Large model method and system for emergency command aided decision making

Abstract

The invention provides a large model method and system for emergency command aided decision making, and the method comprises the steps: obtaining safety monitoring data from a to-be-monitored platform, carrying out the related retrieval of the safety monitoring data based on a large model, obtaining the data related to the production and life data, and storing the data, thereby achieving the comprehensive obtaining of the data based on the large model. The method comprises the following steps: monitoring safety monitoring data in real time based on a large model to obtain risk accident data, performing key data association analysis on the risk accident data, peripheral life data, peripheral emergency data and related law and regulation data, and determining possible secondary accidents according to an analysis result. Based on data association, the hidden relationship between risk accidents is deeply explored to reveal potential risks, association between data is deeply mined, accurate accident information is provided for emergency command assistance, and an accident report document is automatically generated based on risk accident data and specific data conditions of possible secondary accidents. And auxiliary decision making is provided for commanders.

Description

Technical Field

[0001] The present invention relates to the technical field of auxiliary decision making, and in particular to a large model method and system for auxiliary decision making of emergency command. Background Art

[0002] In the fields of enterprise safety production digital platform, enterprise safety production standardization management system, emergency management comprehensive business platform, visual emergency rescue smart platform, urban safety risk dynamic distribution map system, chemical park safety risk digital twin intelligent management and control platform, when coal mine accidents, fire accidents, traffic accidents, gas leakage accidents, hazardous chemicals accidents, industrial and trade accidents and other risk accidents occur, emergency command decision-making assistance large models are usually configured to provide auxiliary decision-making for commanders.

[0003] The existing emergency command decision-making support model uses semantic analysis, images, videos, voice recognition and other technologies to clean and aggregate different data and model algorithm services to form intelligence services, thereby extracting content to form text. However, it is unable to deeply explore the implicit relationship between risk accidents, reveal potential risk points, and independently write report documents. Summary of the invention

[0004] The present invention provides a large model method and system for emergency command decision-making assistance, which are used to solve the problems raised in the background technology.

[0005] A large model method for emergency command decision support, comprising:

[0006] S1: Obtain the safety monitoring data from the platform to be monitored, and perform relevant retrieval on the safety monitoring data based on the big model, obtain the surrounding life data related to the production and life data, surrounding emergency data and relevant laws and regulations data for storage;

[0007] S2: Monitor safety monitoring data in real time based on the big model to obtain risk accident data;

[0008] S3: Conduct key data correlation analysis on risk accident data, surrounding life data, surrounding emergency data and relevant laws and regulations data, and determine possible secondary accidents based on the analysis results;

[0009] S4: Automatically generate accident report documents based on risk accident data and specific data of possible secondary accidents.

[0010] Preferably, in S1, obtaining security monitoring data from the platform to be monitored includes:

[0011] Obtain sensor data and daily record data of the platform to be monitored;

[0012] Preprocessing the sensor data and daily recorded data to obtain standard data;

[0013] Based on the safety monitoring standard, the standard data is specifically processed to obtain safety monitoring data.

[0014] Preferably, in S1, the safety monitoring data is searched based on the big model to obtain surrounding life data, surrounding emergency data and relevant laws and regulations data related to the production and life data for storage, including:

[0015] Based on the monitoring area scope of the security monitoring data, determine the data acquisition area for related retrieval, and extract type keywords, name keywords and content keywords in the security monitoring data based on the data type, data name and data content;

[0016] Performing semantic association on the type keywords, name keywords and content keywords, and integrating the type keywords, name keywords and content keywords according to the semantic association result to obtain target keywords;

[0017] Acquire a relevant data directory from a data acquisition area based on the large model, and perform a relevant search on the target data directory based on the target keyword to obtain initial search data;

[0018] Acquire the safety monitoring data to monitor the safety accident type, and obtain the hazard source characteristics corresponding to the safety accident type;

[0019] Performing data feature matching on the hazard source feature and the initial search data to obtain a feature matching result, and obtaining a secondary search term based on the feature matching result;

[0020] Performing a horizontal search on the relevant data directory based on the secondary search term to obtain first supplementary search data;

[0021] Performing semantic expansion on the secondary search term to obtain an extended search term, performing an extended search on the initial search data based on the extended search term, and obtaining an in-depth search term from the initial search data according to the extended search result;

[0022] Performing an in-depth search on the relevant data directory based on the in-depth search term to obtain second supplementary search data;

[0023] Integrate the initial search data, the first supplementary search data, and the second supplementary search data, and obtain relevant data according to the integration result;

[0024] Based on the data content characteristics, the relevant data are divided into surrounding life data, surrounding emergency data and relevant laws and regulations data and saved.

[0025] Preferably, the obtaining of secondary search terms based on feature matching results includes:

[0026] The search data corresponding to the hazard source feature is obtained from the feature matching result, and the keywords of the search data are extracted. The hazard source feature is combined with the keywords to obtain a secondary search term.

[0027] Preferably, the method of dividing the relevant data into surrounding life data, surrounding emergency data and relevant laws and regulations data based on data content characteristics and saving the data includes:

[0028] Establish data content features based on specific data content of surrounding life, Zhou Biao emergency and relevant laws and regulations;

[0029] Based on the data content characteristics, the relevant data is divided into surrounding life data, surrounding emergency data and relevant laws and regulations data;

[0030] Store and save surrounding life data, surrounding emergency data and relevant legal and regulatory data.

[0031] Preferably, in S2, real-time monitoring of safety monitoring data is performed based on a large model to obtain risk accident data, including:

[0032] Obtain the accident judgment criteria for each risk accident;

[0033] Compare the safety monitoring data with the accident judgment criteria based on the big model, and when an abnormality occurs, determine that a risk accident has occurred;

[0034] Abnormal safety monitoring data will be regarded as risk accident data.

[0035] Preferably, in S3, key data correlation analysis is performed on the risk accident data and surrounding life data, surrounding emergency data and relevant laws and regulations data, and possible secondary accidents are determined based on the analysis results, including:

[0036] Extract key data about accident types and accident situations from risk accident data, and obtain associated information features matching the key data from an information association database;

[0037] Based on the characteristics of the associated data, the surrounding life data, the surrounding emergency data and the relevant laws and regulations data are associated to obtain associated life data, associated emergency data and associated laws and regulations data;

[0038] Calculate the correlation between the risk accident data and the associated life data, the associated emergency data and the associated legal and regulatory data based on a preset correlation algorithm, and select the target associated life data, the target associated emergency data and the target associated legal and regulatory data whose correlation is greater than a preset correlation threshold;

[0039] Obtaining the type of secondary accidents that may be caused by the risk accident data, and based on the data association between the secondary accident type and the risk accident type, associating the accident type of the risk accident data with the life type in the target-associated life data to obtain a type association degree, and associating the accident value of the risk accident data with the life value in the target-associated life data to obtain a value association degree;

[0040] Based on the type correlation, determine the secondary accidents that may be caused by the risk accident data;

[0041] Based on the product of the type association degree and the value association degree, the probability of occurrence of a possible secondary accident is determined.

[0042] Preferably, after determining the secondary accidents that may be caused by the risk accident data, the following steps are also included:

[0043] Associating the accident characteristics of the secondary accident with the target-related emergency data to obtain relevant emergency information;

[0044] Associating the accident characteristics of the secondary accident with target-related laws and regulations data to obtain relevant laws and regulations information;

[0045] Integrate the relevant emergency information and relevant legal and regulatory information and add them to the specific data of the secondary accident.

[0046] Preferably, in S4, the accident report document is automatically generated based on the risk accident data and the specific data of the secondary accidents that may occur, including:

[0047] Integrate the risk accident data and the specific data of possible secondary accidents to obtain a data set;

[0048] The document template selected by the user is retrieved from the large model, the data set is matched with the document template, and an accident report document is generated according to the matching result.

[0049] A large model system for emergency command decision support, comprising:

[0050] A data acquisition module is used to acquire safety monitoring data from the platform to be monitored, and to perform relevant searches on the safety monitoring data based on the big model, and to acquire surrounding life data, surrounding emergency data and relevant laws and regulations data related to the production and life data for storage;

[0051] Real-time monitoring module, used to monitor safety monitoring data in real time based on a large model to obtain risk accident data;

[0052] The correlation analysis module is used to conduct key data correlation analysis on risk accident data and surrounding life data, surrounding emergency data and relevant laws and regulations data, and determine the possible secondary accidents based on the analysis results;

[0053] The report generation module is used to automatically generate accident report documents based on risk accident data and specific data of possible secondary accidents.

[0054] Compared with the prior art, the present invention has achieved the following beneficial effects:

[0055] By acquiring the safety monitoring data from the platform to be monitored and performing relevant retrieval on the safety monitoring data based on the big model, the surrounding life data, surrounding emergency data and relevant legal and regulatory data related to the production and life data are acquired and saved, so as to realize comprehensive data acquisition based on the big model, perform real-time monitoring of the safety monitoring data based on the big model, obtain risk accident data, realize real-time monitoring of the platform to be monitored, perform key data correlation analysis on the risk accident data and the surrounding life data, surrounding emergency data and relevant legal and regulatory data, determine the possible secondary accidents according to the analysis results, and realize in-depth exploration of the implicit relationship between risk accidents based on data correlation to reveal potential risks, realize in-depth mining of the relationship between data, provide accurate accident information for emergency command assistance, and automatically generate accident report documents based on the specific data of risk accident data and possible secondary accidents, so as to provide auxiliary decision-making for commanders.

[0056] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures specifically pointed out in this application document.

[0057] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0059] Figure 1 A flowchart of a large model method for emergency command decision support in an embodiment of the present invention;

[0060] Figure 2 A flowchart of automatically generating an accident report document in an embodiment of the present invention;

[0061] Figure 3 The present invention is a structural diagram of a large model system for emergency command decision support in an embodiment of the present invention. DETAILED DESCRIPTION

[0062] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0063] Embodiment 1:

[0064] The embodiment of the present invention provides a large model method for emergency command auxiliary decision making, such as Figure 1 As shown, including:

[0065] S1: Obtain the safety monitoring data from the platform to be monitored, and perform relevant retrieval on the safety monitoring data based on the big model, obtain the surrounding life data related to the production and life data, surrounding emergency data and relevant laws and regulations data for storage;

[0066] S2: Monitor safety monitoring data in real time based on the big model to obtain risk accident data;

[0067] S3: Conduct key data correlation analysis on risk accident data, surrounding life data, surrounding emergency data and relevant laws and regulations data, and determine possible secondary accidents based on the analysis results;

[0068] S4: Automatically generate accident report documents based on risk accident data and specific data of possible secondary accidents.

[0069] In this embodiment, the platform to be monitored is, for example, an enterprise safety production digital platform, an enterprise safety production standardization management system, an emergency management comprehensive business platform, a visual emergency rescue smart platform, an urban safety risk dynamic distribution map system, a chemical park safety risk digital twin intelligent management and control platform, etc.

[0070] In this embodiment, the safety monitoring data includes temperature, pressure, air quality, production and life data, equipment data, etc.

[0071] In this embodiment, the safety monitoring data is monitored in real time based on the big model to obtain risk accident data. Specifically, when the safety monitoring data is abnormal, the risk accident data is obtained according to the abnormal situation, wherein the risk accidents include coal mine accidents, fire accidents, traffic accidents, gas leakage accidents, hazardous chemicals accidents, industrial and trade accidents and other risk accidents.

[0072] In this embodiment, the surrounding life data includes, for example, the activities of people in the surrounding area and the configuration of facilities.

[0073] In this embodiment, the surrounding emergency data includes, for example, emergency resources such as emergency teams, emergency experts, emergency supplies, and fire stations around the accident.

[0074] In this embodiment, the secondary accident is an accident derived from the risk accident.

[0075] The beneficial effects of the above design scheme are: by obtaining the safety monitoring data from the platform to be monitored, and performing relevant retrieval on the safety monitoring data based on the big model, the surrounding life data, surrounding emergency data and relevant legal and regulatory data related to the production and life data are obtained and saved, so as to realize comprehensive data acquisition based on the big model, and perform real-time monitoring of the safety monitoring data based on the big model to obtain risk accident data, realize real-time monitoring of the platform to be monitored, perform key data correlation analysis on the risk accident data and the surrounding life data, surrounding emergency data and relevant legal and regulatory data, determine the possible secondary accidents according to the analysis results, and realize in-depth exploration of the implicit relationship between risk accidents based on data correlation to reveal potential risks, realize in-depth mining of the relationship between data, provide accurate accident information for emergency command assistance, and automatically generate accident report documents based on the specific data of risk accident data and possible secondary accidents, so as to provide auxiliary decision-making for commanders.

[0076] Embodiment 2:

[0077] Based on Example 1, the embodiment of the present invention provides a large model method for emergency command decision support. In S1, obtaining security monitoring data from the platform to be monitored includes:

[0078] Obtain sensor data and daily record data of the platform to be monitored;

[0079] Preprocessing the sensor data and daily recorded data to obtain standard data;

[0080] Based on the safety monitoring standard, the standard data is specifically processed to obtain safety monitoring data.

[0081] In this embodiment, the preprocessing of the sensor data and daily recorded data includes data cleaning, data conversion, data denoising, etc.

[0082] In this embodiment, specific processing such as statistical analysis, machine learning, text analysis, etc. is performed on the standard data.

[0083] The beneficial effect of the above design scheme is: by obtaining the sensor collection data and daily record data of the platform to be monitored, the sensor collection data and daily record data are pre-processed to obtain standard data, and based on the safety monitoring standards, the standard data is specifically processed to obtain safety monitoring data, thereby providing high-quality data for emergency command and auxiliary decision-making.

[0084] Embodiment 3:

[0085] Based on Example 1, the embodiment of the present invention provides a large model method for emergency command decision-making assistance. In S1, the safety monitoring data is searched based on the large model to obtain surrounding life data related to the production and life data, surrounding emergency data and relevant laws and regulations data for storage, including:

[0086] Based on the monitoring area scope of the security monitoring data, determine the data acquisition area for related retrieval, and extract type keywords, name keywords and content keywords in the security monitoring data based on the data type, data name and data content;

[0087] Performing semantic association on the type keywords, name keywords and content keywords, and integrating the type keywords, name keywords and content keywords according to the semantic association result to obtain target keywords;

[0088] Acquire a relevant data directory from a data acquisition area based on the large model, and perform a relevant search on the target data directory based on the target keyword to obtain initial search data;

[0089] Acquire the safety monitoring data to monitor the safety accident type, and obtain the hazard source characteristics corresponding to the safety accident type;

[0090] Performing data feature matching on the hazard source feature and the initial search data to obtain a feature matching result, and obtaining a secondary search term based on the feature matching result;

[0091] Performing a horizontal search on the relevant data directory based on the secondary search term to obtain first supplementary search data;

[0092] Performing semantic expansion on the secondary search term to obtain an extended search term, performing an extended search on the initial search data based on the extended search term, and obtaining an in-depth search term from the initial search data according to the extended search result;

[0093] Performing an in-depth search on the relevant data directory based on the in-depth search term to obtain second supplementary search data;

[0094] Integrate the initial search data, the first supplementary search data, and the second supplementary search data, and obtain relevant data according to the integration result;

[0095] Based on the data content characteristics, the relevant data are divided into surrounding life data, surrounding emergency data and relevant laws and regulations data and saved.

[0096] In this embodiment, for example, when the safety accident is a fire, the hazardous characteristic source is hazardous chemicals, hazardous chemical processes, and the like.

[0097] In this embodiment, the first supplementary search data is to enrich the initial search data horizontally, for example, to enrich the regional scope.

[0098] In this embodiment, the second supplementary search data is used to further enrich the initial search data, for example, to enrich the situation within a regional range.

[0099] The beneficial effects of the above design scheme are: by performing relevant retrieval on the security monitoring data based on the big model, type keywords, name keywords and content keywords in the security monitoring data are extracted based on the data type, data name and data content, and semantic association is performed on the type keywords, name keywords and content keywords. According to the semantic association results, the type keywords, name keywords and content keywords are integrated to obtain the target keywords, and the relevant data directory from the data acquisition area is obtained based on the big model. The target data directory is searched based on the target keywords to obtain initial retrieval data to ensure the correctness of the retrieval, and then horizontal retrieval and in-depth retrieval are performed again based on the initial retrieval data to obtain supplementary retrieval data, and finally surrounding life data, surrounding emergency data and related legal and regulatory data are obtained and saved to ensure the richness of the relevant data and provide auxiliary data information for emergency command assistance.

[0100] Embodiment 4:

[0101] Based on Example 3, the embodiment of the present invention provides a large model method for emergency command decision support, wherein the method of obtaining a secondary search term based on a feature matching result includes:

[0102] The search data corresponding to the hazard source feature is obtained from the feature matching result, and the keywords of the search data are extracted. The hazard source feature is combined with the keywords to obtain a secondary search term.

[0103] The beneficial effect of the above design scheme is: by obtaining the search data corresponding to the hazard source characteristics from the feature matching results, and extracting the keywords of the search data, the hazard source characteristics are combined with the keywords to obtain secondary search terms, providing a basis for in-depth search.

[0104] Embodiment 5:

[0105] Based on Example 3, the embodiment of the present invention provides a large model method for emergency command auxiliary decision-making, which divides the relevant data based on the data content characteristics to obtain surrounding life data, surrounding emergency data and relevant legal and regulatory data and saves them, including:

[0106] Establish data content features based on specific data content of surrounding life, Zhou Biao emergency and relevant laws and regulations;

[0107] Based on the data content characteristics, the relevant data is divided into surrounding life data, surrounding emergency data and relevant laws and regulations data;

[0108] Store and save surrounding life data, surrounding emergency data and relevant legal and regulatory data.

[0109] The beneficial effect of the above design scheme is: by establishing data content characteristics based on specific data content of surrounding life, surrounding emergency and relevant laws and regulations, based on the data content characteristics, the relevant data is divided to obtain surrounding life data, surrounding emergency data and relevant laws and regulations data, and the surrounding life data, surrounding emergency data and relevant laws and regulations data are stored and preserved to facilitate the retrieval and viewing of relevant data.

[0110] Embodiment 6:

[0111] Based on Example 1, the embodiment of the present invention provides a large model method for emergency command decision-making assistance. In S2, the safety monitoring data is monitored in real time based on the large model to obtain risk accident data, including:

[0112] Obtain the accident judgment criteria for each risk accident;

[0113] Compare the safety monitoring data with the accident judgment criteria based on the big model, and when an abnormality occurs, determine that a risk accident has occurred;

[0114] Abnormal safety monitoring data will be regarded as risk accident data.

[0115] In this embodiment, the accident judgment standard is set in advance according to actual conditions.

[0116] The beneficial effect of the above design scheme is: by obtaining the accident judgment standard of each risk accident, the safety monitoring data is compared with the accident judgment standard based on the big model. When an abnormality occurs, it is determined that a risk accident has occurred, and the abnormal safety monitoring data is used as risk accident data to achieve real-time monitoring of the monitoring platform, providing real-time accident conditions for emergency command assistance.

[0117] Embodiment 7:

[0118] Based on Example 1, the embodiment of the present invention provides a large model method for emergency command decision-making assistance. In S3, key data association analysis is performed on risk accident data and surrounding life data, surrounding emergency data and relevant laws and regulations data, and possible secondary accidents are determined according to the analysis results, including:

[0119] Extract key data about accident types and accident situations from risk accident data, and obtain associated information features matching the key data from an information association database;

[0120] Based on the characteristics of the associated data, the surrounding life data, the surrounding emergency data and the relevant laws and regulations data are associated to obtain associated life data, associated emergency data and associated laws and regulations data;

[0121] Calculate the correlation between the risk accident data and the associated life data, the associated emergency data and the associated legal and regulatory data based on a preset correlation algorithm, and select the target associated life data, the target associated emergency data and the target associated legal and regulatory data whose correlation is greater than a preset correlation threshold;

[0122] Obtaining the type of secondary accidents that may be caused by the risk accident data, and based on the data association between the secondary accident type and the risk accident type, associating the accident type of the risk accident data with the life type in the target-associated life data to obtain a type association degree, and associating the accident value of the risk accident data with the life value in the target-associated life data to obtain a value association degree;

[0123] Based on the type correlation, determine the secondary accidents that may be caused by the risk accident data;

[0124] Based on the product of the type association degree and the value association degree, the probability of occurrence of a possible secondary accident is determined.

[0125] In this embodiment, the preset association algorithm is the Apriori algorithm, which is an algorithm for mining association rules through frequent item sets. The core idea of ​​the Apriori algorithm is to mine frequent item sets through two stages: candidate set generation and downward closed test.

[0126] In this embodiment, after the type association degree is greater than a preset type threshold, the corresponding secondary accident is determined to be a possible accident.

[0127] In this embodiment, the greater the product of the type association degree and the value association degree, the greater the probability of occurrence of the corresponding secondary accident that may be generated.

[0128] In this embodiment, the information association database is pre-designed to provide a basis for data association.

[0129] The beneficial effects of the above design scheme are: by extracting key data about the accident type and accident situation from the risk accident data, obtaining the associated information features matching the key data from the information association database; performing association processing based on the associated data features with surrounding life data, surrounding emergency data and relevant legal and regulatory data to obtain associated life data, associated emergency data and associated legal and regulatory data; calculating the association between the risk accident data and the associated life data, associated emergency data and associated legal and regulatory data based on a preset association algorithm, and selecting target associated life data, target associated emergency data and target associated legal and regulatory data whose association is greater than a preset association threshold; obtaining the secondary effects that the risk accident data may bring. Accident type, based on the data association between the secondary accident type and the risk accident type, associate the accident type of the risk accident data with the life type in the target-associated life data to obtain the type association degree, and associate the accident value of the risk accident data with the life value in the target-associated life data to obtain the value association degree; based on the type association degree, determine the secondary accidents that may be caused by the risk accident data, based on the product of the type association degree and the value association degree, determine the probability of occurrence of the secondary accidents that may be caused, based on data association, realize in-depth exploration of the implicit relationship between risk accidents to reveal potential risks, realize in-depth mining of the association between data, and provide accurate accident information for emergency command assistance.

[0130] Embodiment 8:

[0131] Based on Example 7, the embodiment of the present invention provides a large model method for emergency command decision support, which, after determining the secondary accidents that may be caused by the risk accident data, also includes:

[0132] Associating the accident characteristics of the secondary accident with the target-related emergency data to obtain relevant emergency information;

[0133] Associating the accident characteristics of the secondary accident with target-related laws and regulations data to obtain relevant laws and regulations information;

[0134] Integrate the relevant emergency information and relevant legal and regulatory information and add them to the specific data of the secondary accident.

[0135] The beneficial effects of the above design scheme are: by associating the accident characteristics of the secondary accident with the target-related emergency data, relevant emergency information is obtained, and the accident characteristics of the secondary accident are associated with the target-related legal and regulatory data to obtain relevant legal and regulatory information, and the relevant emergency information and relevant legal and regulatory information are integrated and added to the specific data situation of the secondary accident, providing emergency data for the generation of subsequent accident report documents, and providing auxiliary decision-making for commanders.

[0136] Embodiment 9:

[0137] Based on Example 1, the present invention provides a large model method for emergency command auxiliary decision making, such as Figure 3 As shown, in S4, an accident report document is automatically generated based on the risk accident data and the specific data of the possible secondary accidents, including:

[0138] Integrate the risk accident data and the specific data of possible secondary accidents to obtain a data set;

[0139] The document template selected by the user is retrieved from the large model, the data set is matched with the document template, and an accident report document is generated according to the matching result.

[0140] The beneficial effect of the above design scheme is: by integrating the risk accident data and the specific data of possible secondary accidents, a data set is obtained, the document template selected by the user is retrieved from the large model, the data set is matched with the document template, and an accident report document is generated according to the matching results to provide auxiliary decision-making for the commander.

[0141] Embodiment 10:

[0142] The embodiment of the present invention provides a large model system for emergency command decision support, which is used to implement the steps of the large model method described in any one of Embodiments 1-9, such as Figure 3 As shown, including:

[0143] A data acquisition module is used to acquire safety monitoring data from the platform to be monitored, and to perform relevant searches on the safety monitoring data based on the big model, and to acquire surrounding life data, surrounding emergency data and relevant laws and regulations data related to the production and life data for storage;

[0144] Real-time monitoring module, used to monitor safety monitoring data in real time based on a large model to obtain risk accident data;

[0145] The correlation analysis module is used to conduct key data correlation analysis on risk accident data and surrounding life data, surrounding emergency data and relevant laws and regulations data, and determine the possible secondary accidents based on the analysis results;

[0146] The report generation module is used to automatically generate accident report documents based on risk accident data and specific data of possible secondary accidents.

[0147] In this embodiment, the platform to be monitored is, for example, an enterprise safety production digital platform, an enterprise safety production standardization management system, an emergency management comprehensive business platform, a visual emergency rescue smart platform, an urban safety risk dynamic distribution map system, a chemical park safety risk digital twin intelligent management and control platform, etc.

[0148] In this embodiment, the safety monitoring data includes temperature, pressure, air quality, production and life data, equipment data, etc.

[0149] In this embodiment, the safety monitoring data is monitored in real time based on the big model to obtain risk accident data. Specifically, when the safety monitoring data is abnormal, the risk accident data is obtained according to the abnormal situation, wherein the risk accidents include coal mine accidents, fire accidents, traffic accidents, gas leakage accidents, hazardous chemicals accidents, industrial and trade accidents and other risk accidents.

[0150] In this embodiment, the surrounding life data includes, for example, the activities of people in the surrounding area and the configuration of facilities.

[0151] In this embodiment, the surrounding emergency data includes, for example, emergency resources such as emergency teams, emergency experts, emergency supplies, and fire stations around the accident.

[0152] In this embodiment, the secondary accident is an accident derived from the risk accident.

[0153] The beneficial effects of the above design scheme are: by obtaining the safety monitoring data from the platform to be monitored, and performing relevant retrieval on the safety monitoring data based on the big model, the surrounding life data, surrounding emergency data and relevant legal and regulatory data related to the production and life data are obtained and saved, so as to realize comprehensive data acquisition based on the big model, and perform real-time monitoring of the safety monitoring data based on the big model to obtain risk accident data, realize real-time monitoring of the platform to be monitored, perform key data correlation analysis on the risk accident data and the surrounding life data, surrounding emergency data and relevant legal and regulatory data, determine the possible secondary accidents according to the analysis results, and realize in-depth exploration of the implicit relationship between risk accidents based on data correlation to reveal potential risks, realize in-depth mining of the relationship between data, provide accurate accident information for emergency command assistance, and automatically generate accident report documents based on the specific data of risk accident data and possible secondary accidents, so as to provide auxiliary decision-making for commanders.

[0154] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention belong to the scope of this application document and its equivalent technology, the present invention is also intended to include these changes and variations.

Claims

1. A large model method for emergency command decision support, characterized in that: include: S1: Obtain the safety monitoring data from the platform to be monitored, and perform relevant retrieval on the safety monitoring data based on the big model, obtain the surrounding life data related to the production and life data, surrounding emergency data and relevant laws and regulations data for storage; S2: Monitor safety monitoring data in real time based on the big model to obtain risk accident data; S3: Conduct key data correlation analysis on risk accident data, surrounding life data, surrounding emergency data and relevant laws and regulations data, and determine possible secondary accidents based on the analysis results; S4: Automatically generate accident report documents based on risk accident data and specific data of possible secondary accidents.

2. The large model method for emergency command decision support according to claim 1 is characterized in that: In S1, obtaining security monitoring data from the platform to be monitored includes: Obtain sensor data and daily record data of the platform to be monitored; Preprocessing the sensor data and daily recorded data to obtain standard data; Based on the safety monitoring standard, the standard data is specifically processed to obtain safety monitoring data.

3. The large model method for emergency command decision support according to claim 1 is characterized in that: In S1, the safety monitoring data is searched based on the big model to obtain surrounding life data, surrounding emergency data and relevant laws and regulations data related to the production and life data for storage, including: Based on the monitoring area scope of the security monitoring data, determine the data acquisition area for related retrieval, and extract type keywords, name keywords and content keywords in the security monitoring data based on the data type, data name and data content; Performing semantic association on the type keywords, name keywords and content keywords, and integrating the type keywords, name keywords and content keywords according to the semantic association result to obtain target keywords; Acquire a relevant data directory from a data acquisition area based on the large model, and perform a relevant search on the target data directory based on the target keyword to obtain initial search data; Acquire the safety monitoring data to monitor the safety accident type, and obtain the hazard source characteristics corresponding to the safety accident type; Performing data feature matching on the hazard source feature and the initial search data to obtain a feature matching result, and obtaining a secondary search term based on the feature matching result; Performing a horizontal search on the relevant data directory based on the secondary search term to obtain first supplementary search data; Performing semantic expansion on the secondary search term to obtain an extended search term, performing an extended search on the initial search data based on the extended search term, and obtaining an in-depth search term from the initial search data according to the extended search result; Performing an in-depth search on the relevant data directory based on the in-depth search term to obtain second supplementary search data; Integrate the initial search data, the first supplementary search data, and the second supplementary search data, and obtain relevant data according to the integration result; Based on the data content characteristics, the relevant data are divided into surrounding life data, surrounding emergency data and relevant laws and regulations data and saved.

4. The large model method for emergency command decision support according to claim 3 is characterized in that: The obtaining of secondary search terms based on the feature matching results includes: The search data corresponding to the hazard source feature is obtained from the feature matching result, and the keywords of the search data are extracted. The hazard source feature is combined with the keywords to obtain a secondary search term.

5. The large model method for emergency command decision support according to claim 3 is characterized in that: Based on the data content characteristics, the relevant data is divided into surrounding life data, surrounding emergency data and relevant laws and regulations data and saved, including: Establish data content features based on specific data content of surrounding life, Zhou Biao emergency and relevant laws and regulations; Based on the data content characteristics, the relevant data is divided into surrounding life data, surrounding emergency data and relevant laws and regulations data; Store and save surrounding life data, surrounding emergency data and relevant legal and regulatory data.

6. The large model method for emergency command decision support according to claim 1 is characterized in that: In S2, the safety monitoring data is monitored in real time based on the big model to obtain risk accident data, including: Obtain the accident judgment criteria for each risk accident; Compare the safety monitoring data with the accident judgment criteria based on the big model, and when an abnormality occurs, determine that a risk accident has occurred; Abnormal safety monitoring data will be regarded as risk accident data.

7. The large model method for emergency command decision support according to claim 1 is characterized in that: In S3, key data correlation analysis is performed on risk accident data and surrounding life data, surrounding emergency data and relevant laws and regulations data, and possible secondary accidents are determined based on the analysis results, including: Extract key data about accident types and accident situations from risk accident data, and obtain associated information features matching the key data from an information association database; Based on the characteristics of the associated data, the surrounding life data, the surrounding emergency data and the relevant laws and regulations data are associated to obtain associated life data, associated emergency data and associated laws and regulations data; Calculate the correlation between the risk accident data and the associated life data, the associated emergency data and the associated legal and regulatory data based on a preset correlation algorithm, and select the target associated life data, the target associated emergency data and the target associated legal and regulatory data whose correlation is greater than a preset correlation threshold; Obtaining the type of secondary accidents that may be caused by the risk accident data, and based on the data association between the secondary accident type and the risk accident type, associating the accident type of the risk accident data with the life type in the target-associated life data to obtain a type association degree, and associating the accident value of the risk accident data with the life value in the target-associated life data to obtain a value association degree; Based on the type correlation, determine the secondary accidents that may be caused by the risk accident data; Based on the product of the type association degree and the value association degree, the probability of occurrence of a possible secondary accident is determined.

8. The large model method for emergency command decision support according to claim 7 is characterized in that: After determining the secondary accidents that may be caused by risk accident data, it also includes: Associating the accident characteristics of the secondary accident with the target-related emergency data to obtain relevant emergency information; Associating the accident characteristics of the secondary accident with target-related laws and regulations data to obtain relevant laws and regulations information; Integrate the relevant emergency information and relevant legal and regulatory information and add them to the specific data of the secondary accident.

9. The large model method for emergency command decision support according to claim 1 is characterized in that: In S4, an accident report document is automatically generated based on the risk accident data and the specific data of the possible secondary accidents, including: Integrate the risk accident data and the specific data of possible secondary accidents to obtain a data set; The document template selected by the user is retrieved from the large model, the data set is matched with the document template, and an accident report document is generated according to the matching result.

10. A large model system for emergency command decision support, used to implement the steps of the large model method described in any one of claims 1 to 9, characterized in that: include: A data acquisition module is used to acquire safety monitoring data from the platform to be monitored, and to perform relevant searches on the safety monitoring data based on the big model, and to acquire surrounding life data, surrounding emergency data and relevant laws and regulations data related to the production and life data for storage; Real-time monitoring module, used to monitor safety monitoring data in real time based on a large model to obtain risk accident data; The correlation analysis module is used to conduct key data correlation analysis on risk accident data and surrounding life data, surrounding emergency data and relevant laws and regulations data, and determine the possible secondary accidents based on the analysis results; The report generation module is used to automatically generate accident report documents based on risk accident data and specific data of possible secondary accidents.

Images