Coal mine safety prevention and control method and system based on collaboration of intelligent AI and edge data

Abstract

The present invention belongs to the field of coal mine safety control, and in particular relates to a coal mine safety control method and system based on the collaboration of intelligent AI and edge data, comprising: first, constructing a first-level vertical risk management chain based on the coal mine risk management rank, and constructing a three-level coal mine plane monitoring chain in combination with the mining area distribution and the monitoring network; then, using historical risk type data, mining status and major main risk types at the mine level to cluster and divide the monitoring chain nodes, setting edge assessment nodes and configuring the model, and obtaining the mapping of the second-level risk assessment chain and the vertical chain through the graph algorithm; then, from the process positions and control measures of the monitoring nodes, a library of key position control measures is obtained through algorithms such as knowledge graphs and mapped to each chain to form a three-level coal mine risk assessment and control network; finally, real-time monitoring data is obtained, and dynamic analysis and evaluation is performed on this network to realize upper-level risk control and lower-level process position control, thereby effectively improving the level of coal mine safety control.

Description

Technical Field

[0001] The present invention belongs to the field of coal mine safety control, and in particular relates to a coal mine safety control method and system based on the collaboration of intelligent AI and edge data. Background Art

[0002] With the development of information technology, coal mine safety control is gradually moving towards intelligent and information-based management. However, current coal mine safety control technology still faces many challenges. On the one hand, the coal mine production environment is complex and harsh, and traditional safety monitoring methods cannot comprehensively, accurately, and in real time obtain information on various risk factors. For example, in some remote mining areas or those with complex geological conditions, the deployment of sensors and data transmission may be restricted, resulting in the inability to timely grasp the safety status of some areas. On the other hand, the existing risk identification and control system is not perfect, lacking accurate quantification and classification of risks and corresponding efficient control measures. In the risk control process, the division of responsibilities between different levels is not clear, and coordination is insufficient, which easily leads to control loopholes or duplication of control.

[0003] For example, the Chinese patent with authorization announcement number CN105117857B discloses a dynamic assessment and diagnosis system and method for coal mine safety, which includes a server, an operation terminal, a network switching device, and an underground monitoring device. The server and the operation terminal located above the well are connected to the underground monitoring device through a connecting device. The underground monitoring device is a detection device for detecting underground roof pressure and fire-proof gas data; the underground monitoring device transmits the collected data to the server and the operation terminal via the network switching device, and displays the results on the operation terminal after information processing.

[0004] For example, the Chinese patent application with publication number CN109034612A discloses a safety risk diagnosis method based on a coal mine early warning analysis and prevention and control system, which includes the following steps: using an XML file to define and store a calculation model based on a coal mine safety risk assessment index system; using an ETL tool to regularly extract indicator data values ​​from each data source based on the meaning of the indicators in the indicator system definition, and storing them in static, annual, monthly, and daily indicator data tables in the data warehouse; obtaining the score of the leaf node indicator item based on the data value of each indicator collected in the data warehouse, and then calculating the score of each indicator of the coal mine safety risk assessment index system; generating a risk rating for each indicator based on the score of each indicator of the coal mine safety risk assessment index system and the rating rules of the overall score, thereby generating a dynamic diagnosis report regularly.

[0005] The above-mentioned existing technologies have the following problems: Although the existing technologies have also provided many coal mine risk assessment methods, they are vague in terms of information transmission, hierarchical management and responsibility correspondence, and are unable to achieve rapid response management based on line and rank, and there are also certain deficiencies in the corresponding risk division. For this reason, the present invention provides a coal mine safety prevention and control method and system based on the collaboration of intelligent AI and edge data. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the present invention proposes a coal mine safety control method and system based on the collaboration of intelligent AI and edge data, including: first, constructing a first-level vertical risk management chain according to the coal mine risk management level, and constructing a three-level coal mine plane monitoring chain in combination with the mining area distribution and the monitoring network; then using historical risk type data, mining status and major main risk types at the mine level to cluster the monitoring chain nodes, set up edge evaluation nodes and configure the model, and obtain the second-level risk assessment chain and vertical chain mapping through the graph algorithm; then, the key position control measures library is obtained from the monitoring node process positions and control measures through algorithms such as knowledge graphs and mapped to each chain to form a three-level coal mine risk assessment and control network; finally, the monitoring data is obtained in real time, and through dynamic analysis and evaluation of this network, the upper-level risk control and lower-level process position control are realized, effectively improving the level of coal mine safety control.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] Coal mine safety prevention and control methods based on the collaboration of intelligent AI and edge data include:

[0009] S1. Based on the configured coal mine risk management level, a first-level vertical risk management chain is established. At the same time, based on the distribution of coal mining areas and the configured monitoring network, a third-level coal mine horizontal monitoring chain is established.

[0010] S2. Cluster each monitoring node in the coal mine plane monitoring chain according to the historical risk type data of the corresponding sub-nodes in the coal mine plane monitoring chain, the coal mine mining status, and the specified major risk types of the mine level, and obtain N major risk cluster monitoring node sets;

[0011] S3. Set up N edge assessment nodes and configure a hierarchical holographic modeling assessment model and assessment main risk type for each edge assessment node. At the same time, each edge assessment node is connected to the corresponding main risk cluster monitoring node set through a one-to-many mapping by assessing the main risk type. Based on all edge assessment nodes, a secondary risk assessment chain is obtained, and the secondary risk assessment chain is connected and mapped to the primary vertical risk management chain through the configured risk score-management level mapping relationship.

[0012] S4. Based on the process positions corresponding to each monitoring node in the monitoring network and the control measures configured for each process position, a key position control measures library is obtained through the knowledge graph and graph database algorithm;

[0013] S5. Map the key post control measures database to the first-level vertical risk management chain, the second-level risk assessment chain, and the third-level coal mine plane monitoring chain to obtain a three-level coal mine risk assessment and control network;

[0014] S6. Obtain data from each monitoring node in real time, conduct dynamic analysis and evaluation of the monitoring data through the three-level coal mine risk assessment and control network, and feed the evaluation results back to the first-level vertical risk management chain for upper-level job-level risk control and lower-level process position control.

[0015] Specifically, the major major risks at the mine level include five major risk types: gas, water, fire, roof and electromechanical transportation; the coal mine risk management job levels include coal supervision bureau level, group company level, group company local coal mine level, mine level, coal mine safety department level, district team level and team level; the corresponding job levels at the coal supervision bureau level, group company level, group company local coal mine level, mine level, coal mine safety department level, district team level and team level are as follows: coal supervision bureau level > group company level > group company local coal mine level > mine level > coal mine safety department level > district team level > team level;

[0016] The hierarchical holographic modeling assessment model includes a risk accident assessment sub-model; the risk level standards contained in the risk accident assessment sub-model include: major risk, relatively large risk, general risk and low risk; the level corresponding to major risk is identified as Level I; the level corresponding to relatively large risk is identified as Level II; the level corresponding to general risk is identified as Level III; and the level corresponding to low risk is identified as Level IV.

[0017] Specifically, the risk score-management level mapping relationship is as follows:

[0018] The corresponding risk level standard for the team level is low risk, and the corresponding risk management level score range is [1, 3]; the corresponding risk level standard for the team level is general risk, and the corresponding risk management level score range is [4, 8]; the corresponding risk level standard for the coal mine safety department level is relatively high risk, and the corresponding risk management level score range is [9, 16]; the corresponding risk level standard for the coal supervision bureau level, group company level, group company local coal mine level, and mine level is major risk, and the corresponding risk management level score range is (20, 40];

[0019] The specific risk level corresponding to the standard interval is: low risk corresponding to the standard interval is [1, 3], general risk corresponding to the standard interval is [4, 8], relatively high risk corresponding to the standard interval is [9, 16], and major risk corresponding to the standard interval is [20, 25].

[0020] The risk management rank score range is obtained by multiplying the risk level assessment score of the corresponding accident and the rank risk management coefficient; the rank risk management coefficient for the corresponding risk situations at the mine level, coal mine safety department level, team level and team level is 1; the rank risk management coefficient corresponding to the group company's local coal mine level is 1.5; the rank risk management coefficient corresponding to the group company level is 2; and the rank risk management coefficient corresponding to the coal supervision bureau level is 2.5.

[0021] Specifically, the hierarchical holographic modeling and evaluation model also includes a risk comprehensive prediction sub-model; the steps for constructing the hierarchical holographic modeling and evaluation model include:

[0022] S301. Based on the historical coal mine accident data and the description of the probability of the accident, construct a set of possibility assessment indicators and possibility level value pairs corresponding to the risk comprehensive prediction sub-model. Specifically, the possibility assessment indicators and corresponding possibility level value pairs include [extremely likely to occur, 5], [very likely to occur, 4], [likely to occur, 3], [relatively unlikely to occur, 2], and [basically unlikely to occur, 1].

[0023] S302. Based on the severity standards of historical accidents, construct an impact assessment indicator set and corresponding assessment level value pairs corresponding to the risk accident assessment sub-model. Specifically, the impact assessment indicator and corresponding assessment level value pairs include [particularly significant impact, 5], [significant impact, 4], [significant impact, 3], [moderate impact, 2], and [very small impact, 1].

[0024] S303. Construct a joint assessment risk matrix based on the evaluation indicators and corresponding possibility level value pairs of the comprehensive risk prediction sub-model and the corresponding evaluation indicators and corresponding possibility level value pairs of the risk accident assessment sub-model.

[0025] Specifically, the steps for constructing the hierarchical holographic modeling evaluation model also include:

[0026] S304: Based on the historical accident data of coal mines, a conditional probability function is constructed for the occurrence of the jth impact assessment indicator when the kth possibility assessment indicator occurs. ;in represents the jth impact evaluation indicator in the impact evaluation indicator set, represents the kth possibility evaluation indicator in the possibility evaluation indicator set;

[0027] S305. According to the joint assessment risk matrix and the conditional probability function, a graded holographic modeling assessment model is constructed through a comprehensive fuzzy algorithm to obtain a risk grade determination assessment score and a corresponding risk grade of the corresponding accident.

[0028] Specifically, the steps for constructing the secondary risk assessment chain include:

[0029] S201. Based on the N main risk cluster monitoring node sets obtained in S2, configure N main edge nodes, and set each main edge node as the main edge evaluation calculation node for each main risk cluster monitoring node set, and build the constructed hierarchical holographic modeling evaluation model into the corresponding node;

[0030] S202: Based on the primary edge evaluation calculation node, a secondary risk assessment chain is constructed using a graph algorithm.

[0031] S203. Based on the interactive monitoring information corresponding to each monitoring node in the three-level coal mine plane monitoring chain, an assessment migration chain of nodes in the secondary risk assessment chain is constructed. If the risk type assessed by the current main edge assessment calculation node is not the major main risk type corresponding to the current main edge assessment calculation node, the corresponding risk assessment task is migrated to the main edge assessment calculation node corresponding to the major main risk type through the assessment migration chain to perform assessment calculation and obtain the risk assessment result. The risk assessment task includes the risk value assessment task and the label mark of the corresponding monitoring node.

[0032] Specifically, process positions include hoisting and transporting, electrical operation, pressure operation, rotating parts, roof management, and auxiliary transportation. The first-level vertical risk management chain is constructed through a graph algorithm based on the coal mine risk management level and the corresponding level size.

[0033] The upper management node in the first-level vertical risk management chain only manages the job level corresponding to the next-level node.

[0034] A coal mine safety control system based on the collaboration of intelligent AI and edge data, including: 8. Evaluation and control network construction module;

[0035] The evaluation and control network construction modules include risk management chain unit, plane monitoring chain unit, monitoring cluster unit, and risk assessment chain unit;

[0036] The risk management chain unit is used to build a first-level vertical risk management chain based on the configured coal mine risk management level;

[0037] Plane monitoring chain unit, used to build a three-level coal mine plane monitoring chain based on the distribution and configuration of the monitoring network in the coal mining area;

[0038] The monitoring clustering unit is used to cluster the monitoring nodes in the coal mine plane monitoring chain according to the historical risk type data of the corresponding sub-nodes in the coal mine plane monitoring chain, the coal mine mining status and the specified major risk types of the mine level, and obtain N main risk cluster monitoring node sets;

[0039] The risk assessment chain unit is used to set up N edge assessment nodes and configure a hierarchical holographic modeling assessment model and an assessment main risk type for each edge assessment node. At the same time, each edge assessment node is connected to the corresponding main risk cluster monitoring node set in a one-to-many mapping manner through the assessment main risk type, and a secondary risk assessment chain is obtained based on all edge assessment nodes through a graph algorithm.

[0040] Specifically, the assessment and control network construction module also includes a risk-management mapping unit, a control measures library construction unit, and a risk assessment and control network unit;

[0041] The risk-management mapping unit is used to connect and map the secondary risk assessment chain with the primary vertical risk management chain through the configured risk score-management level mapping relationship;

[0042] The control measures library construction unit is used to obtain the key position control measures library through the knowledge graph and graph database algorithm based on the corresponding process positions of each monitoring node in the monitoring network and the control measures configured for each process position;

[0043] The risk assessment and control network unit is used to map the key post control measures library to the first-level vertical risk management chain, the second-level risk assessment chain and the third-level coal mine plane monitoring chain to obtain a three-level coal mine risk assessment and control network.

[0044] A computer-readable storage medium, characterized in that computer instructions are stored thereon, which, when the computer instructions are run, execute a coal mine safety prevention and control method based on the collaboration of intelligent AI and edge data.

[0045] Compared with the prior art, the present invention has the following beneficial effects:

[0046] In response to the shortcomings of the existing technology, the present invention significantly improves the accuracy and response speed of coal mine risk management by constructing a multi-level coal mine risk assessment and control network. Specifically, first, a first-level vertical risk management chain is constructed according to the configured coal mine risk management level, and a three-level coal mine plane monitoring chain is constructed in combination with the mining area distribution and the monitoring network, forming a comprehensive coverage from high-level management to grassroots monitoring; secondly, by clustering the sub-nodes in the coal mine plane monitoring chain according to historical risk types, mining status and major mine-level main risk types, N main risk cluster monitoring node sets are obtained. This process ensures the scientific nature and pertinence of risk assessment and solves the problem of vague risk division in the existing technology. Thirdly, by obtaining the second-level risk assessment chain and connecting it with the first-level vertical risk management chain, this multi-level and multi-dimensional risk assessment system realizes efficient information transmission and rapid response, ensuring that managers at all levels can grasp risk dynamics in a timely manner and make accurate decisions. Fourthly, based on the process position and its control measures corresponding to each monitoring node, a key position control measures library is established. This step not only refines risk prevention and control measures but also provides clear operational guidelines for each process position, enhancing the feasibility of actual operations. Finally, the key position control measures library is mapped to a three-level risk management chain, forming a three-tier coal mine risk assessment and control network. Real-time monitoring data is acquired for dynamic analysis and evaluation, effectively combining upper-level risk control with lower-level process position control. This closed-loop management system ensures transparency of risk information and clear allocation of responsibilities, overcoming the problems of poor information transmission and unclear correspondence between hierarchical management and responsibilities in existing technologies, significantly improving the efficiency and reliability of coal mine risk management. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is a flow chart of a coal mine safety prevention and control method based on intelligent AI and edge data collaboration according to Example 1 of the present invention;

[0048] Figure 2 This is a quantitative indicator diagram of the possibility of an accident occurring in Example 1 of the present invention;

[0049] Figure 3 This is a quantitative indicator diagram for accident assessment according to Example 1 of the present invention;

[0050] Figure 4 This is a module diagram of the coal mine safety control system based on the collaboration of intelligent AI and edge data in Example 2 of the present invention. DETAILED DESCRIPTION

[0051] Example 1

[0052] See also Figure 1 The present invention provides an embodiment of a coal mine safety control method based on the collaboration of intelligent AI and edge data, comprising the following steps:

[0053] S1. Based on the configured coal mine risk management level, a first-level vertical risk management chain is established. At the same time, based on the distribution of coal mining areas and the configured monitoring network, a third-level coal mine horizontal monitoring chain is established.

[0054] Furthermore, in this embodiment, the coal mine risk management job levels include coal supervision bureau level, group company level, group company local coal mine level, mine level, coal mine safety department level, district team level, and team level; the job levels corresponding to coal supervision bureau level, group company level, group company local coal mine level, mine level, coal mine safety department level, district team level, and team level are as follows: coal supervision bureau level > group company level > group company local coal mine level > mine level > coal mine safety department level > district team level > team level;

[0055] Furthermore, in this embodiment, the three-level coal mine plane monitoring chain is the lowest level in the three-level coal mine risk assessment and control network, and is used to monitor location nodes in various areas of the coal mine;

[0056] Furthermore, in this embodiment, the first-level vertical risk management chain is constructed through a graph algorithm based on the coal mine risk management ranks and the corresponding rank sizes.

[0057] S2. Cluster each monitoring node in the coal mine plane monitoring chain according to the historical risk type data of the corresponding sub-nodes in the coal mine plane monitoring chain, the coal mine mining status, and the specified major risk types of the mine level, and obtain N major risk cluster monitoring node sets;

[0058] Furthermore, the major mine-level risk types specified in this embodiment include five major risk types: gas, water, fire, roof, and electromechanical transportation;

[0059] Historical risk type data include the major risk types with the greatest possibility and the most frequent occurrence in the history of the corresponding monitoring area.

[0060] S3. Set up N edge assessment nodes and configure a hierarchical holographic modeling assessment model and assessment main risk type for each edge assessment node. At the same time, each edge assessment node is connected to the corresponding main risk cluster monitoring node set through a one-to-many mapping by assessing the main risk type. Based on all edge assessment nodes, a secondary risk assessment chain is obtained, and the secondary risk assessment chain is connected and mapped to the primary vertical risk management chain through the configured risk score-management level mapping relationship.

[0061] Furthermore, in this embodiment, the hierarchical holographic modeling assessment model includes a risk accident assessment sub-model; the risk level standards contained in the risk accident assessment sub-model include: major risk, relatively large risk, general risk and low risk; the level identification corresponding to major risk is Level I; the level identification corresponding to relatively large risk is Level II; the level identification corresponding to general risk is Level III; and the level identification corresponding to low risk is Level IV.

[0062] Specifically, the risk score-management level mapping relationship is as follows:

[0063] The corresponding risk level standard for the team level is low risk, and the corresponding risk management level score range is [1, 3]; the corresponding risk level standard for the team level is general risk, and the corresponding risk management level score range is [4, 8]; the corresponding risk level standard for the coal mine safety department level is relatively high risk, and the corresponding risk management level score range is [9, 16]; the corresponding risk level standard for the coal supervision bureau level, group company level, group company local coal mine level, and mine level is major risk, and the corresponding risk management level score range is (20, 40];

[0064] The specific risk level corresponding to the standard interval is: low risk corresponding to the standard interval is [1, 3], general risk corresponding to the standard interval is [4, 8], relatively high risk corresponding to the standard interval is [9, 16], and major risk corresponding to the standard interval is [20, 25].

[0065] The risk management grade score range is obtained by multiplying the risk level assessment score of the corresponding accident by the grade risk management coefficient; the grade risk management coefficient of the corresponding risk situation at the mine level, coal mine safety department level, team level and team level is 1; the grade risk management coefficient corresponding to the local coal mine level of the group company is 1.5; the grade risk management coefficient corresponding to the group company level is 2; the grade risk management coefficient corresponding to the coal supervision bureau level is 2.5. Furthermore, the corresponding grade risk management coefficient in this embodiment is used represents, i represents the i-th management level.

[0066] Further, please refer to Table 1, which is a mapping table of the specific relationship between risk levels and corresponding management ranks in this embodiment, as well as the specific control measures corresponding to each rank. The specific measures in the table show that the upper management node in the first-level vertical risk management chain only manages the ranks corresponding to the nodes in the next level. For example, please refer to the following table, which will not be detailed here.

[0067] Table 1

[0068]

[0069] Furthermore, in this embodiment, please refer to Table 2 for the corresponding relationship between the job level risk management coefficient and the corresponding risk management job level, which will not be described in detail here;

[0070] Table 2

[0071]

[0072] This process significantly improves the accuracy and responsiveness of coal mine risk management by constructing a multi-level coal mine risk assessment and control network. First, a first-level vertical risk management chain was constructed based on seven levels: coal supervision bureau, group company, and mine. Furthermore, a three-level coal mine horizontal monitoring chain was constructed, combining the distribution of mining areas and the monitoring network. This multi-level risk management structure ensures comprehensive coverage from the highest regulatory level to the grassroots working surfaces, resolving the issue of poor information transmission in existing technologies. Secondly, based on historical risk type data (e.g., five major types of risks: gas, water, fire, roof, and electromechanical transportation) and current mining status, the nodes in the coal mine horizontal monitoring chain were clustered and divided to obtain a set of N primary risk cluster monitoring nodes. This scientific classification method makes risk assessment more accurate and avoids the problem of ambiguous risk classification in traditional methods. Third, N edge assessment nodes were set up and configured with a hierarchical holographic modeling assessment model. Through the risk accident assessment sub-model, each edge assessment node was connected to the corresponding set of primary risk cluster monitoring nodes in a one-to-many mapping fashion, forming a two-level risk assessment chain. The risk score-management rank mapping ensured that different ranks could quickly respond to corresponding levels of risk, achieving a differentiated and hierarchical rapid response management model. For example, the team level was responsible for low-level risks (Level IV), the district team level was responsible for general risks (Level III), the coal mine safety department level was responsible for greater risks (Level II), and major risks (Level I) were handled by higher ranks, such as the Coal Supervision Bureau and the group company level. This clear division of responsibilities improved response efficiency. In addition, by introducing job-level risk management coefficients (such as the mine-level coefficient is 1, the group company's local coal mine-level coefficient is 1.5, the group company-level coefficient is 2, and the coal supervision bureau-level coefficient is 2.5), the system can more flexibly adjust the risk management efforts of different job levels to ensure that high-risk situations are handled promptly and effectively; in summary, this process not only solves the problems of unclear hierarchical management and responsibility correspondence in existing technologies by constructing a sophisticated multi-level risk assessment system, but also greatly improves the efficiency and reliability of coal mine risk management. The real-time dynamic analysis and evaluation mechanism ensures that managers at all levels can respond quickly to risk changes, ensure the safety and stability of coal mine operations, and provide strong technical support for safe production in the coal mining industry.

[0073] Furthermore, in this embodiment, the hierarchical holographic modeling and evaluation model also includes a risk comprehensive prediction sub-model; the steps of constructing the hierarchical holographic modeling and evaluation model include:

[0074] S301. Based on the historical coal mine accident data and the description of the probability of the accident, construct a set of possibility assessment indicators and possibility level value pairs corresponding to the risk comprehensive prediction sub-model. Specifically, the possibility assessment indicators and corresponding possibility level value pairs include [extremely likely to occur, 5], [very likely to occur, 4], [likely to occur, 3], [relatively unlikely to occur, 2], and [basically unlikely to occur, 1].

[0075] Further, see Figure 2 , the quantitative indicators of the possibility of an accident include the possibility level value, i.e., level, possibility assessment index and description of the possibility of the accident. For the description of the possibility of the accident, please refer to Figure 2 , I will not describe it here;

[0076] S302. Based on the severity standards of historical accidents, construct an impact assessment indicator set and corresponding assessment level value pairs corresponding to the risk accident assessment sub-model. Specifically, the impact assessment indicator and corresponding assessment level value pairs include [particularly significant impact, 5], [significant impact, 4], [significant impact, 3], [moderate impact, 2], and [very small impact, 1].

[0077] See also Figure 3 , the quantitative indicators of the consequences of accidents include the evaluation level value, i.e. level, impact assessment index and description of the severity of the consequences of accidents. For the description of the severity of the consequences of accidents, please refer to Figure 3 , I will not describe it here;

[0078] S303: construct a joint assessment risk matrix based on the evaluation indicators and corresponding possibility level value pairs of the comprehensive risk prediction sub-model and the corresponding evaluation indicators and corresponding assessment level value pairs of the risk accident assessment sub-model;

[0079] S304: Based on the historical accident data of coal mines, a conditional probability function is constructed for the occurrence of the jth impact assessment indicator when the kth possibility assessment indicator occurs. ;in represents the jth impact evaluation indicator in the impact evaluation indicator set, represents the kth possibility evaluation indicator in the possibility evaluation indicator set;

[0080] S305, based on the joint assessment risk matrix and conditional probability function ,Through the hierarchical holographic modeling and evaluation model constructed through a comprehensive fuzzy ,algorithm, the risk level judgment and evaluation score of the corresponding accident and the ,corresponding risk level are obtained.

[0081] For example, to better illustrate the assessment process of each major risk type in this embodiment, please refer to the specific practical example in Table 3. As described in the table, it includes the serial number (i.e., the number of the accident), the hazard source (for major risk types, fire and gas risk types are used as an example here), the risk description, and the risk management level score calculation process, where M represents the possibility level value and N represents the assessment level value.

[0082] Table 3

[0083]

[0084] Furthermore, in this embodiment, the steps of constructing the secondary risk assessment chain include:

[0085] S201. Based on the N main risk cluster monitoring node sets obtained in S2, configure N main edge nodes, and set each main edge node as the main edge evaluation calculation node for each main risk cluster monitoring node set, and build the constructed hierarchical holographic modeling evaluation model into the corresponding node;

[0086] S202: Based on the primary edge evaluation calculation node, a secondary risk assessment chain is constructed using a graph algorithm.

[0087] S203. Based on the interactive monitoring information corresponding to each monitoring node in the three-level coal mine plane monitoring chain, an assessment migration chain of nodes in the secondary risk assessment chain is constructed. If the risk type assessed by the current main edge assessment calculation node is not the major main risk type corresponding to the current main edge assessment calculation node, the corresponding risk assessment task is migrated to the main edge assessment calculation node corresponding to the major main risk type through the assessment migration chain to perform assessment calculation and obtain the risk assessment result. The risk assessment task includes the risk value assessment task and the label mark of the corresponding monitoring node.

[0088] S4. Based on the process positions corresponding to each monitoring node in the monitoring network and the control measures configured for each process position, a key position control measures library is obtained through the knowledge graph and graph database algorithm;

[0089] Furthermore, in this embodiment, the process positions include transporting and hoisting, electrical operation, pressure operation, rotating parts, top management and auxiliary transportation;

[0090] Among them, the corresponding control measures for transportation and lifting include:

[0091] (1) In addition to normal installation and dismantling operations on the working surface and the transportation system in the transportation area, the night shift is engaged in transportation operations;

[0092] (2) When lifting heavy objects, the connection must be reliable. No one is allowed to approach the bottom of the heavy objects or the direction where they may fall. Ensure that the retreat route is clear.

[0093] (3) The wire rope of the winch at the top of the inclined tunnel is strictly prohibited from exceeding the hook removal point to prevent the trailer from breaking due to excessive rope;

[0094] (4) The inclined tunnel is closed for transportation. The upper entrance vehicle must pass the slope change point to enter the inclined tunnel and the wire rope must be tightened before the inclined tunnel pneumatic door stop (anti-runaway device) can be opened;

[0095] The control measures for electrical operations include:

[0096] Electricians must be proficient in the power-off command procedures. Electrical work must be performed wearing insulated boots or on an insulated platform. Main circuit electrical work at voltage levels above 36V (except for small electrical equipment) must be performed with a three-phase grounding wire. The power-off system of "whoever shuts off the power, puts up the sign, whoever removes the sign, then restores the power" must be strictly implemented. Live maintenance and moving electrical equipment are strictly prohibited. Non-professional personnel are prohibited from operating electromechanical equipment without authorization. TBMs must be parked on flat, dry, and well-supported ground during maintenance.

[0097] The corresponding control measures for pressurized operation include:

[0098] (1) When operating under pressure, the pressure must be released first and personnel must wear goggles;

[0099] (2) U-shaped clamps must be used in accordance with regulations at the connection between the stop valve and the pipeline;

[0100] (3) An elbow must be installed above or below the stop valve on the main pipe;

[0101] (4) A stop valve must be used at the hose outlet, and a sprinkler must be added when sprinkling water;

[0102] (5) Stop valves, pipe joints, etc. must be protected secondary;

[0103] (6) The stop valve of the main pipe at the mining head must be equipped with a power-assisting sleeve (Note: The power-assisting sleeve is made of 4-point steel pipe, 200mm long, with a φ5mm hole punched at one end. The burrs on the steel pipe are removed and it is hung at the stop valve. Workers using it at other locations must carry it with them);

[0104] The control measures corresponding to rotating parts include:

[0105] For equipment with rotating parts, such as conveyor drums, winch drums, rake guide wheels, endless rope guide wheels, and tail wheels, protective covers (rails) must be installed, safety signs must be made, and inspection, repair, and maintenance of rotating parts must be strengthened. Personnel are strictly prohibited from touching rotating parts when they are in operation.

[0106] The corresponding control measures for gang top management include:

[0107] Strictly implement the system of knocking on the wall and asking about the top throughout the entire construction process; the rock tunnel excavation working face must strictly implement the measures of hanging protective nets on the face; all workers entering the unsupported top side must wear leggings and anti-smashing back armor; standardize the installation of roof delamination meters, the establishment of tunnel displacement measurement stations and data collection and analysis, equipment maintenance and replacement; standardize the construction, inspection and recording of anchor rods (cables); focus on hidden dangers such as "rotten nets, broken anchors, and water sheets" on the roof during inspections of in-use tunnels; actively promote fault advance grouting treatment, face-mounted mechanical protection devices, etc.

[0108] The corresponding control measures for auxiliary transportation include:

[0109] Implementation of relevant regulations for large-scale tunnel transportation management, including regulations for overhead passenger devices, inclined tunnel walkers, undulating tunnel passenger devices, and monorail cranes, standardizes maintenance and protective testing. Strengthen management of inclined tunnel transportation, ensuring that safety facilities are complete, that audio and visual signals are sensitive and reliable, and that electrical equipment is in good working order. Promote the networking of monorail cranes, video display systems, stepless speed regulation, and the separation of pedestrians and vehicles.

[0110] S5. Map the key post control measures database to the first-level vertical risk management chain, the second-level risk assessment chain, and the third-level coal mine plane monitoring chain to obtain a three-level coal mine risk assessment and control network;

[0111] S6. Obtain data from each monitoring node in real time, conduct dynamic analysis and evaluation of the monitoring data through the three-level coal mine risk assessment and control network, and feed the evaluation results back to the first-level vertical risk management chain for upper-level job-level risk control and lower-level process position control.

[0112] This process significantly improves the accuracy, response speed and execution efficiency of coal mine risk management by constructing a multi-level and multi-dimensional coal mine risk assessment and control system. First, the hierarchical holographic modeling assessment model not only includes a risk accident assessment sub-model, but also introduces a risk comprehensive prediction sub-model. Through historical accident data and accident probability descriptions, a detailed set of possibility assessment indicators and impact assessment indicators are constructed, and a comprehensive quantitative assessment of risks is achieved through the joint assessment of risk matrices and conditional probability functions. This refined risk assessment method ensures the scientificity and accuracy of risk judgment. Secondly, by configuring the main edge node and the assessment migration chain, a secondary risk assessment chain is constructed, so that each monitoring node can dynamically adjust the risk assessment task according to the actual situation to ensure that major risk types are given priority. This mechanism This system not only improves assessment efficiency but also enhances the flexibility and adaptability of the system, resolving the problems of poor information transmission and ambiguous responsibility allocation in existing technologies. Thirdly, based on knowledge graphs and graph database algorithms, the system has established a library of key control measures covering key process positions such as lifting, electrical operation, and pressure operation. These specific and clear control measures provide standardized operating guidelines for each position, enhancing the feasibility and safety of actual operations. Finally, in S5 and S6, the key position control measure library is mapped to the first-level vertical risk management chain, the second-level risk assessment chain, and the third-level coal mine plane monitoring chain, forming a three-level coal mine risk assessment and control network. Monitoring node data is acquired in real time for dynamic analysis and evaluation, and the results are fed back to higher-level positions for risk control and specific operational guidance for lower-level process positions, achieving seamless integration from high-level management to grassroots operations. This closed-loop management system ensures transparency of risk information and clarity of responsibilities, significantly improving the efficiency and reliability of coal mine risk management.

[0113] Example 2

[0114] See also Figure 4 , another embodiment provided by the present invention: a coal mine safety control system based on the collaboration of intelligent AI and edge data, comprising: an evaluation and control network construction module and a real-time evaluation and control module;

[0115] The assessment and control network construction module is used to construct and configure the risk assessment and control network of the third-level coal mine; the assessment and control network construction module includes the risk management chain unit, the plane monitoring chain unit, the monitoring cluster unit, the risk assessment chain unit, the risk management mapping unit, the control measure library construction unit and the risk assessment and control network unit;

[0116] The risk management chain unit is used to build a first-level vertical risk management chain based on the configured coal mine risk management level;

[0117] Plane monitoring chain unit, used to build a three-level coal mine plane monitoring chain based on the distribution and configuration of the monitoring network in the coal mining area;

[0118] The monitoring clustering unit is used to cluster the monitoring nodes in the coal mine plane monitoring chain according to the historical risk type data of the corresponding sub-nodes in the coal mine plane monitoring chain, the coal mine mining status and the specified major risk types of the mine level, and obtain N main risk cluster monitoring node sets;

[0119] The risk assessment chain unit is used to set up N edge assessment nodes and configure a hierarchical holographic modeling assessment model and the main risk type for each edge assessment node. At the same time, each edge assessment node is connected to the corresponding main risk cluster monitoring node set through a one-to-many mapping through the main risk type, and a secondary risk assessment chain is obtained based on all edge assessment nodes through a graph algorithm;

[0120] The risk-management mapping unit is used to connect and map the secondary risk assessment chain with the primary vertical risk management chain through the configured risk score-management level mapping relationship;

[0121] The control measures library construction unit is used to obtain the key position control measures library through the knowledge graph and graph database algorithm based on the corresponding process positions of each monitoring node in the monitoring network and the control measures configured for each process position;

[0122] The risk assessment and control network unit is used to map the key post control measures library to the first-level vertical risk management chain, the second-level risk assessment chain, and the third-level coal mine plane monitoring chain, thereby obtaining a three-level coal mine risk assessment and control network.

[0123] The real-time assessment and control module is used to obtain data from each monitoring node in real time, dynamically analyze and evaluate the monitoring data through the three-level coal mine risk assessment and control network, and feed back the assessment results to the first-level vertical risk management chain for upper-level job level risk control and lower-level process position control.

[0124] Example 3

[0125] An electronic device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, it implements a coal mine safety control method based on the collaboration of intelligent AI and edge data.

[0126] A computer-readable storage medium stores computer instructions, which, when executed, execute a coal mine safety prevention and control method based on the collaboration of intelligent AI and edge data.

[0127] The embodiments of the present invention are described above in conjunction with the accompanying drawings, but the present invention is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present invention, ordinary technicians in this field can also change, modify, replace and modify the above-mentioned embodiments without departing from the scope of protection of the purpose of the present invention and the claims, and all of these are protected by the present invention.

[0128] If the technical solution disclosed herein involves personal information, the product using the technical solution disclosed herein has clearly informed the individual of the personal information processing rules and obtained the individual's voluntary consent before processing the personal information. If the technical solution disclosed herein involves sensitive personal information, the product using the technical solution disclosed herein has obtained the individual's separate consent before processing the sensitive personal information and at the same time meets the "explicit consent" requirement. For example, on personal information collection devices such as cameras, a clear and prominent sign is set to inform that the individual has entered the personal information collection scope and that personal information will be collected. If the individual voluntarily enters the collection scope, it is deemed that they agree to the collection of their personal information; or on the personal information processing device, when the personal information processing rules are notified by obvious signs / information, the individual's authorization is obtained through pop-up information or by asking the individual to upload their personal information. The personal information processing rules may include information such as the personal information processor, the purpose of personal information processing, the processing method, and the type of personal information processed.

Claims

1. A coal mine safety control method based on the collaboration of intelligent AI and edge data, characterized by: include: S1. Based on the configured coal mine risk management level, a first-level vertical risk management chain is established. At the same time, based on the distribution of coal mining areas and the configured monitoring network, a third-level coal mine horizontal monitoring chain is established. S2. Cluster each monitoring node in the coal mine plane monitoring chain according to the historical risk type data of the corresponding sub-nodes in the coal mine plane monitoring chain, the coal mine mining status, and the specified major risk types of the mine level, and obtain N major risk cluster monitoring node sets; S3. Set up N edge assessment nodes and configure a hierarchical holographic modeling assessment model and assessment main risk type for each edge assessment node. At the same time, each edge assessment node is connected to the corresponding main risk cluster monitoring node set through a one-to-many mapping by assessing the main risk type. Based on all edge assessment nodes, a secondary risk assessment chain is obtained, and the secondary risk assessment chain is connected and mapped to the primary vertical risk management chain through the configured risk score-management level mapping relationship. S4. Based on the process positions corresponding to each monitoring node in the monitoring network and the control measures configured for each process position, a key position control measures library is obtained through the knowledge graph and graph database algorithm; S5. Map the key post control measures database to the first-level vertical risk management chain, the second-level risk assessment chain, and the third-level coal mine plane monitoring chain to obtain a three-level coal mine risk assessment and control network; S6. Obtain data from each monitoring node in real time, conduct dynamic analysis and evaluation of the monitoring data through the three-level coal mine risk assessment and control network, and feed the evaluation results back to the first-level vertical risk management chain for upper-level job-level risk control and lower-level process position control.

2. The coal mine safety control method based on intelligent AI and edge data collaboration as claimed in claim 1, characterized in that: The major mine-level risk types stipulated in the regulations include five major risk types: gas, water, fire, roof and electromechanical transportation; the coal mine risk management job levels stipulated in the regulations include coal supervision bureau level, group company level, group company's local coal mine level, mine level, coal mine safety department level, district team level and team level; the job levels corresponding to the coal supervision bureau level, group company level, group company's local coal mine level, mine level, coal mine safety department level, district team level and team level are as follows: coal supervision bureau level > group company level > group company's local coal mine level > mine level > coal mine safety department level > district team level > team level; The hierarchical holographic modeling assessment model includes a risk accident assessment sub-model; The risk level standards included in the risk accident assessment sub-model include: major risk, relatively large risk, general risk and low risk; the level corresponding to the major risk is identified as Level I; the level corresponding to the relatively large risk is identified as Level II; the level corresponding to the general risk is identified as Level III; and the level corresponding to the low risk is identified as Level IV.

3. The coal mine safety control method based on intelligent AI and edge data collaboration as claimed in claim 2, characterized in that: The risk score-management level mapping relationship is specifically as follows: The risk level standard corresponding to the team level is low risk, and the corresponding risk management level score range is [1, 3]; the risk level standard corresponding to the team level is general risk, and the corresponding risk management level score range is [4, 8]; the risk level standard corresponding to the coal mine safety department level is relatively high risk, and the corresponding risk management level score range is [9, 16]; the risk level standard corresponding to the coal supervision bureau level, group company level, group company local coal mine level, and mine level is major risk, and the corresponding risk management level score range is (20, 40]; The risk level corresponding to the judgment standard interval is specifically: low risk corresponding to the judgment standard interval is [1, 3], general risk corresponding to the judgment standard interval is [4, 8], relatively high risk corresponding to the judgment standard interval is [9, 16], and major risk corresponding to the judgment standard interval is [20, 25]; The risk management grade score range is obtained by multiplying the risk level assessment score of the corresponding accident and the grade risk management coefficient; the grade risk management coefficient for the corresponding risk situations at the mine level, coal mine safety department level, team level and team level is 1; the grade risk management coefficient corresponding to the group company's local coal mine level is 1.5; the grade risk management coefficient corresponding to the group company level is 2; and the grade risk management coefficient corresponding to the coal supervision bureau level is 2.

5.

4. The coal mine safety control method based on intelligent AI and edge data collaboration as claimed in claim 3, characterized in that: The hierarchical holographic modeling and evaluation model also includes a risk comprehensive prediction sub-model; the steps of constructing the hierarchical holographic modeling and evaluation model include: S301. Based on the historical coal mine accident data and the description of the probability of the accident, construct a set of possibility assessment indicators and possibility level value pairs corresponding to the risk comprehensive prediction sub-model, wherein the possibility assessment indicators and corresponding possibility level value pairs include [extremely likely to occur, 5], [very likely to occur, 4], [possible to occur, 3], [relatively unlikely to occur, 2], and [basically unlikely to occur, 1]; S302. Based on the severity standards after historical accidents, construct an impact assessment indicator set and a corresponding assessment level value pair corresponding to the risk accident assessment sub-model, wherein the impact assessment indicator and the corresponding assessment level value pair include [extremely significant impact, 5], [significant impact, 4], [significant impact, 3], [moderate impact, 2], and [very small impact, 1]; S303. Construct a joint assessment risk matrix based on the evaluation indicators and corresponding possibility level value pairs of the comprehensive risk prediction sub-model and the corresponding evaluation indicators and corresponding possibility level value pairs of the risk accident assessment sub-model.

5. The coal mine safety control method based on intelligent AI and edge data collaboration as claimed in claim 4, characterized in that: The steps of constructing the hierarchical holographic modeling evaluation model include: S304: Based on the historical accident data of coal mines, a conditional probability function is constructed for the occurrence of the jth impact assessment indicator when the kth possibility assessment indicator occurs. ;in represents the jth impact evaluation indicator in the impact evaluation indicator set, represents the kth possibility evaluation indicator in the possibility evaluation indicator set; S305. According to the joint assessment risk matrix and the conditional probability function, a graded holographic modeling assessment model is constructed through a comprehensive fuzzy algorithm to obtain a risk grade determination assessment score and a corresponding risk grade of the corresponding accident.

6. The coal mine safety control method based on intelligent AI and edge data collaboration as claimed in claim 5, characterized in that: The steps for constructing the secondary risk assessment chain include: S201. Based on the N main risk cluster monitoring node sets obtained in S2, configure N main edge nodes, and set each main edge node as the main edge evaluation calculation node for each main risk cluster monitoring node set, and build the constructed hierarchical holographic modeling evaluation model into the corresponding node; S202: Based on the primary edge evaluation calculation node, a secondary risk assessment chain is constructed using a graph algorithm. S203. Based on the interactive monitoring information corresponding to each monitoring node in the three-level coal mine plane monitoring chain, an assessment migration chain of nodes in the secondary risk assessment chain is constructed. If the risk type assessed by the current main edge assessment calculation node is not the major main risk type corresponding to the current main edge assessment calculation node, the corresponding risk assessment task is migrated to the main edge assessment calculation node corresponding to the major main risk type through the assessment migration chain to perform assessment calculation and obtain the risk assessment result; the risk assessment task includes a risk value assessment task and a label mark of the corresponding monitoring node.

7. The coal mine safety control method based on intelligent AI and edge data collaboration as claimed in claim 6, characterized in that: The process positions include transporting and lifting, electrical operation, pressure operation, rotating parts, top management and auxiliary transportation; The first-level vertical risk management chain is constructed through a graph algorithm based on the coal mine risk management ranks and the corresponding rank sizes; the upper-level management node in the first-level vertical risk management chain only manages the ranks corresponding to the lower-level nodes.

8. A coal mine safety control system based on intelligent AI and edge data collaboration, which is implemented based on the coal mine safety control method based on intelligent AI and edge data collaboration according to any one of claims 1 to 7, characterized in that: include: Evaluate the control network building modules; The evaluation and control network construction module includes a risk management chain unit, a plane monitoring chain unit, a monitoring cluster unit, and a risk evaluation chain unit; The risk management chain unit is used to build a first-level vertical risk management chain according to the configured coal mine risk management level; The plane monitoring chain unit is used to construct a three-level coal mine plane monitoring chain based on the monitoring network distributed and configured in the coal mining area; The monitoring clustering unit is used to cluster the monitoring nodes in the coal mine plane monitoring chain according to the historical risk type data of the corresponding sub-nodes in the coal mine plane monitoring chain, the coal mine mining status and the specified mine-level major risk types, and obtain N main risk cluster monitoring node sets; The risk assessment chain unit is used to set up N edge assessment nodes and configure a hierarchical holographic modeling assessment model and an assessment main risk type for each edge assessment node. At the same time, each edge assessment node is connected to the corresponding main risk cluster monitoring node set through a one-to-many mapping through the assessment main risk type, and a secondary risk assessment chain is obtained based on all edge assessment nodes through a graph algorithm.

9. The coal mine safety control system based on intelligent AI and edge data collaboration as claimed in claim 8, characterized in that: The assessment and control network construction module also includes a risk-management mapping unit, a control measures library construction unit and a risk assessment and control network unit; The risk-management mapping unit is used to connect and map the secondary risk assessment chain with the primary vertical risk management chain through the configured risk score-management rank mapping relationship; The control measures library construction unit is used to obtain a key position control measures library through a knowledge graph and a graph database algorithm based on the process positions corresponding to each monitoring node in the monitoring network and the control measures configured for each process position; The risk assessment and control network unit is used to map the key post control measures library to the first-level vertical risk management chain, the second-level risk assessment chain and the third-level coal mine plane monitoring chain respectively, to obtain a three-level coal mine risk assessment and control network.

10. A computer-readable storage medium, characterized in that Computer instructions are stored thereon, and when the computer instructions are executed, the coal mine safety prevention and control method based on the collaboration of intelligent AI and edge data as described in any one of claims 1 to 7 is executed.

Images