Mine safety production risk monitoring and early warning system and method

Abstract

The invention discloses a mine safety production risk monitoring and early warning system and method, and relates to the technical field of mine safety, and the system comprises a data collection module, a data integration and storage module, a correlation analysis and model construction module, a real-time monitoring and early warning module and an emergency decision support module. According to the method, a multi-dimensional data acquisition system is constructed, multi-source data of geology, production, equipment, personnel and the like are fused, deep analysis is carried out by applying a multi-dimensional association algorithm based on mutual information, potential relationships among the data are comprehensively mined, comprehensive and accurate assessment of mine safety production risks is realized, and a real-time dynamic risk assessment model is utilized to realize comprehensive and accurate assessment of mine safety production risks. The risk state is updated in real time according to latest collected multi-dimensional data, a server-side timed task obtains data from a data integration and storage module according to the minute-level frequency and inputs the data into a model, an early warning mechanism is triggered immediately when the risk level calculated by the model exceeds a preset threshold value, risk mutation can be perceived in time, and the hysteresis of a traditional assessment mode is overcome.

Description

Technical Field

[0001] The present invention relates to the field of mine safety technology, and in particular to a mine production safety risk monitoring and early warning system and method. Background Art

[0002] As a specific area for the exploitation of mineral resources, mines are the key foundation for supporting the energy industry and industrial system. Their mining operations involve a variety of complex engineering activities, covering underground excavation, ore transportation, ventilation and drainage, and other links. These activities not only face complex and changeable geological conditions, but also rely on a large number of mechanical equipment and a large number of personnel to work together. In the mining process, production safety is always the top priority. Mine production safety risk monitoring is the core link to ensure the safe operation of mines. Through the use of advanced sensor technology, data acquisition equipment and information technology, various safety risk factors in the mining environment are monitored and measured in an all-round and real-time manner. This measure is of immeasurable significance for timely detecting potential safety hazards, preventing accidents, protecting the lives of mine workers, and maintaining the normal production order of mining enterprises.

[0003] However, the existing technology still has certain defects when used. Traditional mine safety production risk assessment mostly adopts periodic assessment or static assessment methods. These methods often cannot reflect the dynamic changes in the mine production process in a timely manner and cannot accurately capture the sudden changes in risks. At the same time, previous assessment methods usually only focus on data in a single dimension or a few dimensions, and lack the comprehensive utilization of multi-dimensional data such as mine geological data, production data, equipment operation data and personnel information. It is difficult to comprehensively and accurately assess safety production risks, resulting in a lack of strong data support in emergency decision-making and an inability to respond to sudden safety incidents in a timely and effective manner. Therefore, it is of great significance to develop mine safety production risk monitoring and early warning systems and methods. Summary of the Invention

[0004] The purpose of this invention is to make up for the shortcomings of the existing technology and provide a mine safety production risk monitoring and early warning system and method. It can achieve comprehensive, real-time and accurate monitoring and early warning of mine safety production risks by establishing a real-time dynamic risk assessment system that integrates multi-dimensional data, providing strong support for emergency decision-making.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: a mine production safety risk monitoring and early warning system, which includes: a data acquisition module, a data integration and storage module, a correlation analysis and model building module, a real-time monitoring and early warning module, and an emergency decision support module;

[0006] The data acquisition module is used to collect mine geological data, production data, equipment operation data and personnel information;

[0007] The data integration and storage module is used to integrate and store the collected data in a unified format;

[0008] The association analysis and model construction module uses a multi-dimensional association algorithm of mutual information, and the calculation formula is: where X and Y are data of different dimensions, p(x,y) is the joint probability and marginal probability, α and β i are adjustment parameters, X i is the sub-dimensional data of X, H(X i |Y) is the conditional entropy. Based on this, a real-time dynamic risk assessment model is constructed. This model calculates the risk level R through the formula where Z j is the feature data after association analysis, γ j is the weight, and f is the risk calculation function;

[0009] The real-time monitoring and warning module is used to update the risk status and give warnings according to the latest data;

[0010] The emergency decision-making support module provides decision-making reference plans based on the risk situation.

[0011] Furthermore, in the data acquisition module, geological data is collected by arranging pressure sensors, displacement sensors, and gas sensors at roadway and stope positions to collect rock mass stress, surface displacement, and gas concentration data. The sensors collect data according to the initial set frequency. When the detected change rate of rock mass stress exceeds the set threshold δ, the collection frequency is switched from T1 to T2, where T2 < T1. Production data is obtained regularly through the data synchronization interface with the existing mine production management system to obtain information on mining progress, output, and transportation volume. Equipment operation data relies on intelligent monitoring devices installed on coal mining machines and ventilators, and the equipment operation status, rotation speed, vibration, and oil temperature parameters are collected in real time through wired and wireless connections to the equipment controller. Personnel information uses the personnel positioning system base station to track personnel positions in real time, and the attendance system synchronizes working hours and attendance record data regularly.

[0012] Even further, the data integration and storage module adopts a blockchain distributed storage structure, uses the hash algorithm to encrypt data blocks, and each data block contains the hash value of the previous data block. When integrating data, first clean the collected original data to remove noise data and outliers, then convert different types of data according to a unified data format, convert the sensor measurement values in geological data to standardized physical quantity units, and finally store them in a distributed database. Different types of data are associated according to timestamps and unique identifiers.

[0013] Furthermore, the mutual information algorithm in the association analysis and model building module introduces dynamic time warping technology when calculating mutual information, aligns different time series data, and adjusts parameters α, β i Through multiple experiments, using different mine historical data as samples, the genetic algorithm was used to iteratively optimize and determine the weight γ in the risk assessment model. j The importance of each associated feature data in risk assessment is determined through the hierarchical analysis method combined with experts' scores.

[0014] Furthermore, in the real-time monitoring and early warning module, the early warning threshold is dynamically adjusted according to the historical data of the risk assessment model. The sliding window algorithm is used to adjust the threshold according to the fluctuation of the risk data in the past N cycles. The server-side scheduled task obtains the latest multi-dimensional data from the data integration and storage module every 1 minute and inputs it into the risk assessment model. When the risk level R calculated by the model exceeds the preset warning thresholds of different levels, the system triggers the early warning mechanism. The early warning information includes the location of the risk, the risk type and the risk level.

[0015] Furthermore, the emergency decision support module establishes an emergency decision knowledge base to store previous mine safety accident cases and corresponding emergency response plans. When an early warning occurs, similar cases are retrieved from the knowledge base based on the current risk situation. A method combining semantic matching and vector space model is used to match the current risk description with the case library. The key features of the current risk situation, including the risk type and the involved area, are converted into text descriptions. Natural language processing technology is used to extract keywords, construct a vector space model, and perform similarity calculations with case vectors in the case library. Similar cases with similarities exceeding the set value are retrieved to provide a reference for emergency decision-making.

[0016] Furthermore, the system has a self-diagnosis function, which regularly checks the operating status of each module. When it detects that the module operation abnormality index exceeds the preset range, it automatically starts the backup module, monitors the indicators of sensor data transmission stability of the data acquisition module, data reading and writing speed of the data integration and storage module, and algorithm operation efficiency of the correlation analysis and model construction module, sets the normal range, and when it exceeds the range, the system switches to the backup module.

[0017] The mine production safety risk monitoring and early warning method is applicable to the above-mentioned mine production safety risk monitoring and early warning system, and the method includes the following steps:

[0018] Data collection: Collect multi-dimensional data on mine geology, production, equipment operation and personnel information;

[0019] Data integration and storage: Integrate the collected multi-dimensional data on mine geology, production, equipment operation and personnel information, and store them in a unified data format;

[0020] Correlation analysis and model building: Use multi-dimensional correlation analysis algorithms to conduct correlation analysis on data, build a real-time dynamic risk assessment model, and calculate risk levels;

[0021] Real-time monitoring and early warning: Real-time data acquisition updates risk status, triggering an early warning when the early warning threshold is reached;

[0022] Emergency decision support: Establish an emergency decision knowledge base to store previous mine safety accident cases and corresponding emergency response plans. When an early warning occurs, similar cases are retrieved from the knowledge base based on the current risk situation, and combined with real-time multi-dimensional data to provide reference plans for emergency decision-making.

[0023] Furthermore, in the data collection step, for the equipment operation data, when the change trend of the equipment operation parameters exceeds the set curve range, additional auxiliary sensors are started to collect supplementary data. In the real-time monitoring and early warning step, the server regularly obtains the latest data from the storage module and inputs it into the risk assessment model. The model output risk level is compared with the preset threshold. When the threshold is exceeded, the system sends an early warning message containing risk location, type and level information.

[0024] Compared with the existing technology, the mine production safety risk monitoring and early warning system and method have the following beneficial effects:

[0025] The present invention constructs a multi-dimensional data acquisition system, integrates multi-source data such as geology, production, equipment and personnel, uses a multi-dimensional correlation algorithm based on mutual information to conduct in-depth analysis, and comprehensively explores the potential relationships between data to achieve a comprehensive and accurate assessment of mine safety production risks. Utilizing a real-time dynamic risk assessment model, the risk status is updated in real time based on the latest collected multi-dimensional data. The server-side scheduled task obtains data from the data integration and storage module at a frequency of minutes and inputs it into the model. When the risk level calculated by the model exceeds the preset threshold, an early warning mechanism is immediately triggered, which can detect risk mutations in a timely manner and overcome the lag of traditional assessment methods.

[0026] Other advantages, objects and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be learned from the practice of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0028] Figure 1 This is a structural diagram of the mine production safety risk monitoring and early warning system;

[0029] Figure 2 This is a flow chart of the mine safety production risk monitoring and early warning method. DETAILED DESCRIPTION

[0030] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0031] Example 1

[0032] See also Figure 1 and Figure 2 In a large underground metal mine, there are multiple mining areas with complex geological conditions, covering a variety of different rock types and geological structures. At the same time, the mining operation is highly intensive, with a large number of various types of machinery and equipment, a large number of operators, and a wide and dispersed distribution of working areas. In order to ensure the working safety of the mine, the mine applies the mine safety production risk monitoring and early warning system and method of the present invention.

[0033] Pressure sensors, displacement sensors, and gas sensors are rationally and densely arranged at key locations in each mining area, such as the intersection of tunnels, the periphery of the stope, and areas prone to geological changes. In the initial stage, these sensors collect data at a preset frequency T1, continuously monitoring key geological parameters such as rock stress, surface displacement, and gas concentration. When the pressure sensor detects that the rock stress change rate exceeds the preset threshold δ, the system automatically triggers the adjustment mechanism and switches the sensor's collection frequency from T1 to a higher frequency T2. This allows for more intensive and timely acquisition of geological data during critical periods when abnormal changes in geological conditions occur, providing richer and more accurate information for subsequent risk assessments.

[0034] Through a specially developed data synchronization interface, a stable and efficient data connection is established with the mine's existing production management system. Utilizing this interface, comprehensive production data is regularly obtained from the production management system at fixed time intervals, including but not limited to the mining progress of each mining area, with detailed records of daily and weekly changes in mining depth and range; production data, which accurately counts the output of ore in different time periods; and transportation volume data, which covers information such as the ore transportation volume in each transportation link from the mining site to the beneficiation plant. These production data can intuitively reflect the production and operation status of the mine and provide an important production dimension reference for risk monitoring.

[0035] Advanced intelligent monitoring devices are installed on various core mechanical equipment, such as coal mining machines, ventilators, hoists, etc. These monitoring devices are stably connected to the equipment controllers through wired or wireless means, and can collect multiple key parameters during the operation of the equipment in real time. For coal mining machines, their operating status (such as start-up, shutdown, failure, etc.), speed, tool vibration, oil temperature and other parameters are collected to evaluate the working efficiency and health of the coal mining machine; for ventilators, their wind pressure, air volume, motor speed and equipment vibration amplitude and other parameters are monitored to ensure the stable operation of the ventilation system and provide a good ventilation environment for underground workers; for hoists, attention is paid to their lifting speed, load, braking system status and other parameters to ensure the safety of ore and personnel transportation.

[0036] With the help of a high-precision personnel positioning system base station, the location information of each operator can be tracked in real time. Regardless of whether they are walking in the underground tunnels, working in the mine, or active in other work areas, the system can accurately capture their location coordinates and transmit the location information to the data collection center in real time. At the same time, the attendance system regularly synchronizes the operator's working hours, attendance records and other information according to the established cycle, including working and leaving time, overtime hours, leave records, etc., in order to fully understand the personnel's work status and attendance.

[0037] All types of collected data are quickly transmitted to the data integration and storage module through a high-speed data transmission network. In this module, the data cleaning program is first started, and a specially designed data cleaning algorithm is used to comprehensively screen the original data to identify and remove noise data and outliers. For example, data collected by the sensor that obviously deviates from the normal range and does not conform to physical laws, or erroneous data caused by signal interference, are all marked and eliminated.

[0038] Subsequently, different types of data are converted and processed according to the unified data format standards. For the sensor measurement values in the geological data, they are converted into standardized physical units according to the internationally accepted physical quantity unit standards to ensure the consistency and comparability of the data. After the cleaning and conversion, the data are associated and integrated according to the timestamp and the unique identifier set in advance for each type of data.

[0039] Finally, a blockchain distributed storage structure is adopted, and advanced hash algorithms are used to encrypt data blocks. Each data block not only contains the currently collected data content, but also embeds the hash value of the previous data block, thereby building a complete, tamper-proof and highly traceable data storage system, which is stored in a distributed database and provides a reliable data foundation for subsequent data processing and analysis.

[0040] The multi-dimensional correlation algorithm based on improved mutual information is used to deeply mine and analyze the integrated multi-dimensional data. The formula is: When calculating mutual information, the dynamic time warping technology is cleverly introduced. This technology can accurately align data of different time series through a specific algorithm, effectively solving the analysis error problem caused by asynchronous data collection time. In the algorithm, the parameters α and β are adjusted. i It is not a random setting, but is verified through a large number of experiments. Using historical data from this mine and other similar mines as samples, a complex experimental model is constructed with the core goal of maximizing the accuracy of multi-dimensional data correlation analysis. Genetic algorithms are used for multiple rounds of iterative optimization to ultimately determine the parameter values that best suit the actual situation of the mine.

[0041] Based on the rich data relationships obtained from association analysis, a real-time dynamic risk assessment model is constructed. In this model, the formula To calculate the risk level R, where the weight γ j The determination adopts the hierarchical analysis method, organizes senior experts in the mining field, and carefully scores the importance of each related characteristic data in the risk assessment process. After rigorous mathematical calculation and analysis, a scientific and reasonable weight value is obtained to ensure that the risk assessment model can accurately reflect the actual risk status of mine safety production.

[0042] On the server side, a special scheduled task program is set up, which automatically starts every 1 minute. After startup, the scheduled task quickly obtains the latest multi-dimensional data from the data integration and storage module, and accurately inputs this data into the risk assessment model. The risk assessment model performs high-speed calculations based on the input data and the established algorithms and formulas to calculate the current mine safety production risk level R in real time.

[0043] At the same time, multiple different levels of warning thresholds are pre-set in the model. These thresholds are determined based on a combination of factors such as the mine's historical risk data, safety standards, and expert experience. When the risk level R calculated by the model exceeds a preset warning threshold, the system will automatically trigger the warning mechanism. The warning information will be quickly sent out through multiple channels, including but not limited to text messages to the mobile phones of mine safety managers, eye-catching sound and light alarm prompts on the large screen of the mine monitoring center, and detailed warning prompt windows popping up on the software interface of the relevant operation terminal.

[0044] The early warning information includes in detail the specific location where the risk occurs, which can be accurately determined through the personnel positioning system and equipment location information; the risk type, which can be clearly determined based on the results of correlation analysis and model judgment to determine whether it is a geological disaster risk, equipment failure risk or personnel operation risk, etc.; and the risk level, so that the recipient can intuitively understand the severity of the risk.

[0045] When the warning information is triggered, the emergency decision support module immediately activates the response mechanism. First, the key features of the current risk situation, such as risk type, involved equipment or area, risk level, etc., are converted into a detailed text description. Then, using advanced natural language processing technology, the text description is deeply analyzed to extract key information and keywords. Based on these keywords, a vector space model is constructed to calculate the similarity between the vector of the current risk situation and the vector of past mine safety accident cases stored in the case library.

[0046] The case library stores a large number of organized and classified historical cases. Each case records in detail the background, cause, handling process and results of the accident. Through similarity calculation, similar cases with a similarity to the current risk situation exceeding the set value are retrieved. Combined with the multi-dimensional data obtained in real time, the retrieved similar cases are analyzed and adjusted in a targeted manner to generate emergency decision-making reference plans suitable for the current actual situation.

[0047] To sum up, this embodiment comprehensively and deeply realizes all-round, real-time and precise monitoring and early warning of mine safety production risks. This embodiment effectively guarantees mine safety production, greatly reduces the probability of accidents and the extent of losses caused by accidents, and lays a solid technical foundation for the continuous, stable and safe production operations of mines, ensuring that mine production activities can proceed smoothly in a safe and efficient environment.

[0048] Example 2

[0049] See also Figure 1 and Figure 2 In an open-pit coal mining scenario, the coal mine covers a vast area, and the mining area has undulating terrain and is greatly affected by natural environmental factors. Extreme weather such as strong winds and heavy rains frequently occur. The coal mine is equipped with large-scale open-pit mining equipment, such as giant excavators and heavy trucks. At the same time, the workers are distributed in different mining, transportation and auxiliary links. In order to ensure the working safety of the coal mine, the mine applies the mine safety production risk monitoring and early warning system and method of the present invention.

[0050] Geological monitoring equipment, including geological radars, inclinometers, and stress gauges, are installed at key locations in open-pit coal mines, such as mining boundaries, slopes, and pit bottoms. The geological radar periodically scans the geological structure within a certain depth underground to obtain information such as stratum distribution and fault location. The inclinometer monitors changes in the slope's inclination angle in real time, while the stress gauge continuously measures the stress state inside the rock mass. When the inclinometer detects that the slope's inclination angle change rate exceeds a preset threshold δ, the data acquisition frequency is increased from the initial T1 to a higher T2 to more intensively capture abnormal changes in geological conditions.

[0051] Through seamless connection with the coal mine production scheduling system, production data is obtained in real time. With the help of sensors installed on excavators, loaders and other equipment, as well as vehicle identification systems on transportation roads, the working hours, excavation volume, loading volume of each equipment, as well as the transportation vehicle's route, load and number of transportation trips are accurately recorded. At the same time, energy consumption data such as electricity and fuel are collected from the coal mine's energy management system for comprehensive evaluation of production efficiency and costs.

[0052] For large-scale open-pit mining equipment, such as giant excavators, heavy-duty trucks and belt conveyors, various sensors are installed at key parts of the equipment. Pressure, displacement and angle sensors are installed on the excavator's bucket, lifting arm and slewing mechanism to monitor the equipment's working status and the operating parameters of its mechanical components in real time. Temperature, pressure and vibration sensors are deployed on the engine, transmission, tires and other parts of the heavy-duty truck to keep abreast of the operating conditions of the vehicle's power system, transmission system and travel system. For belt conveyors, sensors installed on the belts, rollers and drive devices monitor the belt tension, operating speed and rotation of the rollers. These sensors send the collected data to the data acquisition terminal in real time through a wireless transmission module.

[0053] Each worker is equipped with a smart bracelet with positioning and communication functions. Satellite positioning technology and wireless networks are used to track the worker's location information in real time. The smart bracelet can also record the worker's physiological parameters such as heart rate and body temperature, as well as work status information such as working hours and rest time. At the same time, through the coal mine's attendance management system, the worker's attendance records, leave status and training information are regularly synchronized.

[0054] The data obtained from various data acquisition sources are aggregated to the data integration and storage module through an encrypted data transmission link. First, the data cleaning algorithm is used to denoise and process outliers on the original data. For example, obvious erroneous data caused by sensor failure or signal interference, as well as data that deviates too much from the normal operating parameter range of the equipment, are eliminated. Then, according to the unified data format specification, different types of data are standardized and converted. For example, the reflected wave data collected by the geological radar is converted into intuitive geological image data, and the equipment operating parameters are converted into data expressed in international standard units.

[0055] After cleaning and conversion, the data is associated and integrated based on timestamps and unique identifiers, and stored using a distributed file system combined with blockchain technology. The data blocks are encrypted using a hash algorithm. Each data block contains the hash value of the previous data block to ensure the security, integrity and traceability of the data. Different types of data, such as geological data, production data, equipment operation data and personnel information, are stored in different data tables, but are associated through timestamps and unique identifiers to facilitate subsequent data query and analysis.

[0056] The integrated data are deeply correlated using a multi-dimensional correlation algorithm based on improved mutual information. The formula is: The algorithm introduces dynamic time warping technology when calculating mutual information, accurately aligns data of different time series, and uses genetic algorithms to adjust parameters α1, β i1 Optimization is performed to maximize the effectiveness of correlation analysis between multi-dimensional data. For example, the correlation between geological conditions and equipment failures can be analyzed to find out the patterns of equipment failure in specific geological areas; the relationship between researchers' working status and production efficiency and safety risks can be studied.

[0057] Based on the results of correlation analysis, a real-time dynamic risk assessment model is constructed, and the formula Calculate the risk level R, where the weight γ j Through the hierarchical analysis method, experts in the coal mining field are organized to score and determine the importance of each associated characteristic data in risk assessment. The risk calculation function f is designed according to the characteristics of different types of risk factors to accurately reflect the risk status.

[0058] A scheduled task is set up on the data processing server to obtain the latest multi-dimensional data from the data integration and storage module every minute and input it into the risk assessment model. Based on the input data, the model quickly calculates the current open-pit coal mine's production safety risk level R. In the risk assessment model, multiple different levels of warning thresholds are pre-set. These thresholds comprehensively consider factors such as the coal mine's historical accident data, safety regulations and expert experience.

[0059] When the risk level R calculated by the model exceeds a preset warning threshold, the system automatically triggers the warning mechanism, and the warning information is promptly sent to relevant personnel through various means, including sending text message notifications to the mobile phones of safety managers, displaying a striking warning pop-up window on the large screen of the coal mine dispatching center, accompanied by sound and light alarm prompts, and sending a warning voice message to the broadcasting system of the work site. The warning information contains detailed information about the specific location of the risk, which is accurately determined through the personnel positioning system and equipment positioning information; the risk type, such as geological disaster risk (slope landslide, collapse, etc.), equipment failure risk (excavator failure, truck breakdown, etc.) or personnel safety risk (fatigue work, illegal operation, etc.); and the risk level, so that the recipient can quickly understand the severity of the risk and take corresponding measures.

[0060] Once the early warning is triggered, the emergency decision support module is immediately activated. First, the key features of the current risk situation, such as risk type, involved area, risk level and other information, are converted into text descriptions. Then, natural language processing technology is used to analyze the text, extract keywords, construct a vector space model, and calculate the similarity with the historical case vectors in the case library. The case library stores a large number of open-pit coal mine production safety accident cases. Each case records in detail the background, cause, handling process and results of the accident. Through similarity matching, similar cases with a similarity to the current risk situation exceeding the set value are retrieved. Combined with the multi-dimensional data collected in real time, the retrieved similar cases are targeted adjusted and optimized to generate an emergency decision reference plan suitable for the current actual situation.

[0061] To sum up, in this open-pit coal mine scenario, the system constructed in this embodiment comprehensively and efficiently realizes accurate monitoring and early warning of production safety risks. The multi-dimensional data collection system covers various key aspects such as geology, production, equipment and personnel, deeply mines the potential risk correlations between data, and realizes a comprehensive and accurate assessment of production safety risks. The real-time dynamic risk assessment model can closely track data changes in the coal mine production process and update the risk status in a timely manner. Compared with traditional monitoring methods, it can detect risk mutations in advance and gain valuable time for emergency response.

[0062] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as above in terms of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art can, without departing from the scope of the technical solution of the present invention, make some changes or modifications to equivalent embodiments using the technical contents disclosed above. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. Mine safety production risk monitoring and early warning system, characterized by: The system includes: a data acquisition module, a data integration and storage module, an association analysis and model construction module, a real-time monitoring and early warning module, and an emergency decision-making support module; The data acquisition module is used to collect mine geological data, production data, equipment operation data, and personnel information; The data integration and storage module is used to integrate and store the collected data in a unified format; The association analysis and model building module uses a multi-dimensional association algorithm of mutual information, and the calculation formula is: Where X and Y are data of different dimensions, p(x,y) is the joint probability and marginal probability, α and β i is the adjustment parameter, X i is the sub-dimension data of X, H(X i |Y) is the conditional entropy, based on which a real-time dynamic risk assessment model is constructed. The model is expressed by the formula Calculate the risk level R, where Z j is the characteristic data after association analysis, γ j is the weight, f is the risk calculation function; The real-time monitoring and early warning module is used to update the risk status and give early warnings according to the latest data; The emergency decision-making support module provides a decision-making reference plan based on the risk situation.

2. The mine production safety risk monitoring and early warning system according to claim 1 is characterized in that: In the data acquisition module, geological data is collected by arranging pressure sensors, displacement sensors, and gas sensors in roadway and stope positions to collect rock mass stress, surface displacement, and gas concentration data. The sensors collect data according to the initial set frequency. When the detected change rate of rock mass stress exceeds the set threshold δ, the collection frequency is switched from T1 to T2, where T2 < T1. Production data is obtained regularly through the data synchronization interface with the existing mine production management system to obtain information on mining progress, output, and transportation volume. Equipment operation data relies on intelligent monitoring devices installed on coal mining machines and ventilators to collect equipment operation status, rotational speed, vibration, and oil temperature parameters in real time through wired and wireless connections to the equipment controller. Personnel information uses the personnel positioning system base station to track personnel positions in real time, and the attendance system synchronizes working hours and attendance record data regularly.

3. The mine production safety risk monitoring and early warning system according to claim 1 is characterized in that: The data integration and storage module adopts a blockchain distributed storage structure, uses the hash algorithm to encrypt data blocks, and each data block contains the hash value of the previous data block. When integrating data, first clean the collected raw data to remove noise data and outliers, then convert different types of data according to a unified data format, convert the sensor measurement values in geological data into standardized physical quantity units, and finally store them in a distributed database. Different types of data are associated according to timestamps and unique identifiers.

4. The mine production safety risk monitoring and early warning system according to claim 1 is characterized in that: The mutual information algorithm in the association analysis and model building module introduces dynamic time warping technology when calculating mutual information, aligns different time series data, and adjusts parameters α, β i Through multiple experiments, using different mine historical data as samples, the genetic algorithm was used to iteratively optimize and determine the weight γ in the risk assessment model. j The importance of each associated feature data in risk assessment is determined through the hierarchical analysis method combined with experts' scores.

5. The mine production safety risk monitoring and early warning system according to claim 1 is characterized in that: In the real-time monitoring and early warning module, the early warning threshold is dynamically adjusted according to the historical data of the risk assessment model. The sliding window algorithm is adopted to adjust the threshold according to the risk data fluctuation in the recent N cycles. The server-side timing task fetches the latest multi-dimensional data from the data integration and storage module every 1 minute and inputs it into the risk assessment model. When the risk level R calculated by the model exceeds the preset early warning thresholds of different levels, the system triggers the early warning mechanism. The early warning information includes the risk occurrence location, risk type, and risk level.

6. The mine production safety risk monitoring and early warning system according to claim 1 is characterized in that: The emergency decision-making support module establishes an emergency decision-making knowledge base to store past mine safety accident cases and corresponding emergency disposal plans. When an early warning occurs, according to the current risk situation, retrieve similar cases from the knowledge base, and use a method combining semantic matching and vector space model to match the current risk description with the case library. Convert the key features of the current risk situation, including risk type and affected area, into text descriptions, extract keywords using natural language processing technology, construct a vector space model, calculate the similarity with the case vectors in the case library, and retrieve similar cases with a similarity exceeding the set value to provide a reference for emergency decision-making.

7. The mine production safety risk monitoring and early warning system according to claim 1 is characterized in that: The system has a self-diagnosis function and regularly checks the operating status of each module. When it detects that the module operation abnormality index exceeds the preset range, it automatically starts the backup module, monitors the indicators of sensor data transmission stability of the data acquisition module, data reading and writing speed of the data integration and storage module, and algorithm operation efficiency of the correlation analysis and model building module, and sets a normal range. When the range is exceeded, the system switches to the backup module.

8. A mine production safety risk monitoring and early warning method, applicable to the mine production safety risk monitoring and early warning system according to claims 1-7, characterized in that: The method comprises the following steps: Data collection: Collect multi-dimensional data on mine geology, production, equipment operation and personnel information; Data integration and storage: Integrate the collected multi-dimensional data on mine geology, production, equipment operation and personnel information, and store them in a unified data format; Correlation analysis and model building: Use multi-dimensional correlation analysis algorithms to conduct correlation analysis on data, build a real-time dynamic risk assessment model, and calculate risk levels; Real-time monitoring and early warning: Real-time data acquisition updates risk status, triggering an early warning when the early warning threshold is reached; Emergency decision support: Establish an emergency decision knowledge base to store previous mine safety accident cases and corresponding emergency response plans. When an early warning occurs, similar cases are retrieved from the knowledge base based on the current risk situation, and combined with real-time multi-dimensional data to provide reference plans for emergency decision-making.

9. The mine production safety risk monitoring and early warning method according to claim 8, characterized in that: In the data collection step, for the equipment operation data, when the change trend of the equipment operation parameters exceeds the set curve range, additional auxiliary sensors are started to collect supplementary data. In the real-time monitoring and early warning step, the server regularly obtains the latest data from the storage module and inputs it into the risk assessment model. The model output risk level is compared with the preset threshold. When the threshold is exceeded, the system sends an early warning message containing the risk location, type and level information.

Images