Whole mine safety situation analysis and evaluation system and method

Abstract

The invention relates to a whole mine safety situation analysis and evaluation system and method, and belongs to the technical field of coal mine safety production and management. The system comprises a basic data informatization module, a security big data center module, a professional dimension scoring module, a spatial dimension scoring module and a risk visualization analysis module. The basic information informatization module is used for establishing a basic information informatization database according to professions based on a coal mine safety production standardization management system; the security big data center module is used for integrating multi-source heterogeneous data, namely managing, monitoring and analyzing data, and establishing a data unified interaction standard; the professional dimension scoring module is used for carrying out risk assessment by adopting a security check table method for each professional field; and the spatial dimension scoring module is used for realizing automatic, real-time and quantitative evaluation based on quasi-static and dynamic monitoring data by adopting a virtual object situation-based scoring method for typical scenes. According to the invention, the whole mine overall risk and local key area safety risk evaluation can be realized.

Description

Technical Field

[0001] The present invention belongs to the technical field of coal mine safety production and management, and relates to a safety situation analysis and evaluation system and method for the whole coal mine. Background Technique

[0002] There are many, wide-ranging and dynamically changing safety risk points in coal mines, and the requirements for safety control are high with great difficulty. Risks exist in multiple aspects such as unsafe behaviors of people, unsafe states of equipment, unsafe environmental factors and defects in management measures. For this reason, the "Notice on Supporting and Encouraging the Research and Development and Application of Coal Mine Intelligent Technology and Equipment" points out that the key is to research and develop intelligent monitoring and early warning of safety hazards and intelligent emergency rescue for disasters, and focus on carrying out the research and development and application of coal mine intelligent software systems. The "Coal Mine Intelligent Construction Guide (2021 Edition)" points out that the key is to build intelligent early warning systems for major safety hazards, intelligent safety monitoring systems, etc., to achieve personnel reduction, safety increase and efficiency improvement.

[0003] There are many, wide-ranging and dynamically changing safety risk points in coal mines, and the requirements for safety control are high with great difficulty. Risks exist in multiple aspects such as unsafe behaviors of people, unsafe states of equipment, unsafe environmental factors and defects in management measures. Coal mining enterprises have established business systems such as single-disaster monitoring, early warning, intelligent ventilation, and hidden danger management for gas, hydrology, etc. according to their respective actual disaster situations, providing important support for disaster prevention and management. However, the types of safety risks are complex, the forms are diverse, and they affect each other. It is urgent to integrate multi-dimensional information such as "people, machines, environment, and management" to form a large-scale safety intelligent control system.

[0004] In the prior art, the invention patent with the publication number CN110852552A discloses a coal mine safety risk assessment method based on big data. This method mainly groups the personnel of coal mining enterprises, calculates the importance ranking of the rating groups through questionnaires, uses the CESK method for risk assessment to form initial risk data, and then analyzes the initial risk data according to the risk assessment system to obtain the risk assessment results and disposal suggestions. This assessment method mainly uses questionnaires for assessment, which has subjectivity and one-sidedness, and the assessment results are not accurate enough.

[0005] Therefore, in order to realize the large-scale safety intelligent control system, the following problems need to be solved urgently:

[0006] (1) The construction technology of large-scale safety control for the whole coal mine.

[0007] (2) The problems of unified modeling and horizontal risk comparison for the safety situation evaluation of working places under different conditions and coal mines.

[0008] (3) For the existing risk evaluation indicators, grading and quantifying scores at the underlying indicators mostly rely on experience, and the scoring criteria lack theoretical basis. It is necessary to solve the problem of objective quantification of scoring for underlying indicators. Summary of the Invention

[0009] In view of this, the purpose of the present invention is to provide a full-mine safety situation analysis and evaluation system and method. Based on the evaluation of the inherent safety risks of the mine, by integrating business subsystems such as personnel, electromechanics, environment, and management related to coal mine safety, a coal mine safety big data center is formed, and a coal mine safety risk quantification evaluation method is established. The overall risks of the full mine and the risks of local key places are analyzed and evaluated from two dimensions of the spatial line and the professional line, laying a foundation for the targeted control of coal mine safety.

[0010] To achieve the above object, the present invention provides the following technical solutions:

[0011] A full-mine safety situation analysis and evaluation system specifically includes a basic module, a risk assessment module, and a risk visualization analysis module;

[0012] The basic module includes a basic information digitization module and a safety big data center module; the basic information digitization module is based on the coal mine safety production standardization management system, and a basic information digitization database is established in 15 majors such as mining, tunneling, machinery, ventilation, transportation, geology and survey, etc.; the safety big data center module is to integrate multi-source heterogeneous data, namely management, monitoring, and analysis data, and establish a unified data interaction standard.

[0013] The risk assessment module includes a professional dimension scoring module and a spatial dimension scoring module; the professional dimension scoring module uses the safety checklist method for risk assessment, that is, for each professional field, safety checklists are constructed from 4 aspects including supervision and inspection results, safety production standardization work results, hidden dangers and three-dimensional conditions, and monitoring abnormal perception conditions, and the scores of each professional field are calculated dynamically; the spatial dimension scoring module is for 5 typical scenarios such as coal mining faces, tunneling faces, transportation roadways, electromechanical chambers, and surface industrial squares, and uses a virtual object situation scoring method to achieve automatic, real-time, and quantitative evaluation based on quasi-static and dynamic monitoring data.

[0014] The risk visualization analysis module is to establish a GIS single map and a configuration map of local key places to realize the visual display of the risk analysis and evaluation results.

[0015] Further, the basic information digitization module is a management report based on interface interactive configuration that integrates data source configuration, data object configuration, and data report configuration, and realizes the management of multiple reports through configuration.

[0016] Further, the established unified data interaction standards include safety monitoring data interaction standards, early warning analysis data interaction standards, and electromechanical equipment data interaction standards;

[0017] The safety monitoring data interaction standard converts various types of data into defined data, real-time data, abnormal data, historical data, and statistical data through a protocol adapter;

[0018] The early warning analysis data interaction standard includes unified regional coding, unified early warning events, unified early warning levels, and unified interaction formats;

[0019] The electromechanical equipment data interaction standard realizes equipment abstraction and dynamic attribute extension by adopting the EIP object model; specifically, the equipment abstraction abstracts mine equipment into classes (such as fans and motors) and assigns unique codes to support unified management; the dynamic attribute extension supports the classification management of static attributes (equipment models) and operating attributes (temperature, start / stop status) to adapt to different monitoring scenarios.

[0020] Furthermore, in the virtual object situation scoring method, the safety risks of the entire mine are evaluated from the professional line dimension, the grid subdivision method is adopted to evaluate the safety risks of key areas in the mine from the spatial line dimension, the analytic hierarchy process is used to identify risk elements such as people, machines, environment, and management in each area / specialty of the subdivided area, and a hierarchical model of the evaluation object is established; then, the technique for order preference by similarity to ideal solution is used, combined with threshold values such as alarms and explosions, to construct a virtual evaluation object, and a scoring criterion for each index of the object to be evaluated and each index of the virtual evaluation object is established; at the same time, the analytic hierarchy process is used to determine the subjective weight, the historical safety production standardization inspection results are used as samples, the logistic regression method is used to dynamically calculate the objective weight, and the game theory idea is used to minimize the deviation sum of the subjective weight and the objective weight, and the optimal comprehensive weight is calculated; a combination of linear weighting and key indicator veto is adopted for situation assessment. This method specifically includes the following steps:

[0021] S1: Establish an evaluation hierarchy model by adopting the analytic hierarchy process integrating fine network division;

[0022] S2: Adopt the technique for order preference by similarity to ideal solution to establish a neutral reference evaluation rule and mechanism for each evaluation index of the virtual object, mainly including:

[0023] (1) Determine the critical point between safety and insecurity, that is, the index value of the neutral virtual object;

[0024] (2) According to the risk level division, determine the critical values of different levels, that is, the critical values of each index of virtual objects with different risk levels, the optimal values, that is, the index values of the optimal virtual object, and the worst values, that is, the index values of the worst virtual object;

[0025] (3) Determine the index scoring mechanism between different virtual objects: For different indicators at the same level, it is obviously not entirely applicable to calculate the scores of upper-level indicators using linear weighting. It is necessary to determine the logical relationship between indicators at the same level and the upper-level indicators, including: determining whether there are key indicators, which cannot be measured by weights, and whether there are veto indicators. For example, if there are major hidden dangers in the indicators at this level, the evaluation result of the entire level should be high risk; determining whether the importance of indicators at each level varies under different circumstances. For example, for the five major disasters of gas, water, fire, mine pressure, roof, and dust in a mine, if the same set of weights is used for working faces with different geological conditions, the evaluation result obviously cannot reflect the real situation. Therefore, the following method is used for calculation: When all indicators at a certain level are greater than the safety critical score (such as 60 points), linear weighting is used to calculate the total score; otherwise, the lowest score of all indicators is taken as the final score of this level;

[0026] (4) Determine the importance of indicators; Not all indicators can be obtained. Therefore, according to the importance of indicators, determine which are important indicators. For important indicators, they cannot be deleted during specific evaluation; for unimportant indicators, they can be deleted.

[0027] S3: Use the combined weighting strategy to determine the initial weights of each evaluation indicator in the virtual object evaluation hierarchy model;

[0028] S4: Use the method of dynamic modeling to construct the hierarchy model of specific evaluation objects;

[0029] S5: According to the characteristics of each hierarchy model of virtual reference objects, use methods such as linear weighting, key indicator veto, and dynamic weight combination to specifically evaluate the safety situation of evaluation objects.

[0030] Furthermore, in step S1, use the analytic hierarchy process integrating fine network division to establish the evaluation hierarchy model, which specifically includes the following steps:

[0031] S11: Grid subdivision;

[0032] S12: Identify risk factors of people, machines, environment, and management within the grid;

[0033] On the basis of grid division, according to the different equipment, environment monitoring, and management focuses of each area, sort out the risk factors existing in each area from the four dimensions of people, machines, environment, and management;

[0034] In the personnel dimension, it is mainly constructed from aspects such as work experience, education level, shift leading, employee training, physical condition, operating personnel, certification status, and fatigue operation index; among them, the fatigue operation index reflects the comprehensive working state of operating personnel and is related to historical, current, and recent working hours, night shift working hours, and the interval between two shifts;

[0035] Environmental dimension: Identify the five major disasters, including gas disasters, water disasters, fire, roof damage, and dust;

[0036] Electromechanical dimension: This includes the 50 mandatory electromechanical inspection items in the Coal Mine Safety Regulations, as well as identification of factors such as the electromechanical equipment health index. Electromechanical equipment health analysis focuses on monitoring alarm conditions, equipment parameter changes, and equipment operation and maintenance. It identifies sudden changes in monitoring data, upward or downward trends, and frequent fluctuations beyond the specified range, enabling timely detection of equipment abnormalities and faults.

[0037] Management dimension: Risk factors are identified mainly from the aspects of hidden danger investigation and treatment, three violations and risk control implementation;

[0038] S13: Establish a hierarchical evaluation structure for overall objectives;

[0039] Analyze the relationship between various risk factors in the subdivided areas, delete duplicate indicators, and analyze the systemic risks between each subdivided area and the entire region based on the risk factor analysis of each region, so as to build a risk assessment indicator system for the entire region;

[0040] Specifically, with respect to the overall risk of the mine, with reference to the Basic Requirements and Scoring Method for Coal Mine Safety Standardization (Trial), the mine risk is evaluated from seven major disciplines: coal mining, tunneling, electromechanical, transportation, ventilation, geological disaster measurement and prevention, and comprehensive management. Among them, six disciplines, namely coal mining, tunneling, electromechanical, transportation, ventilation, and geological disaster measurement and prevention, are related to the information system of mine construction. The risk of each discipline is mainly evaluated from four aspects: fixed risk, quarterly inspection score of safety standardization, three-dimensional number of professional hidden dangers, and abnormalities of related monitoring systems. The comprehensive management discipline can be subdivided into nine sub-disciplines: concept and goal and safety commitment of mine manager, organizational structure, safety production responsibility system and safety management system, quality of employees, safety risk classification control, accident hazard investigation and treatment, dispatching and emergency management, occupational disease hazard prevention and control and ground facilities, and continuous improvement. Since these disciplines do not involve online monitoring systems, the risks are mainly analyzed and evaluated from the fixed scores of the discipline itself and the dynamic scores of the company's quarterly inspections.

[0041] In response to the safety risks in each key area, evaluation indicators are mainly constructed from four aspects: personnel, mechanical and electrical, environment, and management.

[0042] Furthermore, in step S2, the critical point between safety and unsafety is determined, which specifically includes:

[0043] (1.1) Critical values based on laws, regulations, industry standards, and expert experience;

[0044] (1.2) Calculate the critical value of on-site data based on the approximation of the ideal point, specifically including:

[0045] ① According to the idea of approximating the ideal point, first, based on the evaluation index system, collect and calculate the parameter values of all indicators for each area to be evaluated, and obtain the distribution of the evaluation index values of different types. Then, determine the optimal and worst values of all possible values of this indicator. For benefit-type indicators, the larger the value, the better, such as the equipment health index; for cost-type indicators, the smaller the value, the better, such as the gas concentration. Obtain the optimal and worst values of each indicator in all samples. The optimal value is the positive ideal solution, and the worst value is the negative ideal solution.

[0046] ② Calculate the sample closeness.

[0047] The closeness mainly represents the degree of closeness between the sample and the most ideal solution.

[0048] The distance between the sample index and the ideal solution is calculated using the following formula:

[0049]

[0050] In the formula: d i + , d i - are the distances between the positive and negative ideal solutions and the calculated indicators; c i + , c i - represents the element value of the ideal solution C + , C - corresponding.

[0051] The formula for calculating the closeness is as follows:

[0052] E i = d i - / (d i + + d i - ) (2)

[0053] In the formula: E i corresponds to the sample being the positive ideal solution and the negative ideal solution when its value is 1 and 0 respectively.

[0054] Sort according to the value of E i , and divide the critical values of each level according to the 80 / 20 principle of safety management, that is, 20% (which can be dynamically adjusted according to the safety management mechanism as needed) is classified as unsafe. At the same time, reverse solve the indicator values corresponding to the critical closeness.

[0055] ③ Calculate the critical value of on-site data based on cluster analysis;

[0056] For some indicators that are neither benefit - type nor cost - type and for which there is no relevant specification reference currently, such as the gas fluctuation shape characteristic indicator, clustering is used for division and determination.

[0057] First, collect the indicator samples and manually divide them into safe samples, relatively safe samples, relatively dangerous samples, or dangerous samples according to the actual situation; then, use the K - means clustering method to determine the center points of each sample as the critical values of different - level virtual objects.

[0058] Furthermore, in step S2, to determine the critical values of each indicator for virtual objects of different risk levels, specifically: for the critical value a with laws, regulations, industry norms, or expert experience, first determine the basic critical value according to a; then, conduct a detailed division based on the on - site data critical value b based on the approximation of the ideal point and the on - site data critical value c based on cluster analysis.

[0059] Furthermore, in step S2, to determine the indicator scoring mechanism between different virtual objects, for the two cases of quantifiable indicators or non - quantifiable indicators, specifically include:

[0060] (3.1) For the continuously varying quantifiable indicator c, the score division of different virtual object sub - intervals can be improved according to the linear interpolation method; the specific operation steps are as follows:

[0061] 1) Select the indicator values of different virtual objects and give their interval scales in the order of absolute safety, safety, relatively safe, relatively dangerous, dangerous, and absolute danger, that is, c1, c2, …, c n , and at the same time give the corresponding scores s1, s2, …, s n ;

[0062] 2) If the actual c value of a certain evaluated object exactly falls within the interval [c i , c j , then its corresponding score s can be calculated according to the linear interpolation method, that is

[0063]

[0064] If the actual c value falls outside the pre - calibrated interval, the linear extrapolation method can be used; when c < c1, assuming c1 < c n , there is

[0065]

[0066] Conversely, when c > c n , assuming c1 < c n , there is

[0067]

[0068] When using linear extrapolation for interpolation, if the obtained score s is less than 0 or greater than 100, it can be directly changed to 0 or 100 according to the proximity;

[0069] When constructing the discrete (c i ,s i ) data pairs of the above virtual objects, the following points need to be noted:

[0070] ① Ensure that the larger the s in (c i ,s i ), the safer it is; when s i = 0, it is equivalent to a situation of extremely unsafe; when s i = 100, it is equivalent to a situation of extremely safe; i

[0071] ② When it is difficult to judge when it is extremely unsafe and extremely safe, try not to make regulations on the c i value when s i = 0 and s i = 100, that is, for most indicators, try not to construct the virtual object index values of absolute safety and absolute danger, and only make regulations on the c i value when s i = 10 and s i = 90.

[0072] Furthermore, in step S3, the combined weighting strategy is to determine the weights of indicators by combining subjective and objective weights, specifically including:

[0073] (1) Subjective weight: determined by the analytic hierarchy process;

[0074] (2) Objective weight: Taking the historical work safety standardization inspection results as samples, using the logistic regression method to dynamically calculate the objective weight;

[0075] The logistic regression method is a prediction model in statistics, which reflects the strength and direction of the relationship between independent variables and dependent variables. The mine safety situation can be understood as the probability P of safety. Characterized by the work safety standard inspection results, the probability of being in danger is 1 - P, and its influencing factors are the evaluation index system. According to the logistic regression method, the following logistic regression function is constructed:

[0076]

[0077] As obtained last time, the probability that the mine is in a safe state is:

[0078]

[0079] Where: P is the probability that the mine is in a safe state, characterized by the score result of the work safety standardization inspection; x1, x2, x3, …, x n are evaluation indicators; a1, a2, …, a n are regression coefficients, i.e., weights.

[0080] (3) Comprehensive weight: To reduce the one-sidedness of subjective and objective weights, based on the game theory idea, the minimum deviation sum of subjective and objective weights is made, and the optimal comprehensive weight is calculated;

[0081]

[0082] Where: w1 is the objective weight, w2 is the subjective weight, and β1 and β2 are the optimal coefficient matrices.

[0083] Furthermore, in step S4, a hierarchical model of a specific evaluation object is constructed by using the dynamic modeling method, which specifically includes the following two methods:

[0084] (1) Using the dynamic modeling method: When the pairwise comparison method is used to determine the relative importance of each criterion element relative to the parent element, if it is already known that a certain element cannot obtain a definite score for a certain evaluation object, then for this object, when making pairwise comparisons of each criterion element, without considering the missing criterion element, recalculate the weights of each element; for the evaluation object that can obtain all criterion scores, still determine the weights of each criterion according to the complete model;

[0085] (2) Using the default value method: This method is relatively simpler, that is, when data is missing, use the default value to replace it; when modeling, default values should be set for all indicators; it is recommended to set default values based on the following two situations:

[0086] ① If the relative importance of the indicator to the parent element is not high and the risk of its exceeding the limit is not great, its default value can be set to the neutral value, so that its score is 50 points;

[0087] ② If the relative importance of the indicator to the parent element is high and the risk of its exceeding the limit is great, its default value can be set to a value that relatively causes unsafe consequences, ensuring that its score is less than 50 points (such as setting the default value to the value corresponding to a score of 33 points (100×1 / 3)). This is actually a punishment measure to encourage the mine side to complete the data in this aspect as soon as possible. The specific default value settings should be set separately according to different indicators.

[0088] The above two methods for dealing with missing data each have their applicable conditions. Generally, when there is a lack of evaluation basis for some criteria in the upper-level criterion layer of the hierarchical model, the dynamic modeling method should be used. If only some indicators in the lowest-level indicator layer of the criterion layer are missing, generally use the default value to replace them.

[0089] The beneficial effects of the present invention are as follows: the system of the present invention includes functions such as safety information management, risk analysis and identification, workplace risk, professional dimension risk, and risk visualization display:

[0090] (1) Safety information management: For paper materials and ledger information related to mine ventilation and three preventions, gas prevention and control, coal dust prevention and control, fire prevention management, operating procedures, risk management, and hidden danger control, the system provides functions such as data upload, information management, and digital retrieval, laying the foundation for the management dimension of safety risk analysis.

[0091] (2) Risks in the workplace: The risks of the coal mining face, tunneling face, transport tunnel, electromechanical chamber, and ground sites of the mine are analyzed and evaluated from the four aspects of man, machine, environment, and management.

[0092] (3) Professional dimension risk: The overall risk of the entire mine is analyzed and evaluated from four aspects: major hidden dangers, inherent risks, safety production standardization inspection, three hidden dangers violations, and monitoring anomalies for the seven professions of coal mining, tunneling, electromechanical, transportation, ventilation, geodesy, and comprehensive management.

[0093] (4) Visualization of risks: Establish a GIS map of the safety risks of the entire mine, and overlay and display multi-dimensional information such as safety monitoring, risk assessment, alarm, fault, and disconnection; establish configuration diagrams of coal mining working faces, tunneling working faces, substations, ventilation rooms, air compressor rooms, water pump rooms, transport tunnels and other places, and combine them with thematic charts to overlay and display information such as risk assessment, online monitoring, and risk hazards.

[0094] The method of the present invention uses multi-source heterogeneous data integration technology to achieve data collection from multiple business subsystems related to coal mine safety, establishes a coal mine safety risk quantitative analysis and evaluation model from two dimensions: spatial line and professional line, realizes the overall risk of the entire mine and the safety risk evaluation of local key areas, and uses a GIS map and a Web configuration map to visualize the monitoring information and risk analysis and evaluation results of the mine at the macro and micro levels through multi-dimensional drilling and step-by-step backtracking, providing a technical means for digital disaster prevention and safety targeted management and control. The present invention has the following advantages:

[0095] (1) The integration of multi-source heterogeneous data on coal mine safety has been achieved, forming a safety big data center to provide data support for risk quantification analysis;

[0096] (2) A set of "spatial + professional" mine safety risk quantitative evaluation index system and model was formed to achieve quantitative evaluation of coal mine safety risks;

[0097] (3) A security risk situation map and a local scenario Web configuration map were constructed to achieve multi-dimensional, multi-level, and multi-scenario penetrating query and aggregation analysis, laying an important foundation for targeted security management and control.

[0098] After comprehensive and detailed system testing and preliminary application, the system of the present invention is convenient to use, has good reliability, and high versatility.

[0099] Other advantages, objectives, and features of the present invention will be described to some extent in the subsequent specification, and to some extent, will be obvious to those skilled in the art based on the study of the following text, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the following specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be described in preferred detail below in conjunction with the drawings, where:

[0101] Figure 1 is the overall architecture of the safety situation analysis and evaluation system for the entire mine;

[0102] Figure 2 is the overall architecture of the coal mine safety situation evaluation method;

[0103] Figure 3 is the interactive data report generation architecture;

[0104] Figure 4 is the overall architecture of multi-source heterogeneous data integration;

[0105] Figure 5 is the standard code set;

[0106] Figure 6 is the composition of standard data for safety monitoring;

[0107] Figure 7 is the content structure of early warning analysis data exchange;

[0108] Figure 8 is the EIP object model association relationship;

[0109] Figure 9 is the gas disaster risk assessment model for the coal mining face in a low-gas mine;

[0110] Figure 10 is the gas disaster risk assessment model for the coal mining face in a high-gas mine;

[0111] Figure 11 is the analysis and evaluation index for the health of major equipment;

[0112] Figure 12 is the overall hierarchical structure of the safety situation evaluation for the entire mine;

[0113] Figure 13 is the quantitative evaluation index system for the safety risks of the coal mining profession;

[0114] Figure 14 is a quantitative evaluation index system for safety risks in coal mining faces;

[0115] Figure 15 is the functional architecture diagram of the safety situation analysis and evaluation system for the whole mine. Specific implementation manners

[0116] The following uses specific specific examples to illustrate the implementation manners of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0117] Please refer to Figures 1 to 15 , an embodiment of the present invention provides a safety situation analysis system for coal mines in the whole mine, including the following modules:

[0118] 1. Basic module

[0119] (1) Basic information informatization module

[0120] Taking the relevant documents of the coal mine safety production standardization management system as a guide, a basic information informatization database is established in 15 majors such as mining, tunneling, machinery, ventilation, transportation, geology and survey, etc., to realize the digitization of disaster prevention and occupational health.

[0121] 1) Basic information informatization: For a large number of informatization management functions involved in coal mine safety, if developed item by item, not only the R & D investment is large, it is not easy to maintain, but also the system redundancy will be increased. Therefore, by developing an automatic generation function of management reports based on interface interactive configuration that integrates data source configuration, data object configuration, and data report configuration, the R & D of multiple report management functions is completed in the form of configuration, improving the flexibility and variable development degree of the platform. The overall technical architecture is as Figure 3 shown.

[0122] 2) Collection of multi-source heterogeneous data in coal mine safety

[0123] 2.1) Overall architecture of data integration

[0124] The unified collection of coal mine safety data and the semantic expression under unified description are the basis for carrying out safety situation awareness analysis. There are many manufacturers involved in the relevant business subsystems of safety monitoring, with a wide variety of data types, significant differences in professionalism, and large differences in data formats. It is difficult to interact between the monitoring data of each family, between the analysis data, and between the monitoring data and the analysis data. Therefore, the project establishes a coal mine safety data interaction standard, constructs a unified data model, and realizes the data fusion and integration of multiple business subsystems related to coal mine safety. The overall architecture is as Figure 4 shown.

[0125] 2.2) Data unified interaction standard

[0126] Referring to the data element standardization rules such as GB / T 18391 Information Technology - Specification and Standardization of Data Elements, JT / T 697 Basic Data Elements for Traffic Information, and WS361 Directory of Health Information Data Elements, the attributes of coal mine safety monitoring data elements are shown in Table 1 below.

[0127] Table 1 Basic attribute table of coal mine safety monitoring data elements

[0128]

[0129] a. Identification number: Uniquely identifies the metadata type.

[0130] b. Chinese name: The Chinese name of the metadata element, metadata subset, or metadata entity.

[0131] c. English name: The English name of the metadata entity, metadata element, or metadata subset, which is the English name of a certain characteristic of the information resource, generally in full English.

[0132] d. Definition: Describes the basic content and attributes of the metadata subset, metadata entity, or metadata element.

[0133] e. Data type: Describes the data type of the metadata subset, metadata entity, or metadata element, and allows a value range description for the values within its value range. For example, composite type, integer type, boolean type, string, date type, etc.

[0134] f. Value range: Describes the range of values that the metadata element can take.

[0135] Based on the data element description specification, a normalized method is used to define and describe the metadata entities and metadata elements of the core metadata of coal mine safety information resources. The attribute descriptions used include: Chinese name, definition, English name, data type, value range, short name, annotation, and the RDF resource description framework and XML markup language are used to represent the description.

[0136] According to the above method, the following is established as Figure 5The shown standard code set of more than 800 items in 2 major categories and 14 sub - categories.

[0137] The data interaction specification includes the safety monitoring data interaction standard, the early warning analysis data interaction standard, and the electromechanical equipment data interaction standard.

[0138] 2.2.1) Safety monitoring data interaction standard

[0139] There are many manufacturers and a wide variety of safety monitoring data designs. Mainly by developing protocol adapters, various types of data are converted into defined data, real - time data, abnormal data, historical data, and statistical data, as Figure 6 shown.

[0140] 2.2.2) Early warning analysis data interaction standard

[0141] The early warning analysis data interaction standard includes four parts: unified area coding, unified early warning event, unified early warning level, and unified interaction format. Data interaction is carried out by means of WebAPI interface calls.

[0142] Unified area coding uniquely identifies an early warning object and is uniformly defined by the platform.

[0143] Unified early warning event makes a unified agreement on the start and end of the early warning event.

[0144] Unified early warning level is divided into 5 levels, namely:

[0145] Normal (no early warning): The early warning condition is not reached and it is in a normal state;

[0146] Blue early warning (Level I): The early warning condition is reached, which is the lowest early warning level and the risk is relatively low;

[0147] Yellow early warning (Level II): The early warning condition is reached and it is more serious than the blue early warning;

[0148] Orange early warning (Level III): The early warning condition is reached and it is more serious than the yellow early warning;

[0149] Red early warning (Level IV): The highest early warning level condition is reached and it is more serious than the orange early warning;

[0150] Unified interaction content and format.

[0151] yThe early warning analysis data exchange content includes early warning definition information, real - time data, and historical data, as Figure 7 shown.

[0152] 2.2.3) Electromechanical equipment data interaction standard

[0153] 2.2.3.1) EIP object model construction method

[0154] The EIP data object model is a model used to describe the internal and external data exchange and sharing of a system. It classifies data into different objects, each object contains a set of related data attributes, and defines the data types of these attributes. This model enables users to better manage and utilize data resources. In the EIP data object model, data objects are the basic units, which represent certain things in the real world, such as customers, products, orders, etc. Data attributes are components of data objects, used to describe the states and characteristics of data objects. Data types define the value ranges and formats of data attributes, such as character type, numeric type, date type, etc. The EIP data object model is the essential embodiment of realizing data standardization and normalization, making the data exchange between different systems more efficient and smooth. Secondly, through this model, data integration and analysis can be achieved, thus providing support for the fusion decision-making analysis of various data in the whole mine.

[0155] The EIP object model is also called the mine equipment model. This model abstracts each piece of equipment in the mine's human-machine-environment-management into a class and assigns a unique code to achieve unified management. Based on the EIP data object idea, this project constructs data models for monitoring objects such as water hazards, fires, mine pressure, and intelligent ventilation according to the idea of "model - object - attribute". The construction idea is as Figure 8 shown. Among them, the EIP model is used to describe a set of collections with the same objects. For example, a ventilation system can be abstracted into an EIP model.

[0156] An EIP object is used to describe a class of monitoring objects with the same attributes. For example, the fans, motors, valves, etc. under the ventilation system. Object attributes are the physical quantities specifically monitored or virtual quantities with physical meanings fitted from physical quantities, such as the start / stop of the fan, equipment temperature, etc. Each attribute contains 6 basic description fields: sequence, read / write attributes (static attribute, writable attribute, readable attribute), importance (general, important, core, internal attribute), category (static attribute, general attribute, operation indication attribute, control class attribute, fault indication class attribute), attribute name, type (gas, temperature, etc.), value range, etc.

[0157] 2.2.3.2) Establishment of the EIP object model

[0158] To achieve an accurate understanding and expression of various monitoring objects, based on the EIP object model construction method, more than 20 EIP object models have been established, realizing the metadata management of various systems, supporting functions such as classification management, batch import / export of point tables, data point legality verification and data governance, standard data release, and supporting the conversion of the encoding of coal mine perception data access specifications and other custom encodings, achieving an accurate understanding and expression of the whole mine's all-element and whole-process data.

[0159] (2) Safety big data center module: Establish a multi-source heterogeneous data integration model for coal mine safety to realize the integration of management data, monitoring data, and analysis data of business subsystems related to coal mine safety.

[0160] 2. Risk Assessment Module

[0161] (1) Professional dimension scoring module

[0162] From a systemic perspective, 15 professional scoring models were established: Based on the Basic Requirements and Scoring Methods for Coal Mine Safety Standardization Management System, 15 professional subsystems of the mine were analyzed in terms of mining, excavation, machinery, communication, transportation, and ground surveying, to dynamically perceive, analyze, and evaluate the safety situation of the entire mine over a period of time.

[0163] (2) Spatial dimension scoring module

[0164] From the spatial dimension, a typical scenario scoring model is established: five typical scenarios are constructed, namely coal mining working face, tunneling working face, transportation tunnel, electromechanical chamber, and ground industrial square. The hierarchical analysis method based on neutral reference objects is used to establish a safety situation evaluation model for typical scenarios from the four dimensions of people, machines, environment, and management.

[0165] (3) Coal mine safety situation analysis method for the entire mine

[0166] 1) Mine-wide safety situation objectives and regional division

[0167] Based on collected safety supervision and inspection data, mine-submitted data, real-time monitoring data, and mine map model data, the mine's overall safety situation is broken down into a safety performance evaluation and direct safety risk analysis, tailored to the needs of different business departments within the mine. The safety performance evaluation focuses on systemic risks, while the direct safety risk analysis is a spatial assessment.

[0168] In terms of professional expertise, specifically safety performance evaluation, the program is divided into three sub-specialties: coal mining, tunneling, electromechanical, transportation, ventilation, geological hazard measurement and prevention, and comprehensive management, in accordance with coal mine safety production standards. Comprehensive management encompasses nine sub-specialties: philosophy and objectives, mine manager's safety commitment, organizational structure, safety production responsibility system and safety management system, employee quality, safety risk classification and control, accident hazard investigation and treatment, dispatching and emergency management, occupational disease prevention and control, surface facilities, and continuous improvement.

[0169] The spatial division is based on the "whole mine - mining area - key work sites" approach. Underground, key areas include the coal mining face, tunneling face, main transport tunnel, auxiliary transport tunnel, underground substation, and drainage pump room. Above ground, key areas include the ventilation room, air compressor room, surface substation, and gas extraction pump station.

[0170] 2) Security Situation Risk Assessment Method Based on Virtual Objects

[0171] For the collected data such as safety supervision and inspection data, data independently submitted by coal mines, real-time monitoring data, and mine map model data, etc., decompose them into safety work situation evaluation and direct safety risk analysis. The safety work situation evaluation is a systematic risk, and mainly uses the safety checklist method for risk assessment, that is, for each professional field, construct a safety checklist from 4 aspects respectively: supervision and inspection results, work results of work safety standardization, hidden dangers and three-dimensional situation, and monitoring abnormal perception situation, and dynamically calculate the scores of each professional field. The direct safety risk mainly refers to the risks of 5 types of key specific places such as coal mining faces, tunneling faces, transportation roads, electromechanical chambers, and surface industrial squares. The present invention proposes a situation scoring method based on virtual objects to achieve automatic, real-time and quantitative evaluation based on quasi-static and dynamic monitoring data.

[0172] Establish different virtual objects according to the different types of evaluation objects divided above. The virtual object has the following characteristics:

[0173] ① The evaluation indicators of the virtual object are the same as those of the object to be evaluated.

[0174] ② The index values of each evaluation indicator of the virtual object are relatively independent with respect to the total evaluation target and do not change with external conditions.

[0175] ③ The index values of each index of the virtual object are the demarcation points between safety and insecurity. For the evaluation of coal mine safety situation, it is generally divided into high risk, relatively high risk, general risk, and low risk. Therefore, the demarcation point between safety and insecurity is usually the index value corresponding to the score demarcation point between relatively high risk and general risk. When using early warning to represent the safety situation, the demarcation point between safety and insecurity is usually the index value corresponding to the score demarcation point when early warning starts to occur.

[0176] ④ The result of comparing each object to be evaluated with the virtual object can represent the score of each evaluation object for the total evaluation target, without the need for pairwise comparison like the analytic hierarchy process.

[0177] The specific steps of the situation scoring method based on virtual objects are as follows:

[0178] 2.1) Establish an evaluation hierarchy model using the analytic hierarchy process integrated with fine network division.

[0179] For evaluation objects of different types and under different conditions, the present invention does not attempt to establish a unified evaluation hierarchy model, but rather establishes a hierarchical structure that is generally consistent but differs in specific hierarchical structures and index details. The so-called general consistency means that the evaluation objectives are the same, but the specific indicators may vary. Specifically, for the same evaluation objective, it can be characterized by different secondary criteria, but the results they reflect are consistent and can be compared horizontally. For example, when evaluating the gas risk of a coal mining face in a low-gas coal seam, its evaluation indicators are as Figure 9 shown. When evaluating the gas risk of a coal mining face in a high-gas mine, its evaluation indicators are as Figure 10 shown.

[0180] The construction steps of the evaluation hierarchy model are as follows:

[0181] ① Grid subdivision: Taking the coal mining face as an example, combined with the coal mining production process, on the basis of the operation cycle chart, the static-level grid is defined as 5 regions. The working face is divided into 5 regions in the face length direction, namely the return airway end region, the intake airway end region, the middle coal mining region, the head triangular coal region, and the tail triangular coal region. The middle coal mining region can be further subdivided. Combining the three processes of coal cutting, support moving, and scraper conveyor pushing during the production process, the middle coal mining region is further dynamically grid-divided during the production process, divided into coal cutting region, support moving region, scraper conveyor pushing region, and transition region, etc. The heading face is divided into the heading machine head region, the middle region of the heading roadway, and the intake port region according to the working face position. The main haulage roadway is divided into the machine head region, the machine tail region, the belt middle region, etc.

[0182] ② Identification of risk factors of people, machines, environment, and management within the grid

[0183] On the basis of grid division, according to the different equipment, environment monitoring, and management focuses of each region, the risk factors existing in each region are sorted out from the four dimensions of people, machines, environment, and management.

[0184] In the dimension of personnel, it is mainly constructed from aspects such as work experience, education level, follow-up and leading, employee training, physical condition, operating personnel, certification status, and fatigue operation index. Among them, the fatigue operation index reflects the comprehensive working state of operating personnel and is related to the working hours of history, current, and recent, the working hours of night shifts, and the interval between two shifts.

[0185] Environment dimension: Element identification is carried out from 5 major disasters such as gas disasters, water hazards, fires, roof falls, and dust.

[0186] Electromechanical dimension: including the 50 electromechanical inspection items in the Coal Mine Safety Regulations, and the electromechanical equipment health index and other factors for identification. The electromechanical equipment health analysis mainly analyzes the monitoring alarm situation, equipment parameter changes, equipment operation and maintenance, etc., identifies the characteristics of monitoring data mutation, rising or falling trends, frequent fluctuation range crossing, etc., and promptly discovers abnormal equipment fault information. Figure 11 shown.

[0187] Management dimension: Risk factors are mainly identified from the aspects of hidden danger investigation and control, three violations, and risk control implementation.

[0188] ③Establishment of overall goal level evaluation structure

[0189] Analyze the relationship between various risk factors in the subdivided areas, delete duplicate indicators, and based on the risk factor analysis of each area, analyze the systemic risks between each subdivided area and the entire area, so as to construct a risk assessment indicator system for the entire area.

[0190] Specifically, the overall risk assessment for a mine is based on the "Basic Requirements and Scoring Method for Coal Mine Safety Standardization (Trial)" and is conducted across seven major disciplines: coal mining, tunneling, electromechanical engineering, transportation, ventilation, geological hazard measurement and prevention, and comprehensive management. Six of these disciplines are related to the mine's information technology system. The risk assessment for each discipline is primarily based on four key dimensions: fixed risk, scores from quarterly safety standardization inspections, the three-dimensional number of hidden dangers within the discipline, and anomalies in related monitoring systems. The comprehensive management discipline is further subdivided into nine sub-disciplines: philosophy and goals, mine manager's safety commitment; organizational structure; safety responsibility system and safety management system; employee quality; safety risk grading and control; accident hazard investigation and management; dispatching and emergency management; occupational disease prevention and control and surface facilities; and continuous improvement. Since these disciplines do not involve online monitoring systems, risk assessment is primarily based on the discipline's own fixed scores and dynamic scores from the company's quarterly inspections.

[0191] In response to the safety risks in each key area, evaluation indicators are mainly constructed from four aspects: personnel, mechanical and electrical, environment, and management.

[0192] Coal mine safety risk quantitative evaluation index system Figure 12 shown.

[0193] Taking the coal mining specialty as an example, if any of the following situations related to coal mining occur in the coal mining specialty: overproduction beyond capacity, intensity, or quota as defined in the "Criteria for Judging Major Accident Hidden Dangers in Coal Mines", out-of-bounds mining, or the use of equipment and technologies explicitly prohibited or phased out, then the entire risk assessment of the coal mining specialty will be 0 points. If there are no major hidden dangers in the coal mining specialty, then the fixed risks of the mine, the quarterly inspection of work safety standardization, the number of hidden dangers and violations, and the abnormalities of the monitoring system will be considered. For the fixed risk score of the mine, it is analyzed from four aspects: the number of coal mining faces in the mine, the stability of the coal seam, the classification of the coal seam roof, and the coal mining technology; for the score of the quarterly inspection of work safety standardization, the score of the coal mining specialty in the latest inspection is taken as the risk score; for the number of hidden dangers and violations, the existing hidden dangers and violations related to the coal mining specialty are analyzed and scored in the form of negative deductions; for the abnormalities of the monitoring system, the alarms, disconnections, failures, early warnings, etc. related to the roof monitoring system, fully mechanized mining system, roof disaster early warning system, video monitoring system, etc. related to the coal mining specialty are calculated and scored in the form of negative deductions. The risk assessment indicators for the coal mining specialty are as Figure 13 shown.

[0194] Spatial dimension risk assessment indicators: Taking the coal mining face as an example, the risks are evaluated from four aspects: personnel, electromechanics, environment, and management. For personnel risks, the evaluation is carried out from aspects such as the work experience, education level, leadership and shift leading, employee training rate, physical health status, fatigue operation, number of underground workers, and certificate holding; for electromechanics risks, the evaluation is carried out from three aspects: the degree of intelligence of the coal mining face, equipment monitoring and alarm, and equipment data status; for environmental risks, the evaluation is carried out from five aspects: gas disasters, water hazards, fires, roof, and dust in the coal mining face; for management risks, the evaluation is carried out from aspects such as the compilation, review, publicity, and implementation of the operation regulations, risk control, hidden danger treatment, violation situations, and monthly production index completion in the coal mining face. The risk assessment indicators for the coal mining face are as Figure 14 shown.

[0195] 2.2) Adopt the method of integrating and approaching the ideal point to establish the neutral reference evaluation rules and mechanisms for each evaluation index of the virtual object.

[0196] It mainly includes:

[0197] 2.2.1) Determine the critical point between safety and insecurity (the index value of the neutral virtual object).

[0198] 2.2.2) According to the situation of risk level division, determine the critical values of different levels (the index values of virtual objects with different risk levels), the optimal values (the index values of the optimal virtual object), and the worst values (the index values of the worst virtual object).

[0199] The reference evaluation rules for each evaluation index are established with reference to industry laws and regulations, industry norms, expert experience, and on-site data.

[0200] 2.2.2.1) Critical values based on laws, regulations, industry standards, and expert experience

[0201] First, collect the index values corresponding to the indicators that clearly describe the critical states such as safe, relatively safe, relatively dangerous, and dangerous in laws, regulations, industry standards, and expert experience. For example, taking the value range of the gas pressure index when evaluating the outburst risk of coal seams as an example, according to the Interim Provisions on the Standardization of Coal Mine Gas Drainage, the gas pressure compliance value for outburst coal seams is 0.74 MPA. Therefore, 0.74 MPA is selected as the critical value between relatively safe and dangerous for gas pressure.

[0202] 2.2.2.2) Calculation of critical values of on-site data based on the technique for order preference by similarity to an ideal solution

[0203] ① According to the idea of the technique for order preference by similarity to an ideal solution, first, based on the evaluation index system, collect and calculate the parameter values of all indicators for all regions to be evaluated, and obtain the distributions of various evaluation index values of different types. Then, determine the optimal and worst values of all possible values of this index. For benefit-type indicators, the larger the value, the better, such as the equipment health index; for cost-type indicators, the smaller the value, the better, such as the gas concentration; obtain the optimal and worst values of each indicator in all samples. The optimal value is the positive ideal solution. The worst value is the negative ideal solution.

[0204] ② Calculate the sample closeness. The closeness mainly represents the degree of closeness between the sample and the most ideal solution.

[0205] The distance between the sample index and the ideal solution is calculated using the following formula:

[0206]

[0207] where: d i + , d i - are the distances between the positive and negative ideal solutions and the calculated index; c i + , c i - represents the element value corresponding to the ideal solution C + , C - .

[0208] The formula for calculating the closeness is as follows:

[0209] E i = d i - / (d i + + d i - ) (2)

[0210] where: E i When the value is 1 and 0, it corresponds to the positive ideal solution and negative ideal solution of the sample respectively.

[0211] According to E i value, perform sorting, and divide the critical values of each level according to the 80 / 20 principle of safety management, that is, 20% (which can be dynamically adjusted according to the safety management mechanism as needed) is divided into unsafe. At the same time, reverse solve the index value corresponding to the critical closeness.

[0212] 2.2.2.3) Calculation of critical values of on-site data based on cluster analysis

[0213] For some indicators that are neither benefit type nor cost type and there is no relevant specification reference currently, such as the gas fluctuation shape characteristic index, clustering is used for division and determination.

[0214] First, collect the index samples and manually divide them into safe samples, relatively safe samples, relatively dangerous samples, and dangerous samples according to the actual situation. Then use the K-means clustering method to determine the center points of each sample as the critical values of different-level virtual objects.

[0215] 2.2.2.4) Determine the critical values of each index of different-level virtual objects according to the following method

[0216] For the critical value a with laws, regulations, industry specifications, and expert experience, first determine the basic critical value according to a; then make a detailed division according to the on-site data critical value b based on approaching the ideal point and the on-site data critical value c based on cluster analysis.

[0217] For example, the "Coal Mine Safety Regulations" stipulate that the gas concentration alarm threshold at the return air corner of the coal mining face is 1.0%, the power-off threshold is 2.0%, and the gas concentration explosion critical limit is 5%; therefore, it can be determined that the corresponding value of 5% is the worst critical value, and the score is 0; 1.5 is the critical value for danger, and the score is 50 (corresponding to the critical value for danger); 1.0 is the relatively dangerous critical value, and the score is 65 (corresponding to the relatively dangerous critical value); and according to the on-site data critical value c based on cluster analysis, it is calculated that 80% of the gas concentration is below 0.1%, then 0.1 is used as the relatively safe critical value, and the score is 80 points.

[0218] 2.2.2.4) Index scoring mechanism between different virtual objects

[0219] The following will explain for two cases of quantifiable indicators or non-quantifiable indicators.

[0220] For the continuously changing quantifiable indicator c, the score division of different virtual object sub-intervals can be improved according to the linear interpolation method.

[0221] The steps of this method are:

[0222] ① Select different virtual object indicator values and give their interval scales in the order of absolutely safe, safe, relatively safe, relatively dangerous, dangerous, and absolutely dangerous, such as c1, c2,…, cn, and give the scores s1, s2,…, sn corresponding to each value.

[0223] ② If the actual c value of an evaluated object is exactly in [c i ,c j ] interval, the corresponding score can be calculated according to the linear interpolation method, that is,

[0224]

[0225] If the actual c value falls outside the pre-calibrated range, linear extrapolation can be used. <c1(假定c1<c n )

[0226]

[0227] On the contrary, when c>c n (Assuming c1 <c n )

[0228]

[0229] When using linear extrapolation, the score s may be less than 0 or greater than 100. In this case, it can be directly converted to 0 or 100.

[0230] In constructing the above virtual object discrete (c i ,s i ) When aligning data, please note the following points:

[0231] ① Guarantee (c i ,s i ) in i The bigger, the safer; when s i = 0, which is an extremely unsafe situation; when s i =100, which is equivalent to an extremely safe situation.

[0232] ② When it is difficult to judge when it is extremely unsafe or extremely safe, try not to i =0 and s i =100 when c i The value is stipulated, that is, for most indicators, try not to construct absolutely safe virtual object indicator values and absolutely dangerous virtual object indicator values, and only for s i =10 and s i = c when 90 iSpecify the value.

[0233] 2.2.3) For different indicators at the same level, if the scores of the upper-level indicators are obtained by linear weighting, it is obviously not fully applicable. It is necessary to determine the logical relationship of the indicators at the same level relative to the upper-level indicators. This includes: determining whether there are key indicators, which cannot be measured by weights, but are veto indicators. For example, once there are major hidden dangers in the indicators at this level, the evaluation result of the entire level should be high risk; determining whether the indicators at each level are relatively important indicators under different circumstances. For example, for the five major disasters of gas, water, fire, mine pressure, roof, and dust in a mine, if a set of weights is used for working faces with different geological conditions, the evaluation result obviously cannot reflect the real situation. Therefore, the following method is used for calculation: when all the indicators at this level are greater than the safety critical score (such as 60 points), the total score is calculated by linear weighting; otherwise, the lowest score of all the indicators is taken as the final score of this level.

[0234] 2.2.4) Determine the importance of the indicators. Not all indicators can be obtained. Therefore, according to the importance of the indicators, determine which are important indicators. For important indicators, they cannot be deleted during specific evaluation; for unimportant indicators, they can be deleted.

[0235] 2.3) Adopt a combined weighting strategy to determine the initial weights of each evaluation indicator in the virtual object evaluation hierarchy model.

[0236] Virtual object evaluation indicator weighting strategy: Determine the weights of the indicators by combining subjective and objective weights.

[0237] 2.3.1) Subjective weight: The subjective weight is determined by the analytic hierarchy process.

[0238] 2.3.2) Objective weight: Taking the historical work safety standardization inspection results as samples, use the logistic regression method to dynamically calculate the objective weight. The logistic regression method is a prediction model in statistics that reflects the strength and direction of the relationship between independent variables and dependent variables. The mine safety situation can be understood as the probability P of safety. Characterized by the work safety standard inspection results, the probability of being in danger is 1 - P, and its influencing factors are the evaluation index system. According to the logistic regression method, the following logistic regression function is constructed:

[0239]

[0240] As obtained from the previous step, the probability that the mine is in a safe state is:

[0241]

[0242] In the formula: P is the probability that the mine is in a safe state, characterized by the work safety standardization inspection score result; x1, x2, x3,..., x nis the evaluation index; a1, a2, …, a n are the regression coefficients, i.e., the weights.

[0243] (3) Comprehensive weight: To reduce the one-sidedness of subjective and objective weights, based on the game theory idea, the deviation sum between the subjective weight and the objective weight is minimized, and the optimal comprehensive weight is calculated.

[0244]

[0245] In the formula: w1 is the objective weight, w2 is the subjective weight, and β1 and β2 are the optimal coefficient matrices.

[0246] 2.4) The hierarchical model of the specific evaluation object is constructed by using the dynamic modeling method.

[0247] Coal mine safety evaluation involves many aspects. To conduct a complete evaluation, users need to collect, organize, and input a large amount of data. In the initial stage of the work, this will be a difficult task, and it is almost certain that the situation of incomplete input data will be encountered. If the evaluation results cannot be seen due to incomplete data, it will be a frustrating thing and will reduce people's enthusiasm for this work. Therefore, it is necessary to take certain measures to obtain relatively reasonable evaluation results even in the case of incomplete data. For this problem, the following two coping methods are proposed.

[0248] 2.4.1) Using the dynamic modeling method

[0249] When using the pairwise comparison method to determine the relative importance of each criterion element relative to the parent element, if it is already known that a certain element cannot obtain a definite score for a certain evaluation object, then for this object, when making pairwise comparisons of each criterion element, without considering the missing criterion element, recalculate the weights of each element. For the evaluation object that can obtain all criterion scores, still determine the weights of each criterion according to the complete model.

[0250] 2.4.2) Using the default value method

[0251] This method is relatively simpler, that is, when data is missing, use the default value to replace it. When modeling, default values should be set for all indicators. It is recommended to set default values based on the following two situations:

[0252] 2.4.2.1) If the relative importance of the indicator to the parent element is not high and the risk of its exceeding the limit is not large, its default value can be set to the neutral value, so that its score is 50 points.

[0253] 2.4.2.2) If the relative importance of an indicator to its parent element is high and the risk of exceeding the limit is large, its default value can be set to a value that relatively causes unsafe consequences to ensure that its score is less than 50 points (for example, set the default value to the value corresponding to a score of 33 points (100×1 / 3)). This is actually a punishment measure to encourage the mining party to complete the data in this area as soon as possible. The specific setting of the default value should be set separately according to different indicators.

[0254] The above two methods for dealing with missing data each have their applicable conditions. Generally, when there is a lack of evaluation basis for some criteria in the upper-level criterion layer of the hierarchical model as a whole, the dynamic modeling method should be used. If only some indicators in the lowest-level indicator layer of the criterion layer are missing, the default value is generally used to replace them.

[0255] 2.5) According to the characteristics of each hierarchical model of the virtual reference object, methods such as linear weighting, key indicator veto, and dynamic weight combination are used to specifically evaluate the safety situation of the evaluation object.

[0256] 3. Risk Visualization Analysis Module: Establish a GIS integrated map and a configuration map of local key sites to realize the visual display of the risk analysis and evaluation results.

[0257] (1) GIS Integrated Map of Coal Mine Safety Risks Based on Spatial Lines

[0258] 1) Visualization Framework of GIS Integrated Map

[0259] The GIS integrated map mainly includes a front-end web page based on Openlayers, a GIS server, a spatial database, and external interfaces. The general process of data display is that the front-end sends a request through Openlayers, GeoServer accepts the request, calls postgresql to query data, performs operation and analysis through postGIS, returns the result from postgresql to the GeoServer server, and then passes it to Openlayers after rendering or processing, and finally presents it to the user.

[0260] ① Front-End Web Page Based on Openlayers

[0261] The front-end interface is used for GIS data synchronization, attribute supplementation, etc. On the one hand, it obtains business data from external interfaces and geographical data from the GIS server respectively, and generates the final display graph by integrating the two types of data; on the other hand, it receives the editing operations of users and uploads the editing results to the GIS server for display. With the map generated by the openlayers api as the core, it displays the already rendered roadway map obtained from the GIS server, and shows various types of equipment and working faces underground according to the filtering of the layer control tree.

[0262] ② GIS Server

[0263] The GIS server is a bridge connecting the front-end web page and the spatial database. Different from the amorphous data structure and front-end and back-end interaction methods of general programs, the format of GIS data is relatively complex and fixed. Therefore, a general GIS server program that does not require programming came into being and gradually developed unique common functions, such as rendering vector data into raster images and publishing them.

[0264] The Geoserver server publishes data interfaces for equipment, roadways, and working faces externally, providing editing and query functions; it publishes call interfaces for generating personnel trajectories and generating roadway topologies; by writing slg rendering styles, it publishes the rendered roadway images.

[0265] ③Spatial database

[0266] Postgis is a spatial data extension of postgresql. On the one hand, it provides the geometry data format for storing point, line, and surface vector data with spatial references. On the other hand, it also provides hundreds of sql query methods for spatial operations, which can be used for functions such as spatial data conversion, relational operations, processing, and editing. Combined with the function type of the postgresql database and the pgsql language, Postgis has become the only currently internally programmable spatial database.

[0267] In this module, the Postgis spatial database undertakes two major functions: spatial data storage and spatial operations. The supported geometry spatial data format is the guarantee for conveniently and completely accessing equipment positions, roadway trends, and working ranges, and it is also the basis for performing spatial operations.

[0268] ④External interface

[0269] The external interfaces of the one-map module are all provided on the web page, mainly to adapt to multi-system calls and reduce coupling. They are divided into data acquisition interfaces and function call interfaces: the data acquisition interface is a webservice interface that returns JSON data, including equipment definition and monitoring data interfaces, and working face definition and monitoring data interfaces; the function call interface is a postMessage interface for front-end interaction, including interfaces for highlighting equipment, personnel trajectory playback, equipment click event callback interfaces, etc.

[0270] 2) Visual display of security risks based on the GIS one-map

[0271] Display the safety situation analysis and evaluation results according to the spatial line of "mine - mining area - working face", in terms of personnel dimension, electromechanical dimension, environmental dimension, and management dimension. At the same time, taking a GIS map as the carrier, overlay and display monitoring and alarm data, disaster early warning data, risk and hidden danger data, drawing and document data, regional anomaly data, etc.

[0272] ① Basic interaction functions

[0273] The basic interaction functions include functions such as zooming in and out, perspective conversion, coordinate display, and attribute display, which facilitate the exploration of any location and any perspective of the three - dimensional geological body by the staff. Functions such as zooming in and out and perspective conversion can be operated using the mouse, and the operation needs to be achieved by moving the position of the camera component in WebGL. The position where the "camera" is located is the position where the "observer" is located, and the picture taken by the "camera" is the picture displayed on our screen. The coordinate display and attribute display functions can be achieved by listening to the mouse position. By listening to the mouse position and combining with the text display code, the coordinate information and attribute information are displayed.

[0274] ② Visualization of local scenes in the area

[0275] Through model processing technology, realize the three - dimensional dynamic rendering display and 3D visual simulation of local scenes such as coal mining faces, tunneling faces, electromechanical chambers, and main haulage roadways in coal mines. Combine the monitoring and analysis data, draw a data dashboard, realize the integration of static models and data - driven, and construct a realistic and credible digital twin model of the local scene, laying a foundation for the visualization display of the safety situation in key areas.

[0276] ③ Visualization of roadway engineering

[0277] The roadway engineering scene realizes the real - time monitoring and prediction of the roadway structure and operation status. Using WebGL and Web3D visualization technologies, present the digital twin of roadway engineering as an intuitive and three - dimensional scene, providing visualization support for mine safety management. It has functions such as roadway positioning, clicking, and attribute query.

[0278] ④ Visualization of the safety situation of coal mining faces

[0279] The visualization of the safety situation of the coal mining face first marks the location of the coal mining face on a single GIS map, configures the local three-dimensional scene model of the coal mining face area, and realizes the visual display of the roadheader, hydraulic support, intake airway, return airway, etc. The overall scene of the whole mine and the local scene of the coal mining face are connected in series by clicking the mouse and zooming with the scroll wheel. When entering the local scene of the coal mining face, the safety situation analysis and evaluation results of the personnel dimension, electromechanical dimension, environmental dimension, and management dimension of the coal mining face are displayed in the form of a special topic dashboard. At the same time, according to the monitoring, analysis, and management data of the safety monitoring system, transparent geology system, disaster warning system, and safety information management system integrated on the platform, the monitoring abnormal data such as alarms, faults, and disconnections, the geological analysis data such as geological structure belts and faults crossed, the disaster warning data such as gas, fire, and roof, and the management data such as risk points, hidden danger points, three violations of personnel, and regulations and measures are comprehensively and integrally displayed.

[0280] ⑤ Visualization of the safety situation of the tunneling face

[0281] Since there are relatively few relevant subsystems in the tunneling face, mainly consider the water supply, local air supply (local fan) system related to tunneling, and abnormal data such as monitoring alarms of personnel, management, and environment related to the tunneling specialty. Consistent with the processing flow of the coal mining face, the tunneling face also needs to mark the spatial position on a single GIS map, and at the same time configure the local three-dimensional scene model of the tunneling face area to realize the visual display of the roadheader and the heading face. The overall scene of the whole mine and the local scene of the tunneling face are connected in series by clicking the mouse and zooming with the scroll wheel. When entering the local scene of the tunneling face, the safety situation analysis and evaluation results of the personnel dimension, electromechanical dimension, environmental dimension, and management dimension of the tunneling face are displayed in the form of a special topic dashboard. At the same time, according to the monitoring, analysis, and management data of the safety monitoring system, water supply system, local fan system, and safety information management system integrated on the platform, the monitoring abnormal data such as alarms, faults, and disconnections, and the management data such as risk points, hidden danger points, three violations of personnel, and regulations and measures are comprehensively and integrally displayed.

[0282] ⑥ Visualization of the safety situation of the main haulage roadway

[0283] The main haulage roadway mainly presents comprehensive visual displays based on the monitoring data of the main transportation system, combined with regional personnel positioning data, safety management data, disaster warning result data, etc. For the visualization of the safety situation, first mark the position of the haulage roadway on a single GIS map, configure the local 3D scene model of the haulage roadway area, and realize the visual display of belt conveyors and transportation roadways. The overall mine scene and the local scene of the haulage roadway are connected in series by clicking the mouse and zooming with the scroll wheel. When entering the local scene of the haulage roadway, use the form of a special topic dashboard to display the safety situation analysis and evaluation results in terms of personnel dimension, electromechanical dimension, environmental dimension, and management dimension of the haulage roadway. At the same time, according to the monitoring, analysis, and management data of the safety monitoring system, transportation monitoring system, disaster warning system, and safety information management system integrated on the platform, comprehensively integrate and display monitoring abnormal data such as alarms, faults, and disconnections, belt fire warning data, and management data such as risk points, potential hazard points, personnel violations of regulations, and procedures and measures.

[0284] ⑦ Visualization of the safety situation in the electromechanical chamber

[0285] The visualization of the safety situation in the electromechanical chamber includes the visualization of the safety situation in the underground substation and the visualization of the safety situation in the pump house.

[0286] The visualization of the safety situation in the underground substation is based on the local scene of the underground substation chamber. On this basis, superimpose and display the safety situation analysis and evaluation results in terms of personnel dimension, electromechanical dimension, environmental dimension, and management dimension of the substation. Consistent with the previous typical scenes, first mark the position of the substation on a single GIS map, configure the local 3D scene model of the substation area, and realize the visual display of comprehensive protection equipment and chamber areas. The overall mine scene and the local scene of the substation are connected in series by clicking the mouse and zooming with the scroll wheel. At the same time, according to the monitoring, analysis, and management data of the power monitoring system, disaster warning system, and safety information management system integrated on the platform, comprehensively integrate and display monitoring abnormal data such as alarms, faults, and disconnections, chamber fire warning data, and management data such as risk points, potential hazard points, and procedures and measures.

[0287] The safety situation in the pump house mainly focuses on the local violation scene in the pump house, superimpose and display the operation status of the drainage pumps, and use the form of a special topic dashboard inside the pump house to display the safety situation analysis and evaluation results in terms of personnel dimension, electromechanical dimension, environmental dimension, and management dimension of the area. The overall mine scene and the local scene of the pump house are connected in series by clicking the mouse and zooming with the scroll wheel. At the same time, according to the monitoring, analysis, and management data of the drainage monitoring system and safety information management system integrated on the platform, comprehensively integrate and display monitoring abnormal data such as alarms, faults, and disconnections, and management data such as risk points, potential hazard points, and procedures and measures.

[0288] ⑧ Visualization of the safety situation in the industrial square

[0289] The industrial square includes 4 sub-scenarios: the ventilation machine room, the surface substation, the air compressor room, and the gas drainage pump station.

[0290] The visualization of the safety situation in the ventilation machine room is achieved by three-dimensionally modeling the internal scene model of the ventilation machine room. The camera lens is used to roam from far to near to display the overall view and internal scene of the ventilation machine room. At the same time, combined with a special topic display board, it shows the monitoring of equipment parameters, the operation of equipment, and the safety situation analysis and evaluation results in terms of personnel dimension, mechanical and electrical dimension, environmental dimension, and management dimension of the ventilation fan. According to the monitoring, analysis, and management data integrated by the main ventilation fan monitoring system and the safety information management system on the platform, the monitoring abnormal data such as alarms, faults, and disconnections, and the management data such as risk points, hidden danger points, and regulations and measures are comprehensively integrated and displayed.

[0291] The display effect of the surface substation is similar to that of the underground substation. It also visually displays the comprehensive protection equipment, electrical equipment, and chamber areas inside the substation. At the same time, combined with a special topic display board, it shows the monitoring of equipment parameters, the operation of equipment, and the safety situation analysis and evaluation results in terms of personnel dimension, mechanical and electrical dimension, environmental dimension, and management dimension of the power comprehensive protection. According to the monitoring, analysis, and management data integrated by the power monitoring system and the safety information management system on the platform, the monitoring abnormal data such as alarms, faults, and disconnections, and the management data such as risk points, hidden danger points, and regulations and measures are comprehensively integrated and displayed.

[0292] The visualization of the safety situation in the air compressor room is achieved by three-dimensionally modeling the internal scene model of the air compressor room. The camera lens is used to roam from far to near to display the overall view and internal scene of the air compressor room. At the same time, combined with a special topic display board, it shows the monitoring of equipment parameters, the operation of equipment, and the safety situation analysis and evaluation results in terms of personnel dimension, mechanical and electrical dimension, environmental dimension, and management dimension of the air compressor. According to the monitoring, analysis, and management data integrated by the air compressor monitoring system and the safety information management system on the platform, the monitoring abnormal data such as alarms, faults, and disconnections, and the management data such as risk points, hidden danger points, and regulations and measures are comprehensively integrated and displayed.

[0293] The visualization of the safety situation in the gas drainage pump station is mainly built in high-gas mines. It is similar to the display effects of the ventilation machine room and the air compressor room. The camera lens is also used to roam from far to near to display the overall view and internal scene of the gas drainage pump station. At the same time, combined with a special topic display board, it shows the monitoring of equipment parameters, the operation of equipment, and the safety situation analysis and evaluation results in terms of personnel dimension, mechanical and electrical dimension, environmental dimension, and management dimension of the gas drainage pump. According to the monitoring, analysis, and management data integrated by the gas drainage monitoring system and the safety information management system on the platform, the monitoring abnormal data such as alarms, faults, and disconnections, and the management data such as risk points, hidden danger points, and regulations and measures are comprehensively integrated and displayed.

[0294] (2) Web configuration display of typical coal mine scenarios based on professional lines

[0295] 1) Web configuration principles and basic functions

[0296] The Web Cloud Configuration Platform utilizes a configurable, front-end / back-end separated microservices architecture, supporting comprehensive configuration-based development and the rapid construction of application interfaces and business logic. It can meet the functional and performance requirements of various smart mining applications. It utilizes mainstream Web technologies, boasts an advanced and user-friendly UI, and supports multi-platform, multi-terminal, and multi-browser access. It utilizes WebGL to render 2D and 3D scenes and provides a rich set of graphics components and powerful functional design for creating interactive Web-based 3D visualization applications.

[0297] Web Configuration provides core functionality for building 2D and 3D visualization applications. Data Process is responsible for processing and converting data, while Data Visualization uses ht.Node, ht.Edge, and ht.View to visualize data. UserInteraction is responsible for user interaction, while Event Handler is responsible for responding to and processing user interaction events. DataSource is the data source. The core of Web Configuration is to transform data into visualization elements and respond to user interactions.

[0298] 2) Monitoring visualization plug-in based on Web configuration

[0299] A configuration graphics framework with scenarios such as security situation, monitoring and surveillance, and disaster warning is designed, and a graphics API interface and service interface are provided to interact with the presentation layer, supporting the loading, editing, and display of vector graphics (graphic elements).

[0300] 3) Visual display of security risks based on professional lines

[0301] The professional safety situation awareness analysis is based on the "Basic Requirements and Scoring Methods for Coal Mine Safety Standardization." It analyzes and evaluates the system through four secondary components: fixed risks of mine conditions, quarterly inspection scores for mine safety standardization companies, the number of hidden dangers and violations in each professional field, and abnormalities in related monitoring systems. The focus is on analyzing the overall performance of the mine from all professional dimensions. The professional line includes seven major categories: coal mining, tunneling, electromechanical, transportation, ventilation, geological hazard prevention and measurement, and comprehensive management. The comprehensive management major is further divided into nine sub-specialties: philosophy and goals and the mine manager's safety commitment; organizational structure; safety production responsibility system and safety management system; employee quality; safety risk classification and control; accident hazard investigation and control; dispatching and emergency management; occupational disease hazard prevention and control and surface facilities; and continuous improvement.

[0302] The professional line display is based on visualization chart technologies such as Sankey diagrams and bar charts. The core content of the comprehensive analysis portal for each profession is the index hierarchy of the profession model. For example, in the coal mining profession, the core content includes the analysis results of four aspects: inherent risks, work safety standardization, investigation and treatment of potential hazards and violations of disciplines and regulations, and abnormal monitoring and control of mechanical and electrical equipment.

[0303] The system of this embodiment includes functions such as safety informatization management, risk analysis and identification, risk in the workplace, risk in the professional dimension, and risk visualization display. The functional architecture is as Figure 15 shown.

[0304] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the purpose and scope of the present technical solution, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A full-mine safety situation analysis and evaluation system, characterized in that, The system includes a basic module, a risk assessment module, and a risk visualization and analysis module; The basic module includes a basic information digitization module and a safety big data center module; the basic information digitization module is based on the coal mine safety production standardization management system, and a basic information digitization database is established by specialty; the safety big data center module integrates multi-source heterogeneous data, namely management, monitoring, and analysis data, and establishes a unified data interaction standard; The risk assessment module includes a specialty dimension scoring module and a spatial dimension scoring module; the specialty dimension scoring module uses the safety checklist method for risk assessment, that is, for each specialty field, a safety checklist is constructed respectively from the results of supervision and inspection, the results of safety production standardization work, hidden dangers, three-dimensional conditions, and monitoring anomaly perception conditions, and the scores of each specialty field are calculated dynamically; the spatial dimension scoring module is aimed at 5 typical scenarios including coal mining faces, tunneling faces, transportation gateways, electrical and mechanical chambers, and surface industrial squares, and uses the virtual object situation scoring method to realize automatic, real-time, and quantitative evaluation based on quasi-static and dynamic monitoring data; The risk visualization and analysis module establishes a GIS single map and a configuration map of local key places to realize the visual display of the risk analysis and evaluation results.

2. The full mine safety situation analysis and evaluation system according to claim 1, characterized in that The basic information digitization module is a management report based on interface interactive configuration that integrates data source configuration, data object configuration, and data report configuration.

3. The full mine safety situation analysis and evaluation system according to claim 1, characterized in that, The established unified data interaction standard includes a safety monitoring data interaction standard, a warning analysis data interaction standard, and an electrical and mechanical equipment data interaction standard; The safety monitoring data interaction standard converts various types of data into defined data, real-time data, abnormal data, historical data, and statistical data through a protocol adapter; The warning analysis data interaction standard includes a unified area code, a unified warning event, a unified warning level, and a unified interaction format; The electrical and mechanical equipment data interaction standard uses the EIP object model to realize equipment abstraction and dynamic attribute extension; the equipment abstraction specifically abstracts mine equipment into classes and assigns a unique code to support unified management; the dynamic attribute extension supports the classification management of static attributes and running attributes and adapts to different monitoring scenarios.

4. The full-mine safety situation analysis and evaluation system according to claim 1, characterized in that The virtual object situation scoring method specifically includes the following steps: S1: Establish an evaluation hierarchy model using the analytic hierarchy process integrated with fine network division; S2: Use the technique for order preference by similarity to an ideal solution (TOPSIS) integrated method to establish a neutral reference evaluation rule and mechanism for each evaluation index of the virtual object, including: (1) Determine the critical point between safety and insecurity, that is, the index value of the neutral virtual object; (2) According to the risk level division, determine the critical values of different levels, that is, the critical values of each index of virtual objects of different risk levels, the optimal values, that is, the index values of the optimal virtual object, and the worst values, that is, the index values of the worst virtual object; (3) Determine the index scoring mechanism between different virtual objects: Determine the logical relationship of the same-level indicators relative to the upper-level indicators, including: determining whether there are key indicators; determining whether the indicators at each level are relatively important indicators under different circumstances, and calculate using the following method: When all indicators at a certain level are greater than the safety critical score, the total score is calculated using linear weighting, otherwise the lowest score of all indicators is taken as the final score of this level; (4) Determine the importance of the indicators; S3: Use a combined weighting strategy to determine the initial weights of each evaluation indicator in the virtual object evaluation hierarchical model; S4: Use the method of dynamic modeling to construct the hierarchical model of the specific evaluation object; S5: According to the characteristics of each hierarchical model of the virtual reference object, use the combination method of linear weighting, key indicator veto, and dynamic weight to evaluate the safety situation of the specific evaluation object.

5. The full-mine safety situation analysis and evaluation system according to claim 4, characterized in that In step S1, use the analytic hierarchy process integrating fine network division to establish an evaluation hierarchical model, which specifically includes the following steps: S11: Grid subdivision; S12: Identify the risk elements of people, machines, environment, and management within the grid; On the basis of grid division, according to the different equipment, environmental monitoring, and management focuses of each area, sort out the risk elements existing in each area from the four dimensions of people, machines, environment, and management; For the personnel dimension, it is constructed from aspects such as work experience, education level, shift leading, employee training, physical condition, operators, certification status, and fatigue operation index; among them, the fatigue operation index reflects the comprehensive working state of the operators and is related to the working hours of history, current, and recent periods, night shift working hours, and the interval between two shifts; For the environmental dimension: Identify the elements from 5 major disasters including gas disasters, water disasters, fires, roof falls, and dust; For the electromechanical dimension: Include 50 mandatory inspection items for electromechanics in the "Coal Mine Safety Regulations" and identify the electromechanical equipment health index; The analysis of the health of electromechanical equipment is carried out from the aspects of monitoring and alarm conditions, changes in equipment parameters, and equipment operation and maintenance conditions, identifying sudden changes in monitoring data, upward or downward trends, and frequent out-of-bounds of the fluctuation range, and timely discovering abnormal equipment failure information; For the management dimension: Identify risk elements from the aspects of hidden danger investigation and treatment, three violations, and risk control implementation; S13: Establish the overall goal hierarchical evaluation structure; Analyze the relationships between the risk elements in each subdivision area, delete duplicate indicators, and on the basis of the risk element analysis in each area, analyze the risks between each subdivision area and the overall regional system, so as to construct the risk evaluation index system for the entire area. Specifically, in view of the overall risk of the mine, the mine risk is evaluated from seven major disciplines: coal mining, tunneling, electromechanical, transportation, ventilation, geological disaster measurement and prevention, and comprehensive management. Among them, six disciplines, namely coal mining, tunneling, electromechanical, transportation, ventilation, geological disaster measurement and prevention, are related to the information system of mine construction. The risk of each discipline is evaluated from four aspects: fixed risk, quarterly inspection score of safety production standardization, three-dimensional number of professional hidden dangers, and abnormalities of related monitoring systems. The comprehensive management discipline is subdivided into nine sub-disciplines: concept goals and safety commitment of mine managers, organizational structure, safety production responsibility system and safety management system, quality of employees, safety risk classification management and control, accident hazard investigation and treatment, dispatching and emergency management, occupational disease hazard prevention and control and ground facilities, and continuous improvement. Since these disciplines do not involve online monitoring systems, the risks are analyzed and evaluated based on the discipline's own fixed scores and the dynamic scores of the company's quarterly inspections. In response to the safety risks in each key area, evaluation indicators are constructed from four aspects: personnel, mechanical and electrical, environment, and management.

6. The full mine safety situation analysis and evaluation system according to claim 4, characterized in that, In step S2, the critical point between safety and unsafe is determined, which specifically includes: (1.1) Critical values based on laws, regulations, industry standards, and expert experience; (1.2) Calculate the critical value of the field data based on the approximation to the ideal point, specifically including: ① According to the idea of approximating the ideal point, first, based on the evaluation index system, collect and calculate the parameters of each indicator of all areas to be evaluated, and calculate the distribution of the values of each evaluation indicator of different types; then determine the optimal and worst values of all possible values of this indicator; for benefit-type indicators, the larger the value, the better; for cost-type indicators, the smaller the value, the better; obtain the optimal and worst values of each indicator in all samples; the optimal value is the positive ideal solution; the worst value is the negative ideal solution; ②Calculate the sample proximity; ③ Calculate the critical value of field data based on cluster analysis; For indicators that are neither benefit-based nor cost-based and for which there are no relevant normative references, clustering is used for division and determination; First, indicator samples are collected and manually divided into safe samples, relatively safe samples, relatively dangerous samples or dangerous samples according to actual conditions; then the K-means clustering method is used to determine the center point of each sample as the critical value of virtual objects of different levels.

7. The full mine safety situation analysis and evaluation system according to claim 4, characterized in that In step S2, the critical values of various indicators of virtual objects with different risk levels are determined, specifically including: for the critical value a based on laws, regulations, industry norms, and expert experience, first determine the basic critical value based on a; then subdivide it based on the critical value b of the field data based on the approximate ideal point and the critical value c of the field data based on cluster analysis.

8. The full mine safety situation analysis and evaluation system according to claim 4, characterized in that In step S2, the indicator scoring mechanism between different virtual objects is determined, specifically including: (3.1) For the continuously changing quantitative index c, the fractional division of different virtual object subintervals is improved according to the linear interpolation method; the specific operation steps are as follows: 1) Select different virtual object index values and give their interval scales in the order of absolute safety, safety, relatively safe, relatively dangerous, dangerous, and absolute danger, i.e., c1, c2, …, c n , and at the same time give the corresponding scores s1, s2, …, s n ; 2) If the actual c value of a certain evaluated object exactly falls within the interval of [c i , c j , then its corresponding score s is calculated according to the linear interpolation method, that is If the actual value of c falls outside the pre-calibrated interval, linear extrapolation is used; when c < c1, assuming c1 < c n at this time, there is Conversely, when c > c n , assuming c1 < c n , there is When using linear extrapolation, if the score s is less than 0 or greater than 100, it is directly converted to 0 or 100. When constructing the above discrete (c i , s i ) data pairs, the following points need to be noted: ① Ensure (c i , s i ) the s in i The larger, the safer; when s i = 0, it is equivalent to a very unsafe situation; when s i = 100, it is equivalent to a very safe situation; ② When it is difficult to determine when it is extremely unsafe and extremely safe, no stipulation is made for the c i value when s i = 0 and s i = 100. Only stipulations are made for the c i value when s i = 10 and s i = 90.

9. The full-mine safety situation analysis and evaluation system according to claim 4, wherein, In step S3, the combined weighting strategy determines the weights of indicators by combining subjective and objective weights, specifically including: (1) Subjective weight: determined by the analytic hierarchy process; (2) Objective weight: taking the historical work safety standardization inspection results as samples, using the logistic regression method to dynamically calculate the objective weight; (3) Comprehensive weight: based on the game theory idea, minimizing the deviation sum of the subjective weight and the objective weight to calculate the optimal comprehensive weight.

10. The full mine safety situation analysis and evaluation system according to claim 4, characterized in that, In step S4, a hierarchical model of a specific evaluation object is constructed by using the dynamic modeling method, specifically including the following two methods: (1) Using the dynamic modeling method: When using the pairwise comparison method to determine the relative importance of each criterion element relative to the parent element, if it is already known that a certain element cannot obtain a definite score for a certain evaluation object, then for this object, when pairwise comparing each criterion element, without considering the missing criterion element, recalculate the weights of each element; For the evaluation object that can obtain all criterion scores, still determine the weights of each criterion according to the complete model; (2) Using the default value method: When data is missing, use the default value to replace it; when modeling, set default values for all indicators; it is recommended to set default values based on the following two situations: ① If the relative importance of the indicator to the parent element is not high and the risk of its exceeding the limit is not great, then set its default value to the neutral value so that its score is 50 points; ② If the relative importance of the indicator to the parent element is high and the risk of its exceeding the limit is great, then set its default value to a value that causes unsafe consequences to ensure that its score is less than 50 points.

Images