] Furthermore, the system also includes:

[0092] The traffic stability evaluation module is used to obtain traffic route information, perform traffic stability evaluation based on the historical traffic data of the traffic route, and obtain traffic stability evaluation results.

[0093] The deviation time window construction module is used to construct a deviation time window based on the traffic stability evaluation results and the road segment distance information.

[0094] The time adjustment module is used to adjust the time fitting result through the deviation time window, and to perform emergency plan matching and correction based on the time adjustment result.

[0095] Furthermore, the system also includes:

[0096] The filter threshold setting module is used to set the feature value filter threshold;

[0097] The feature filtering module is used to perform feature filtering on the modified feature value identifier through the feature value filtering threshold, and extract the initial feature extraction result of the application.

[0098] The sorting module is used to sort the initial feature extraction results of the application by correcting the order of feature value identifiers, and obtain the feature sorting result.

[0099] Through the foregoing detailed description of a digital emergency response plan management method, those skilled in the art can clearly understand the digital emergency response plan management method and system in this embodiment. As for the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and relevant parts can be referred to the method section description.

[0100] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A digital emergency plan management method, characterized by, The method comprises: Monitoring scene information collection through big data, and constructing a training scene set, wherein the training scene set has a granularity clustering identifier; Setting safety event type information, respectively mapping the training scene set through the safety event type information, and obtaining a mapping result; Building an additional feature set, taking the mapping result and the additional feature set as matching features, and performing plan matching; Obtaining a plan matching result, and performing plan simulation under a corresponding scene through the plan matching result; Outputting a plan simulation result, and generating plan compensation data; Modifying the plan matching result through the plan compensation data, and associating and storing the modified plan with the corresponding mapping result and additional features; Constructing a training scene set comprises: Obtaining demand information of a plan database; Demand analysis is performed on the demand information to generate granularity constraint data; The clustering constraint of the monitoring scene information collection result is performed through the granularity constraint data, the training scene set is constructed based on the clustering result, and the granularity constraint data is taken as the granularity clustering identifier; Plan matching comprises: Obtaining accident site information, and performing information quality evaluation based on the accident site information to obtain a quality identifier; Accident analysis is performed on the accident site information to obtain accident type information, feature extraction is performed through the accident type information and the accident site information, and an initial feature extraction result is obtained, wherein the initial feature extraction result has a feature value identifier; The feature value identifier is modified through the quality identifier, and feature sorting of the initial feature extraction result is performed according to the modified feature value identifier to obtain a feature sorting result; Emergency plan matching is performed according to the feature sorting result, the accident type information, and the associated storage data.

2. The method of claim 1, wherein, The method comprises: Obtaining an initial feature correlation value mapping set through the accident type information; Accident correlation feature extraction is performed on the accident site information, and an accident feature basic value is generated; Correlation value mapping matching is performed according to the accident correlation feature extraction result and the initial feature correlation value mapping set, and the feature value identifier is generated based on the correlation value mapping matching result and the accident feature basic value.

3. The method of claim 1, wherein, The method comprises: Obtaining road section distance information according to the accident site information and warehouse location information; Reading real-time traffic data, performing time fitting based on the real-time traffic data and the road section distance information, and obtaining a time fitting result; Performing emergency plan matching modification according to the time fitting result.

4. The method of claim 3, wherein, The method comprises: Obtaining traffic route information, performing traffic stability evaluation based on historical traffic data of the traffic route, and obtaining a traffic stability evaluation result; Building a deviation time window based on the traffic stability evaluation result and the road section distance information; Performing time adjustment on the time fitting result through the deviation time window, and performing emergency plan matching modification through the time adjustment result.

5. The method of claim 1, wherein, The method comprises: Setting a feature value screening threshold; Performing feature screening on the modified feature value identifier through the feature value screening threshold, and extracting an application initial feature extraction result; The application initial feature extraction result is corrected with feature value identification sequence sorting to obtain the feature sorting result.

6. A digital emergency plan management system, characterized by, The system comprises: A training scene set construction module is configured to collect monitoring scene information through big data and construct a training scene set, wherein the training scene set has a granularity cluster identification; A scene mapping module is configured to set security event type information, map the training scene set through the security event type information respectively, and obtain a mapping result; A preplan matching module is configured to construct an additional feature set, take the mapping result and the additional feature set as matching features, and perform preplan matching; A preplan simulation module is configured to obtain a preplan matching result, and perform preplan simulation under a corresponding scene through the preplan matching result; A preplan compensation data generation module is configured to output a preplan simulation result and generate preplan compensation data; A preplan matching result correction module is configured to correct the preplan matching result through the preplan compensation data, and store the corrected preplan in association with the corresponding mapping result and additional features; A demand information acquisition module is configured to obtain demand information of a preplan database; A demand analysis module is configured to perform demand analysis on the demand information and generate granularity constraint data; A cluster constraint module is configured to perform cluster constraint on the monitoring scene information collection result through the granularity constraint data, construct a training scene set based on the cluster result, and take the granularity constraint data as the granularity cluster identification; An information quality evaluation module is configured to collect accident site information and perform information quality evaluation based on the accident site information to obtain a quality identification; An accident analysis module is configured to perform accident analysis on the accident site information to obtain accident type information, perform feature extraction through the accident type information and the accident site information, and obtain an initial feature extraction result, wherein the initial feature extraction result has feature value identification; A feature sorting module is configured to correct the feature value identification through the quality identification, perform feature sorting of the initial feature extraction result according to the corrected feature value identification, and obtain a feature sorting result; An emergency preplan matching module is configured to perform emergency preplan matching according to the feature sorting result, the accident type information, and the associated storage data.

Images