Security intelligent decision optimization method based on knowledge graph model

Abstract

The invention discloses a safety intelligent decision optimization method based on a knowledge graph model, and relates to the field of tobacco safety. The system comprises a knowledge graph construction module, an environmental parameter decision module, a data acquisition and processing module, a risk assessment module and a decision generation module. The knowledge graph construction module comprises a data acquisition unit, an entity recognition unit, a relationship extraction unit and a knowledge storage unit; wherein the data acquisition unit is used for acquiring multi-source data in the tobacco industry, and the multi-source data comprises safety specification clause texts, equipment operation data and historical risk cases. According to the method, the knowledge graph construction module is used for carrying out association modeling on entities such as safety specification terms, risk points, environmental parameters and emergency resources, data islands are broken, a searchable knowledge network is formed, and the comprehensiveness and relevance of risk analysis are improved.

Description

Technical Field

[0001] The present invention relates to the field of tobacco safety technology, and in particular to a safety intelligent decision-making optimization method based on a knowledge graph model. Background Art

[0002] Traditional tobacco safety management relies on manual review of paper documents or scattered electronic documents. The average time to locate key clauses exceeds 30 minutes, and due to information fragmentation, the clause coverage rate is less than 60%. For example, when warehouse personnel search for the "tobacco storage" specifications, it is difficult to associate them with the implicit clause "inadequate ventilation-spontaneous combustion risk" in the "Tobacco Storage Safety Specifications", resulting in incomplete risk identification. Management risk assessment relies on experience and judgment, and cannot link the "safety specifications-risk points-emergency plans-resource allocation" data chain in real time. The emergency plan is disconnected from the risk scenario. When responding to emergencies, it is necessary to manually review paper plans or search scattered electronic documents online. The average handling delay exceeds 45 minutes, resulting in missing the best time for initial handling.

[0003] Therefore, we make improvements to this and propose a security intelligent decision optimization method based on the knowledge graph model. Summary of the Invention

[0004] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions: The specific application is as follows: A safety intelligent decision-making optimization system based on a knowledge graph model, comprising: a knowledge graph construction module, an environmental parameter decision module, a data acquisition and processing module, a risk assessment module, and a decision generation module; The knowledge graph construction module includes data acquisition unit, entity recognition unit, relationship extraction unit and knowledge storage unit; The data collection unit is used to collect multi-source data from the tobacco industry, including safety specification clauses, equipment operation data, and historical risk cases; An entity recognition unit, which identifies entities based on NLP natural language processing technology, wherein the entities include at least safety specification clauses, risk points, environmental parameters, and emergency resources; Relationship extraction unit, which establishes relationships between entities, including: specification-risk control relationship, parameter-risk trigger relationship, and plan-resource allocation relationship; The knowledge storage unit uses Neo4j graph database to store entity-relationship triples; Environmental parameter decision module, including threshold configuration unit, risk rule configuration unit and weight configuration unit; Among them, the threshold configuration unit sets three levels of threshold ranges for environmental parameters: normal range, warning range, and danger range; Risk rule configuration unit, which defines the risk level classification rules; The weight configuration unit calculates the weight coefficient of each parameter based on the AHP analytical hierarchy process; Data acquisition and processing module, including sensor interface unit, data cleaning unit and data storage unit; Among them, the sensor interface unit is connected to the temperature and humidity sensor and the dust concentration monitor; The data cleaning unit removes abnormal data through median filtering denoising algorithm and standardization processing; Data storage unit, using a time series database to store real-time parameter data; Risk assessment module, including data acquisition unit, knowledge query unit, risk calculation unit and risk level determination unit; Among them, the data acquisition unit regularly synchronizes the latest parameter data from the time series database; Knowledge query unit, which searches for risk points and regulatory clauses corresponding to the parameters based on the knowledge graph; The risk calculation unit uses a piecewise linear interpolation algorithm to calculate the individual risk scores. If Ri = 0, it is within the normal range; if Ri = 5 × (Vi − Nmax) / (Wmax − Nmax), it is within the warning range; if Ri = 5 + 5 × (Vi − Wmax) / (Dmax − Wmax), it is within the danger range, where Ri is the risk score of the i-th parameter, Vi is the actual value of the parameter, Nmax is the upper limit of the normal range, Wmax is the upper limit of the warning range, and Dmax is the upper limit of the danger range. The risk level determination unit determines the risk level based on the comprehensive risk score. The formula is: ; Decision-making module, including plan retrieval unit, measure matching unit, report generation unit and resource scheduling unit; Among them, the emergency plan retrieval unit retrieves and calls the emergency plan from the knowledge graph according to the risk level; The measure matching unit automatically associates specification clauses based on parameter anomaly types and matches preventive measures; Report generation unit, which outputs a decision report containing a list of violation clauses, a risk heat map, and a ranking of the priority of measures; The resource scheduling unit combines the emergency resource location information and the path planning algorithm to generate the optimal scheduling plan.

[0005] A security intelligent decision-making optimization method based on a knowledge graph model includes: the security intelligent decision-making optimization system based on a knowledge graph model described above, and the specific method is as follows: SA1. The knowledge graph construction module constructs the knowledge graph of the tobacco industry and outputs the knowledge graph to the risk assessment module and decision generation module; SA2, the environmental parameter decision module defines the environmental parameter decision model and outputs the threshold, risk rules and weights to the risk assessment module; SA3, the data acquisition and processing module collects and processes data in real time and outputs the processed data to the risk assessment module; SA4, the risk assessment module performs risk assessment and analysis and outputs the risk level to the decision-making module; SA5. The decision generation module generates an intelligent decision report.

[0006] The specific steps for constructing the knowledge graph of the tobacco industry in SA1 are as follows: SB1, start; SB2. Collect multi-source data from the tobacco industry through the data collection unit. Multi-source data includes safety regulations, enterprise operating procedures, and equipment manuals. SB3, the entity recognition unit uses natural language processing technology to identify entities and relationships. The entities include at least safety regulations, risk points, environmental parameters, and emergency resources. The environmental parameters include at least temperature, humidity, ventilation ratio, and dust concentration. SB4. The relationship extraction unit establishes relationships between entities, including but not limited to: the control relationship between regulatory clauses and risk points, the trigger relationship between environmental parameters and risk points, and the configuration relationship between emergency plans and emergency resources; SB5, the knowledge storage unit stores entities and relationships in a graph database to form a knowledge graph; SB6, end.

[0007] The specific steps for defining the environmental parameter decision model in SA2 are as follows: SC1, start; SC2, the threshold configuration unit sets multi-level threshold ranges for each environmental parameter, including normal range, warning range and danger range; SC3, the risk rule configuration unit establishes risk level classification rules and determines the risk level based on the degree and duration of environmental parameters exceeding the threshold; SC4, weight configuration unit defines the weight coefficient of each environmental parameter, which is used to calculate the comprehensive risk score; SC5, end; The specific steps for real-time data collection and processing in SA3 are as follows: SD1, start; SD2, sensor interface unit collects environmental parameters in real time through temperature and humidity sensors and dust concentration monitors; SD3, the data cleaning unit performs denoising and standardization on the collected data using a median filter denoising algorithm to remove abnormal data; SD4, the data storage unit stores the processed data into the time series database, with a sampling interval of ≤5 minutes; SD5, end.

[0008] The specific steps for risk assessment and analysis in the risk assessment module of SA4 are as follows: SE1, start; SE2, the data acquisition unit obtains the latest environmental parameter data from the time series database; SE3, the knowledge query unit retrieves the risk points and regulatory clauses corresponding to the current environmental parameters based on the "trigger relationship between environmental parameters and risk points" in the knowledge graph; SE4, the risk calculation unit calculates the individual risk score of each risk point and calculates the comprehensive risk score based on the weight coefficient; SE5, the risk level determination unit determines the current risk level; SE6, end.

[0009] The specific steps for generating an intelligent decision report in SA5 are as follows: SF1, start; SF2, the emergency plan retrieval unit calls the corresponding emergency plan from the "configuration relationship between emergency plans and emergency resources" in the knowledge graph according to the current risk level; SF3, the measure matching unit matches preventive measures based on the parameter anomaly type through the "control relationship between regulatory clauses and risk points" in the knowledge graph; SF4, the report generation unit generates a decision report containing references to regulatory clauses, risk analysis, preventive measures, and emergency plans; SF5, the resource scheduling unit generates resource scheduling suggestions based on the emergency resource allocation relationship; SF6, end; Among them, the generation rules of preventive measures in SF5: the threshold configuration unit sets default thresholds for environmental parameters such as temperature, humidity, ventilation ratio, and dust concentration, which can be adjusted manually; The default temperature threshold is 30°C. When the temperature exceeds the threshold, the air conditioning system is activated and the stacking layout is adjusted to enhance ventilation. The default humidity threshold is 65%. When the humidity exceeds the threshold, the dehumidifier is started and the storage seal is checked; The default ventilation threshold is 0.6 times / h. When the ventilation ratio is lower than the threshold, mechanical ventilation equipment is added and the stacking method is optimized; The default dust threshold is 10mg / m³. When the dust concentration exceeds the threshold, the dust removal equipment will be activated and the cleaning frequency will be increased.

[0010] The steps for generating a decision report in SF5 include: SF5.1, start; SF5.2. Extract all regulatory clauses associated with the current risk point in the knowledge graph; SF5.3. Sort the regulatory clauses by importance and relevance; SF5.4. Generate compliance analysis and indicate the specific clauses that the current environmental parameters violate; SF5.5. Generate prioritized preventive measures based on the severity of the violation; SF5.6. Generate executable emergency plans based on the real-time status of emergency resources; SF5.7, end.

[0011] Compared with the prior art, the present invention has the following beneficial effects: 1. Use the knowledge graph construction module to conduct association modeling for entities such as safety regulations, risk points, environmental parameters, and emergency resources, breaking down data silos, forming a searchable knowledge network, and improving the comprehensiveness and relevance of risk analysis.

[0012] 2. The environmental parameter decision module calculates individual and comprehensive risk scores through three-level threshold division and AHP weight configuration, combined with a piecewise linear interpolation algorithm, to achieve dynamic quantitative assessment from single parameter alarm to multi-dimensional risk level, significantly improving the timeliness and accuracy of risk identification.

[0013] 3. The decision-making module searches for emergency plans in real time based on the knowledge graph, automatically matches the preventive measures corresponding to the regulatory clauses, and optimizes emergency resource scheduling through path planning algorithms, shortening the traditional manual decision-making process and improving emergency response efficiency.

[0014] 4. Historical risk cases and disposal plans are stored in the knowledge graph, supporting rapid matching of similar risks and experience reuse. At the same time, entity relationships can be dynamically updated through real-time data feedback to achieve continuous iterative optimization of the security decision-making model. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a schematic diagram of the security intelligent decision-making optimization system architecture for this application; Figure 2 A schematic diagram of the process of constructing the knowledge graph for this application; Figure 3 This is a flow chart of the environmental parameter decision model for this application; Figure 4 This is a schematic diagram of the data collection and processing flow for this application; Figure 5 This is a schematic diagram of the risk assessment and analysis process for this application; Figure 6 A flowchart of the intelligent decision-making report generation process for this application; Figure 7 This is a table diagram of the precautionary measures rules for this application. DETAILED DESCRIPTION

[0016] To make the purpose, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them.

[0017] Therefore, the following detailed description of the embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely represents some embodiments of the present invention. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without creative effort shall fall within the scope of protection of the present invention.

[0018] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions therein may be combined with each other.

[0019] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0020] In the description of the present invention, it should be noted that the terms "upper" and "lower" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, or the orientations or positional relationships in which the inventive product is typically placed when in use, or the orientations or positional relationships commonly understood by those skilled in the art. Such terms are intended solely to facilitate the description of the present invention and simplify the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limitations on the present invention. Furthermore, the terms "first" and "second" and the like are used solely for distinction and should not be construed as indicating or implying relative importance.

[0021] The present invention provides the following technical solutions: Please refer to Figure 1-7 , a safety intelligent decision optimization system based on the knowledge graph model, including: a knowledge graph construction module, an environmental parameter decision module, a data acquisition and processing module, a risk assessment module and a decision generation module; The knowledge graph construction module includes data acquisition unit, entity recognition unit, relationship extraction unit and knowledge storage unit; The data collection unit is used to collect multi-source data from the tobacco industry, including safety specification clauses, equipment operation data, and historical risk cases; An entity recognition unit, which identifies entities based on NLP natural language processing technology, wherein the entities include at least safety specification clauses, risk points, environmental parameters, and emergency resources; Relationship extraction unit, which establishes relationships between entities, including: specification-risk control relationship, parameter-risk trigger relationship, and plan-resource allocation relationship; The knowledge storage unit uses Neo4j graph database to store entity-relationship triples; Environmental parameter decision module, including threshold configuration unit, risk rule configuration unit and weight configuration unit; Among them, the threshold configuration unit sets three levels of threshold ranges for environmental parameters: normal range, warning range, and danger range; Risk rule configuration unit, which defines the risk level classification rules; The weight configuration unit calculates the weight coefficient of each parameter based on the AHP analytical hierarchy process; Data acquisition and processing module, including sensor interface unit, data cleaning unit and data storage unit; Among them, the sensor interface unit is connected to the temperature and humidity sensor and the dust concentration monitor; The data cleaning unit removes abnormal data through median filtering denoising algorithm and standardization processing; Data storage unit, using a time series database to store real-time parameter data; Risk assessment module, including data acquisition unit, knowledge query unit, risk calculation unit and risk level determination unit; Among them, the data acquisition unit regularly synchronizes the latest parameter data from the time series database; Knowledge query unit, which searches for risk points and regulatory clauses corresponding to the parameters based on the knowledge graph; The risk calculation unit uses a piecewise linear interpolation algorithm to calculate the individual risk scores. If Ri = 0, it is within the normal range; if Ri = 5 × (Vi − Nmax) / (Wmax − Nmax), it is within the warning range; if Ri = 5 + 5 × (Vi − Wmax) / (Dmax − Wmax), it is within the danger range, where Ri is the risk score of the i-th parameter, Vi is the actual value of the parameter, Nmax is the upper limit of the normal range, Wmax is the upper limit of the warning range, and Dmax is the upper limit of the danger range. The risk level determination unit determines the risk level based on the comprehensive risk score. The formula is: ; Decision-making module, including plan retrieval unit, measure matching unit, report generation unit and resource scheduling unit; Among them, the emergency plan retrieval unit retrieves and calls the emergency plan from the knowledge graph according to the risk level; The measure matching unit automatically associates specification clauses based on parameter anomaly types and matches preventive measures; Report generation unit, which outputs a decision report containing a list of violation clauses, a risk heat map, and a ranking of the priority of measures; The resource scheduling unit combines the emergency resource location information and the path planning algorithm to generate the optimal scheduling plan.

[0022] A security intelligent decision-making optimization method based on a knowledge graph model includes: the security intelligent decision-making optimization system based on a knowledge graph model described above, and the specific method is as follows: SA1. The knowledge graph construction module constructs the knowledge graph of the tobacco industry and outputs the knowledge graph to the risk assessment module and decision generation module; SA2, the environmental parameter decision module defines the environmental parameter decision model and outputs the threshold, risk rules and weights to the risk assessment module; SA3, the data acquisition and processing module collects and processes data in real time and outputs the processed data to the risk assessment module; SA4, the risk assessment module performs risk assessment and analysis and outputs the risk level to the decision-making module; SA5. The decision generation module generates an intelligent decision report.

[0023] The specific steps for constructing the knowledge graph of the tobacco industry in SA1 are as follows: SB1, start; SB2. Collect multi-source data from the tobacco industry through the data collection unit. Multi-source data includes safety regulations, enterprise operating procedures, and equipment manuals. SB3, the entity recognition unit uses natural language processing technology to identify entities and relationships. The entities include at least safety regulations, risk points, environmental parameters, and emergency resources. The environmental parameters include at least temperature, humidity, ventilation ratio, and dust concentration. SB4. The relationship extraction unit establishes relationships between entities, including but not limited to: the control relationship between regulatory clauses and risk points, the trigger relationship between environmental parameters and risk points, and the configuration relationship between emergency plans and emergency resources; SB5, the knowledge storage unit stores entities and relationships in a graph database to form a knowledge graph; SB6, end.

[0024] The specific steps for defining the environmental parameter decision model in SA2 are as follows: SC1, start; SC2, the threshold configuration unit sets multi-level threshold ranges for each environmental parameter, including normal range, warning range and danger range; SC3, the risk rule configuration unit establishes risk level classification rules and determines the risk level based on the degree and duration of environmental parameters exceeding the threshold; SC4, weight configuration unit defines the weight coefficient of each environmental parameter, which is used to calculate the comprehensive risk score; SC5, end; The specific steps for real-time data collection and processing in SA3 are as follows: SD1, start; SD2, sensor interface unit collects environmental parameters in real time through temperature and humidity sensors and dust concentration monitors; SD3, the data cleaning unit performs denoising and standardization on the collected data using a median filter denoising algorithm to remove abnormal data; SD4, the data storage unit stores the processed data into the time series database, with a sampling interval of ≤5 minutes; SD5, end.

[0025] The specific steps for risk assessment and analysis in the risk assessment module of SA4 are as follows: SE1, start; SE2, the data acquisition unit obtains the latest environmental parameter data from the time series database; SE3, the knowledge query unit retrieves the risk points and regulatory clauses corresponding to the current environmental parameters based on the "trigger relationship between environmental parameters and risk points" in the knowledge graph; SE4, the risk calculation unit calculates the individual risk score of each risk point and calculates the comprehensive risk score based on the weight coefficient; SE5, the risk level determination unit determines the current risk level; SE6, end.

[0026] The specific steps for generating an intelligent decision report in SA5 are as follows: SF1, start; SF2, the emergency plan retrieval unit calls the corresponding emergency plan from the "configuration relationship between emergency plans and emergency resources" in the knowledge graph according to the current risk level; SF3, the measure matching unit matches preventive measures based on the parameter anomaly type through the "control relationship between regulatory clauses and risk points" in the knowledge graph; SF4, the report generation unit generates a decision report containing references to regulatory clauses, risk analysis, preventive measures, and emergency plans; SF5, the resource scheduling unit generates resource scheduling suggestions based on the emergency resource allocation relationship; SF6, end; Among them, the generation rules of preventive measures in SF5: the threshold configuration unit sets default thresholds for environmental parameters such as temperature, humidity, ventilation ratio, and dust concentration, which can be adjusted manually; The default temperature threshold is 30°C. When the temperature exceeds the threshold, the air conditioning system is activated and the stacking layout is adjusted to enhance ventilation. The default humidity threshold is 65%. When the humidity exceeds the threshold, the dehumidifier is started and the storage seal is checked; The default ventilation threshold is 0.6 times / h. When the ventilation ratio is lower than the threshold, mechanical ventilation equipment is added and the stacking method is optimized; The default dust threshold is 10mg / m³. When the dust concentration exceeds the threshold, the dust removal equipment will be activated and the cleaning frequency will be increased.

[0027] The steps for generating a decision report in SF5 include: SF5.1, start; SF5.2. Extract all regulatory clauses associated with the current risk point in the knowledge graph; SF5.3. Sort the regulatory clauses by importance and relevance; SF5.4. Generate compliance analysis and indicate the specific clauses that the current environmental parameters violate; SF5.5. Generate prioritized preventive measures based on the severity of the violation; SF5.6. Generate executable emergency plans based on the real-time status of emergency resources; SF5.7, end.

[0028] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0029] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features and technical solutions therein may be combined with each other.

[0030] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings.

[0031] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described in the present invention. Although this specification has described the present invention in detail with reference to the above embodiments, the present invention is not limited to the above specific implementation methods. Therefore, any modification or equivalent replacement of the present invention; and all technical solutions and improvements thereof that do not depart from the spirit and scope of the invention are included in the scope of the claims of the present invention.

Claims

1. A security intelligent decision-making optimization system based on a knowledge graph model, characterized in that: include: Knowledge graph construction module, environmental parameter decision module, data acquisition and processing module, risk assessment module and decision generation module; The knowledge graph construction module includes data acquisition unit, entity recognition unit, relationship extraction unit and knowledge storage unit; The data collection unit is used to collect multi-source data from the tobacco industry, including safety specification clauses, equipment operation data, and historical risk cases; An entity recognition unit, which identifies entities based on NLP natural language processing technology, wherein the entities include at least safety specification clauses, risk points, environmental parameters, and emergency resources; Relationship extraction unit, which establishes relationships between entities, including: specification-risk control relationship, parameter-risk trigger relationship, and plan-resource allocation relationship; The knowledge storage unit uses Neo4j graph database to store entity-relationship triples; Environmental parameter decision module, including threshold configuration unit, risk rule configuration unit and weight configuration unit; Among them, the threshold configuration unit sets three levels of threshold ranges for environmental parameters: normal range, warning range, and danger range; Risk rule configuration unit, which defines the risk level classification rules; The weight configuration unit calculates the weight coefficient of each parameter based on the AHP analytical hierarchy process; Data acquisition and processing module, including sensor interface unit, data cleaning unit and data storage unit; Among them, the sensor interface unit is connected to the temperature and humidity sensor and the dust concentration monitor; The data cleaning unit removes abnormal data through median filtering denoising algorithm and standardization processing; Data storage unit, using a time series database to store real-time parameter data; Risk assessment module, including data acquisition unit, knowledge query unit, risk calculation unit and risk level determination unit; Among them, the data acquisition unit regularly synchronizes the latest parameter data from the time series database; Knowledge query unit, which searches for risk points and regulatory clauses corresponding to the parameters based on the knowledge graph; The risk calculation unit uses a piecewise linear interpolation algorithm to calculate the individual risk scores. If Ri = 0, it is within the normal range; if Ri = 5 × (Vi − Nmax) / (Wmax − Nmax), it is within the warning range; if Ri = 5 + 5 × (Vi − Wmax) / (Dmax − Wmax), it is within the danger range, where Ri is the risk score of the i-th parameter, Vi is the actual value of the parameter, Nmax is the upper limit of the normal range, Wmax is the upper limit of the warning range, and Dmax is the upper limit of the danger range. The risk level determination unit determines the risk level based on the comprehensive risk score. The formula is: ; Decision-making module, including plan retrieval unit, measure matching unit, report generation unit and resource scheduling unit; Among them, the emergency plan retrieval unit retrieves and calls the emergency plan from the knowledge graph according to the risk level; The measure matching unit automatically associates specification clauses based on parameter anomaly types and matches preventive measures; Report generation unit, which outputs a decision report containing a list of violation clauses, a risk heat map, and a ranking of the priority of measures; The resource scheduling unit combines the emergency resource location information and the path planning algorithm to generate the optimal scheduling plan.

2. A security intelligent decision-making optimization method based on a knowledge graph model, characterized in that: include: The security intelligent decision-making optimization system based on the knowledge graph model described in claim 1, the specific method is as follows: SA1. The knowledge graph construction module constructs the knowledge graph of the tobacco industry and outputs the knowledge graph to the risk assessment module and decision generation module; SA2, the environmental parameter decision module defines the environmental parameter decision model and outputs the threshold, risk rules and weights to the risk assessment module; SA3, the data acquisition and processing module collects and processes data in real time and outputs the processed data to the risk assessment module; SA4, the risk assessment module performs risk assessment and analysis and outputs the risk level to the decision-making module; SA5. The decision generation module generates an intelligent decision report.

3. A security intelligent decision-making optimization method based on a knowledge graph model according to claim 2, characterized in that: The specific steps for constructing the knowledge graph of the tobacco industry in SA1 are as follows: SB1, start; SB2. Collect multi-source data from the tobacco industry through the data collection unit. Multi-source data includes safety regulations, enterprise operating procedures, and equipment manuals. SB3, the entity recognition unit uses natural language processing technology to identify entities and relationships. The entities include at least safety regulations, risk points, environmental parameters, and emergency resources. The environmental parameters include at least temperature, humidity, ventilation ratio, and dust concentration. SB4. The relationship extraction unit establishes relationships between entities, including but not limited to: the control relationship between regulatory clauses and risk points, the trigger relationship between environmental parameters and risk points, and the configuration relationship between emergency plans and emergency resources; SB5, the knowledge storage unit stores entities and relationships in a graph database to form a knowledge graph; SB6, end.

4. A security intelligent decision-making optimization method based on a knowledge graph model according to claim 2, characterized in that: The specific steps of defining the environmental parameter decision model in SA2 are as follows: SC1, start; SC2, the threshold configuration unit sets multi-level threshold ranges for each environmental parameter, including normal range, warning range and danger range; SC3, the risk rule configuration unit establishes risk level classification rules and determines the risk level based on the degree and duration of environmental parameters exceeding the threshold; SC4, weight configuration unit defines the weight coefficient of each environmental parameter, which is used to calculate the comprehensive risk score; SC5, end.

5. The security intelligent decision-making optimization method based on the knowledge graph model according to claim 2 is characterized in that: The specific steps of real-time data collection and processing in SA3 are as follows: SD1, start; SD2, sensor interface unit collects environmental parameters in real time through temperature and humidity sensors and dust concentration monitors; SD3, the data cleaning unit performs denoising and standardization on the collected data using a median filter denoising algorithm to remove abnormal data; SD4, the data storage unit stores the processed data into the time series database, with a sampling interval of ≤5 minutes; SD5, end.

6. A security intelligent decision-making optimization method based on a knowledge graph model according to claim 2, characterized in that: The specific steps of risk assessment and analysis performed by the risk assessment module in SA4 are as follows: SE1, start; SE2, the data acquisition unit obtains the latest environmental parameter data from the time series database; SE3, the knowledge query unit retrieves the risk points and regulatory clauses corresponding to the current environmental parameters based on the "trigger relationship between environmental parameters and risk points" in the knowledge graph; SE4, the risk calculation unit calculates the individual risk score of each risk point and calculates the comprehensive risk score based on the weight coefficient; SE5, the risk level determination unit determines the current risk level; SE6, end.

7. The security intelligent decision-making optimization method based on the knowledge graph model according to claim 2 is characterized in that: The specific steps for generating the intelligent decision report in SA5 are as follows: SF1, start; SF2, the emergency plan retrieval unit calls the corresponding emergency plan from the "configuration relationship between emergency plans and emergency resources" in the knowledge graph according to the current risk level; SF3, the measure matching unit matches preventive measures based on the parameter anomaly type through the "control relationship between regulatory clauses and risk points" in the knowledge graph; SF4, the report generation unit generates a decision report containing references to regulatory clauses, risk analysis, preventive measures, and emergency plans; SF5, the resource scheduling unit generates resource scheduling suggestions based on the emergency resource allocation relationship; SF6, end.

8. The security intelligent decision-making optimization method based on the knowledge graph model according to claim 7 is characterized in that: The generation rules of the preventive measures in the SF5 are as follows: the threshold configuration unit sets default thresholds for the environmental parameters of temperature, humidity, ventilation ratio, and dust concentration, which can be adjusted manually; The default temperature threshold is 30°C. When the temperature exceeds the threshold, the air conditioning system is activated and the stacking layout is adjusted to enhance ventilation. The default humidity threshold is 65%. When the humidity exceeds the threshold, the dehumidifier is started and the storage seal is checked; The default ventilation threshold is 0.6 times / h. When the ventilation ratio is lower than the threshold, mechanical ventilation equipment is added and the stacking method is optimized; The default dust threshold is 10mg / m³. When the dust concentration exceeds the threshold, the dust removal equipment will be activated and the cleaning frequency will be increased.

9. The security intelligent decision-making optimization method based on the knowledge graph model according to claim 7 is characterized in that: The steps for generating the decision report in SF5 include: SF5.1, start; SF5.

2. Extract all regulatory clauses associated with the current risk point in the knowledge graph; SF5.

3. Sort the regulatory clauses by importance and relevance; SF5.

4. Generate compliance analysis and indicate the specific clauses that the current environmental parameters violate; SF5.

5. Generate prioritized preventive measures based on the severity of the violation; SF5.

6. Generate executable emergency plans based on the real-time status of emergency resources; SF5.7, end.

Images