Mine risk area dynamic management method and system based on industrial internet of things

Abstract

This invention discloses a method and system for dynamic management of mine risk areas based on the Industrial Internet of Things (IIoT), relating to the technical field of mine management. The method includes: constructing a network of sensing devices to collect multi-source mine area data streams and transmitting them to a cloud processing center; performing data fusion modeling and situation analysis to construct a three-dimensional real-time mine situation map, while simultaneously pre-setting a risk rule base; predicting and marking risk areas to obtain a set of potential mine risk areas, while simultaneously retrieving multi-physics monitoring data; verifying and reviewing the set of potential mine risk areas to determine the target mine risk area set and perform risk quantification assessment and collaborative dynamic control. This invention solves the technical problems of lagging mine risk identification, severe information silos, and a lack of collaborative risk control methods in existing technologies, achieving comprehensive real-time perception of mine risk areas, dynamic and proactive prediction of risk situations, and improving the accuracy and reliability of risk identification and control.

Description

Technical Field

[0001] This invention relates to the technical field of mine management, specifically to a method and system for dynamic management of mine risk areas based on the Industrial Internet of Things. Background Technology

[0002] The geological structure within mines is highly variable, with risks from gas, coal dust, water, fire, and roof collapse intertwined and exhibiting significant dynamic evolution and spatiotemporal correlation. Traditional mine safety management models primarily rely on manual inspections, fixed-point monitoring, and periodic risk assessments. These methods suffer from limited monitoring scope, severe information silos, insufficient global situational awareness, and delayed risk identification. Responses are often only taken after disasters become apparent, hindering dynamic, accurate, and proactive risk management. Furthermore, traditional mine monitoring largely employs fixed sensors and wired data transmission, resulting in inflexible deployment, limited coverage, high data transmission latency, and weak anti-interference capabilities. Particularly in the dynamic identification and assessment of risk areas, there is a lack of comprehensive analysis of the coupling effects of multiple factors and environmental evolution, making it difficult to adapt to the rapid evolution and migration of risk areas during mining activities and hindering effective prediction and prevention of potential risks.

[0003] Therefore, current technologies suffer from technical problems such as lagging mine risk identification, severe information silos, and a lack of coordination in risk management. Summary of the Invention

[0004] This application provides a dynamic management method and system for mine risk areas based on the Industrial Internet of Things, which solves the technical problems of lagging mine risk identification, serious information silos, and lack of coordination in risk control measures in the existing technology. It achieves the technical effect of realizing comprehensive real-time perception of mine risk areas, dynamic and advanced prediction of risk situation, and improving the accuracy and reliability of risk identification and control.

[0005] This application provides a method for dynamic management of mine risk areas based on the Industrial Internet of Things (IIoT). The method includes: identifying key areas and deploying sensors in the target mine to construct a sensing device array network; collecting multi-source mine area data streams through the sensing device array network and transmitting the multi-source mine area data streams to a cloud processing center via an industrial-grade wireless network; performing data fusion modeling and situation analysis on the multi-source mine area data streams based on the cloud processing center to construct a three-dimensional real-time mine situation map, while pre-setting a risk rule base; using the risk rule base to predict and mark risk areas on the three-dimensional real-time mine situation map to obtain a set of potential mine risk areas, and simultaneously retrieving multi-physics field monitoring data of the potential mine risk area set and the upstream risk areas; verifying and reviewing the potential mine risk area set based on the multi-physics field monitoring data to determine the target mine risk area set, and performing risk quantification assessment and collaborative dynamic control on the target mine risk area set.

[0006] In a possible implementation, a sensing device array network is constructed, including: collecting geological exploration data, historical accident data, and work plan data of the target mine; identifying key areas from the geological exploration data, historical accident data, and work plan data to obtain a set of geological risk areas, a set of accident-prone areas, and a set of intensive work areas; taking the union of the set of geological risk areas, the set of accident-prone areas, and the set of intensive work areas as the key area set of the mine, and obtaining the monitoring requirement parameters of the key area set of the mine; designing and optimizing a sensor deployment scheme for the key area set of the mine according to the monitoring requirement parameters, and constructing a sensing device array network.

[0007] In a possible implementation, constructing a real-time 3D situation map of the mine includes: calling a data cleaning program based on the cloud processing center, the data cleaning program including outlier removal, missing value filling, spatiotemporal alignment, and format standardization; cleaning the multi-source mine area data stream according to the data cleaning program to obtain a usable multi-source mine area data stream; performing data fusion modeling by combining the geological exploration data of the target mine and the usable multi-source mine area data stream to generate a 3D spatial model of the mine; and performing situation analysis and visualization rendering on the 3D spatial model of the mine to construct a real-time 3D situation map of the mine.

[0008] In a possible implementation, generating a three-dimensional spatial model of the mine includes: using the geological exploration data of the target mine as a static base for three-dimensional modeling to obtain a three-dimensional model of the mine base; performing entity relationship mapping on the available multi-source mine area data streams to obtain a mine area entity relationship mapping diagram, specifically a four-level entity relationship diagram of mine-area-equipment-sensor; weighting and fusion of the available multi-source mine area data streams according to the accuracy information of sensor types to obtain a fused mine area data stream; and dynamically modeling the fused mine area data streams based on the mine area entity relationship mapping diagram, thereby generating a three-dimensional spatial model of the mine.

[0009] In a possible implementation, the three-dimensional spatial model of the mine is subjected to situation analysis and visualization rendering to construct a real-time three-dimensional situation map of the mine. This includes: constructing a situation assessment index system, which includes environmental safety indicators, equipment health indicators, and operational safety indicators; performing situation assessment and prediction on the fused data stream of the mine area based on the situation assessment index system to obtain mine situation index assessment parameters; performing color coding and parsing on each situation index in the situation assessment index system according to mine safety standards to construct mine situation index color coding rules; and performing visualization rendering on the three-dimensional spatial model of the mine based on the mine situation index color coding rules and the mine situation index assessment parameters to construct a real-time three-dimensional situation map of the mine.

[0010] In possible implementations, a pre-set risk rule base includes: pre-setting static rules for mine safety thresholds based on industry experience standards; performing safety rule mining and confidence screening on historical accident data of the target mine to obtain dynamic correlation rules for mine safety; editing and combining the static rules for mine safety thresholds and the dynamic correlation rules for mine safety according to a pre-set rule structure to construct an initial risk rule set; and conducting simulation tests and adjustments on the initial risk rule set in conjunction with actual mine risk cases to pre-set the risk rule base.

[0011] In a possible implementation, obtaining a set of potential mine risk areas includes: dividing the real-time three-dimensional mine situation map according to a preset grid size to obtain a three-dimensional mine grid situation map; using the risk rule base to perform traversal grid scanning and risk area prediction on the three-dimensional mine grid situation map to obtain a set of mine risk grid areas; and aggregating and marking adjacent areas in the set of mine risk grid areas to obtain a set of potential mine risk areas.

[0012] In a possible implementation, risk quantification assessment and collaborative dynamic management of the target mine risk area set include: extracting and assessing risk factors from the target mine risk area set to obtain a mine risk area factor set and a corresponding risk area factor degree set; conducting risk level assessment and management strategy analysis based on the mine risk area factor set and the corresponding risk area factor degree set to determine a mine area hierarchical management strategy; and using the mine area hierarchical management strategy to conduct collaborative dynamic management of the target mine risk area set.

[0013] In possible implementation methods, determining the mine area hierarchical management and control strategy includes: performing risk weighted assessment and level classification based on the mine risk area factor set and the corresponding risk area factor degree set according to the risk level classification system to obtain the risk area level information set; constructing a mine management and control strategy library, and performing management and control strategy matching and analysis on the mine risk area factor set and the risk area level information set based on the mine management and control strategy library to determine the mine area hierarchical management and control strategy.

[0014] This application also provides a dynamic management system for mine risk areas based on the Industrial Internet of Things (IIoT). The system includes: a mine data acquisition module for identifying key areas and deploying sensors in the target mine, constructing a sensing device array network, acquiring multi-source mine area data streams through the sensing device array network, and transmitting the multi-source mine area data streams to a cloud processing center via an industrial-grade wireless network; a data modeling and analysis module for performing data fusion modeling and situation analysis on the multi-source mine area data streams based on the cloud processing center, constructing a three-dimensional real-time mine situation map, and simultaneously pre-setting a risk rule base; a risk prediction and marking module for predicting and marking risk areas on the three-dimensional real-time mine situation map using the risk rule base, obtaining a set of potential mine risk areas, and simultaneously retrieving multi-physics field monitoring data from the potential mine risk area set and upstream risk areas; and a risk assessment and control module for verifying and reviewing the potential mine risk area set based on the multi-physics field monitoring data, determining the target mine risk area set, and performing risk quantification assessment and collaborative dynamic control of the target mine risk area set.

[0015] This application proposes a dynamic management method and system for mine risk areas based on the Industrial Internet of Things (IIoT). This system involves constructing a network of sensing devices to collect multi-source mine area data streams and transmit them to a cloud processing center. Data fusion modeling and situational analysis are performed to construct a real-time 3D mine situation map, while simultaneously pre-setting a risk rule base. Risk areas are predicted and marked to obtain a set of potential mine risk areas, while simultaneously retrieving multi-physics monitoring data. The set of potential mine risk areas is then verified and reviewed to determine the target mine risk area set, and risk quantification assessment and collaborative dynamic control are conducted. This addresses the technical problems of lagging mine risk identification, severe information silos, and a lack of collaborative risk control methods in existing technologies. It achieves comprehensive real-time perception of mine risk areas, dynamic and proactive prediction of risk situations, and improves the accuracy and reliability of risk identification and control. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the accompanying drawings of the embodiments of this disclosure will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 A schematic diagram of the process for dynamic management of mine risk areas based on the Industrial Internet of Things provided in this application embodiment.

[0018] Figure 2 A schematic diagram of the structure of a dynamic management system for mine risk areas based on the Industrial Internet of Things provided in this application embodiment.

[0019] Figure labeling: Mine data acquisition module 10, data modeling and analysis module 20, risk prediction and labeling module 30, risk assessment and control module 40. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description is provided in conjunction with the accompanying drawings and preferred embodiments, based on the specific implementation methods, structure, features, and effects of the present invention.

[0021] This application provides a method for dynamic management of mine risk areas based on the Industrial Internet of Things, such as... Figure 1 As shown, the method includes: Step S100: Identify key areas and deploy sensors in the target mine, construct a sensing device array network, collect multi-source mine area data streams through the sensing device array network, and transmit the multi-source mine area data streams to the cloud processing center through an industrial-grade wireless network.

[0022] Step S100 further includes: collecting geological exploration data, historical accident data, and work plan data of the target mine; identifying key areas in the geological exploration data, historical accident data, and work plan data to obtain a set of geological risk areas, a set of accident-prone areas, and a set of intensive work areas; taking the union of the set of geological risk areas, the set of accident-prone areas, and the set of intensive work areas as the key area set of the mine, and obtaining the monitoring requirement parameters of the key area set of the mine; designing and optimizing a sensor deployment scheme for the key area set of the mine according to the monitoring requirement parameters, and constructing a sensing device array network.

[0023] Preferably, key areas of the target mine are identified and sensors are deployed. Specifically, geological exploration data of the target mine, such as rock strata structure, fault distribution, and gas occurrence, historical accident data, such as accident type, location, and time, and work plan data, such as mining face advancement plan and roadway excavation sequence, are collected. Then, key areas are identified. For geological exploration data, areas with complex geological structures, rich gas, and abnormal hydrological conditions are identified to form a set of geological risk areas. For historical accident data, accident frequency and spatial distribution are statistically analyzed to delineate a set of high-incidence accident areas. For work plan data, areas where personnel and equipment are concentrated in the current and near future are determined to form a set of intensive work areas. Then, the geological risk area set, the high-incidence accident area set, and the intensive work area set are selected. The union of the domain sets yields a set of key mine areas covering geological risks, historical accident probabilities, and operational activity intensity. For different areas within this set of key mine areas, monitoring requirement parameters are analyzed and obtained, including but not limited to monitoring type (such as gas concentration, microseismic activity, and stress), sensor accuracy, sampling frequency, deployment density, and communication requirements. Finally, a sensor deployment scheme is designed according to the monitoring requirement parameters, including sensor selection, location planning, and network topology design. The deployment scheme for the key mine area set is then adjusted and optimized through coverage optimization, energy consumption balancing, and maximizing communication reliability to form an feasible sensing device array network deployment scheme. This constructs a sensing device array network to ensure that multi-source data from key areas can be effectively and reliably collected.

[0024] Preferably, a network of sensing devices is used to collect multi-source mine area data streams from the mine's physical environment according to its preset monitoring types and sampling frequencies. This includes, but is not limited to, environmental monitoring data such as gas and carbon monoxide concentrations, temperature, humidity, wind speed, and dust concentration; geomechanical data such as rock stress, strain, microseismic events, ground sound signals, and electromagnetic radiation intensity; equipment status data such as the operating parameters (current, voltage, vibration, temperature) and start / stop status of ventilation fans, water pumps, hoists, and mining equipment; and personnel and location data such as the precise location information and movement trajectories of workers and vehicles. Sensor nodes or local gateways in the sensing device network perform preliminary processing on the multi-source mine area data streams, such as analog-to-digital conversion, local filtering, and data packaging, and transmit them through an industrial-grade wireless network in the mine according to a preset communication protocol. This industrial-grade wireless network may be based on Wi-Fi. 6. A network with high anti-interference, low latency, and high reliability, using industrial protocols such as 5G private network, LoRa, and Zigbee, is responsible for transmitting data streams stably and in real time from various monitoring points underground to the cloud processing center on the mine surface through multi-hop or direct methods. Encryption and redundant transmission mechanisms are used during the transmission process to ensure the security and integrity of the data.

[0025] Step S200: Based on the cloud processing center, perform data fusion modeling and situation analysis on the multi-source mine area data stream to construct a three-dimensional real-time situation map of the mine, and at the same time, pre-set a risk rule base.

[0026] Step S200 further includes, according to the cloud processing center, invoking a data cleaning program, the data cleaning program including outlier removal, missing value filling, spatiotemporal alignment, and format standardization; cleaning the multi-source mine area data stream according to the data cleaning program to obtain a usable multi-source mine area data stream; combining the geological exploration data of the target mine and the usable multi-source mine area data stream to perform data fusion modeling to generate a three-dimensional spatial model of the mine; and performing situation analysis and visualization rendering on the three-dimensional spatial model of the mine to construct a three-dimensional real-time situation map of the mine.

[0027] Preferably, after receiving data streams from multiple mine areas, the cloud processing center automatically invokes a preset data cleaning program to perform standardization processing, including outlier removal, missing value imputation, spatiotemporal alignment, and format standardization. Outlier removal identifies and removes erroneous or invalid data points that significantly deviate from the normal range based on statistical methods (such as the 3σ principle) or physical thresholds. Missing value imputation fills in data gaps caused by transmission interruptions or sensor malfunctions using interpolation (such as spatiotemporal interpolation), data extrapolation based on associated sensors, or historical averages. Spatiotemporal alignment calibrates data from different sensors with different timestamps and spatial coordinates to the same time reference and the same spatial coordinate system (such as the mine's absolute coordinate system) to ensure data comparability and consistency in time and space. Format standardization converts all data into a unified structured data format, such as a specific JSON or binary protocol.

[0028] Preferably, the multi-source mine area data stream is cleaned according to the data cleaning procedure to output a complete, consistent, and formatted usable multi-source mine area data stream. This data is then combined with geological exploration data such as static roadways, strata, and structural models of the target mine to perform data fusion modeling. Specifically, based on the geological exploration data, an accurate three-dimensional static base model of the mine is established. Environmental parameters, equipment locations, stress values, etc., contained in the usable multi-source mine area data stream are mapped as attributes or dynamic layers according to their spatial coordinates through entity relationship mapping and superimposed onto the corresponding spatial locations of the three-dimensional static base model of the mine, such as roadway cross-sections and mining faces. Then, through spatial interpolation and data assimilation, discrete sensor data points are transformed into continuous or semi-continuous spatial fields, such as gas concentration fields and stress fields, thereby generating a digitally computable three-dimensional spatial model of the mine that integrates static geological structure and dynamic multi-source monitoring information.

[0029] Preferably, a situational analysis is performed on the 3D spatial model of the mine. This involves real-time calculation and evaluation of dynamic data in the 3D spatial model based on predefined rules, safety thresholds, or trend predictions to identify the current safety status, risk trends, and anomaly patterns. Then, visualization rendering is performed, including mapping the analysis results such as safety levels, exceeding limits, and warning signals to graphical attributes according to preset visualization coding rules, such as color, transparency, isosurfaces, and particle effects. The preset visualization coding rules may include using red to represent danger, green to represent safety, and isosurfaces to represent concentration distribution. Finally, a 3D graphics engine renders the visualization attributes with real-time analysis results onto the 3D spatial model of the mine, ultimately constructing a real-time 3D situational map of the mine that intuitively and in real-time reflects the overall safety and operational status of the mine.

[0030] Furthermore, step S200 also includes: using the geological exploration data of the target mine as a static base to perform three-dimensional modeling, obtaining a three-dimensional model of the mine base; performing entity relationship mapping identification on the available multi-source mine area data stream to obtain a mine area entity relationship mapping diagram, specifically a four-level entity relationship diagram of mine-area-equipment-sensor; weighting and weighting decision fusion of the available multi-source mine area data stream according to the accuracy information of sensor type to obtain a fused mine area data stream; and dynamically modeling the fused mine area data stream based on the mine area entity relationship mapping diagram, thereby generating a three-dimensional spatial model of the mine.

[0031] Preferably, 3D modeling software or engines are used to accurately reconstruct the 3D geometry and spatial structure of the mine's roadway network, goaf, strata, and coal seams based on geological exploration data such as the target mine's roadway layout map, stratigraphic profile, fault and structural distribution, and coal seam occurrence status. This yields a 3D model of the mine's base containing only fixed geological and engineering structures, serving as the static spatial framework and coordinate reference for all dynamic data overlay and spatial analysis. Then, entity relationship mapping is performed on the available multi-source mine area data streams, identifying each dynamic data record within the overall mine management context. The logical attribution relationship is specifically mapped as a four-level entity relationship diagram of mine-region-equipment-sensor. Among them, mine represents the mine entity to which the data belongs, region represents the specific area or working face where the data is collected, such as "103 upper roadway" and "5101 coal mining face", equipment represents the equipment or device that carries the sensor, such as "No. 1 gas extraction pump" and "roof pressure gauge array", and sensor represents the specific sensor unit that generates the data, such as "gas sensor_01" and "stress sensor_A3", and then outputs the entity relationship mapping diagram of the mine area.

[0032] Preferably, the same monitoring target may be provided by multiple sensors of different types or locations. Different confidence weights are assigned to each sensor type based on its accuracy information. The sensor accuracy is obtained from calibration certificates and historical performance statistics. Then, weighted decision fusion is performed using weighted average, evidence theory and other methods to comprehensively calculate the available multi-source mine area data streams from different sensors to obtain the mine area fused data stream, thereby improving the overall quality and reliability of the mine area data. The fused data stream of the mine area is superimposed onto the 3D model of the mine base for dynamic modeling. Specifically, the precise spatial coordinates of the specific sensor corresponding to each data point in the fused data stream of the mine area are determined in the 3D model of the mine base using the entity relationship mapping map of the mine area. Then, dynamic data such as gas concentration, stress, and equipment status in the fused data stream of the mine area are used as attributes and associated with the corresponding spatial coordinates. Through spatial interpolation, the attribute data of discrete points are transformed into a continuous or semi-continuous spatial distribution field covering the entire roadway or area, such as gas concentration cloud map and stress isosurface. Finally, a 3D spatial model of the mine that deeply integrates static geological structure and dynamic multi-source monitoring attributes is generated, which simultaneously presents the dynamic distribution and changes of information such as the physical structure of the mine and its internal environment and equipment status in space.

[0033] Furthermore, step S200 also includes: constructing a situation assessment index system, which includes environmental safety indicators, equipment health indicators, and operational safety indicators; performing situation assessment and prediction on the fused data stream of the mine area based on the situation assessment index system to obtain mine situation indicator assessment parameters; performing color coding analysis on each situation indicator in the situation assessment index system according to mine safety standards to construct mine situation indicator color coding rules; and performing visualization rendering on the three-dimensional spatial model of the mine based on the mine situation indicator color coding rules and the mine situation indicator assessment parameters to construct a real-time three-dimensional situation map of the mine.

[0034] Preferably, a structured situation assessment index system is defined to quantitatively assess the real-time safety and operational status of the mine, including environmental safety indicators, equipment health indicators, and operational safety indicators. Among them, environmental safety indicators are used to assess the safety of the working environment, such as gas concentration, carbon monoxide concentration, oxygen content, temperature, humidity, dust concentration, wind speed, roof delamination, roadway convergence, and micro-vibration energy. Equipment health indicators are used to assess the operational status of key equipment, such as the main ventilation fan's air pressure / volume, drainage pump operating current / head, coal mining machine / tunneling machine vibration and temperature, power supply system insulation status, conveyor belt misalignment and tension, etc. Operational safety indicators are used to assess the compliance and risks of production operations, such as whether personnel have entered restricted areas, overcrowding in work areas, special operation certification status, inspection task completion rate, and process connection timeliness, etc.

[0035] Preferably, the integrated data stream of the mine area is input into the situation assessment indicator system for situation assessment and prediction. This involves calculating, analyzing, and predicting each indicator based on its corresponding sensor readings, equipment status signals, and personnel positioning information. This may include comparing real-time values ​​with safety thresholds, predicting short-term trends based on historical data, and analyzing the correlation between multiple indicators. The resulting output is the mine situation indicator assessment parameter, i.e., the specific quantitative assessment result or predicted state of each indicator. For example, a gas concentration of 1.2% indicates a warning state with an upward trend. Color coding is applied to each situation indicator in the situation assessment indicator system. Based on industry and company-specific mine safety standards, different color codes are predefined for each indicator's "normal," "warning," "alarm," and "dangerous" states. For example, "gas concentration < 1.0%" is displayed as green (normal), "1.0% ≤ concentration < 1.5%" as yellow (warning), and "concentration ≥ 1.5%" as red (alarm). This establishes a complete and standardized color coding rule for mine situation indicators.

[0036] Preferably, the three-dimensional spatial model of the mine is visualized and rendered by combining the color coding rules and evaluation parameters of the mine status indicators. That is, on the three-dimensional spatial model of the mine, according to the evaluation results of the indicators associated with each spatial location (such as roadway cross section, equipment model), the location or area is colored, highlighted or overlaid with special effects according to the color coding rules. The three-dimensional graphics engine performs real-time rendering, and finally constructs a three-dimensional real-time status map of the mine. This map is used to intuitively, dynamically and comprehensively display the safety status, equipment health and operational risk status of the entire mine in the form of heat maps, etc., to ensure that managers can accurately grasp the real-time status of the entire mine and quickly locate risk areas.

[0037] Furthermore, step S200 also includes: pre-setting static rules for mine safety thresholds based on industry experience standards; performing safety rule mining and confidence screening on historical accident data of the target mine to obtain dynamic correlation rules for mine safety; editing and combining the static rules for mine safety thresholds and the dynamic correlation rules for mine safety according to the preset rule structure to construct an initial risk rule set; and conducting simulation tests and adjustments on the initial risk rule set in conjunction with actual mine risk cases to pre-set a risk rule library.

[0038] Preferably, industry experience standards for coded mines, such as coal mine safety regulations, metal and non-metal mine safety regulations, and clearly defined safety thresholds, are used to generate static rules for mine safety thresholds. For example, if the gas concentration sensor reading is ≥1.0%, an early warning is triggered; if the carbon monoxide concentration is ≥24ppm, an alarm is triggered; if the main ventilation fan stops operating for more than 10 minutes, it is determined to be a major ventilation failure. Historical accident data is obtained from the target mine's historical records, including multi-source monitoring data before and after the accident, environmental parameters, work records, etc., and safety rules are mined using Apriori association rule learning to analyze and determine the incident. Therefore, multiple factors that frequently co-occur during or before an event are identified, and confidence levels are then used to filter them. Associations with high confidence and practical explanatory power are retained to obtain dynamic association rules for mine safety. These rules directly reflect the unique, data-driven risk precursor patterns of a specific mine, making them more personalized and predictive. For example, when the frequency of microseismic events increases by 50% within 3 hours and the electromagnetic radiation intensity in a specific area increases by 30% simultaneously, the confidence level for a roof collapse accident occurring within the next 6 hours is 85%. When the mining face advances to within 50 meters of a fault, the amount of gas emission and the ground stress show a strong positive correlation, with a correlation coefficient > 0.8.

[0039] Preferably, the static rules for mine safety thresholds and the dynamic correlation rules for mine safety are uniformly formatted and logically combined according to a preset rule structure. This preset rule structure may consist of standardized condition-action logical statements, thereby constructing an initial risk rule set that includes both general safety baselines and mine-specific risk warning modes. Then, the initial risk rule set is simulated and tested using actual mine risk cases, such as independent historical cases not used for rule mining or simulated typical risk scenario datasets. Performance indicators such as warning accuracy, false alarm rate, missed alarm rate, and warning lead time are used for evaluation. The test results are then used to optimize and adjust the initial risk rules, including adjusting thresholds in static rules or correlation strength thresholds in dynamic rules, adding effective new rules, and removing rules that cause numerous false alarms or are invalid. Finally, a stable and reliable pre-set risk rule library is output.

[0040] Step S300: Use the risk rule base to predict and mark the risk areas of the three-dimensional real-time situation map of the mine to obtain a set of potential mine risk areas, and simultaneously retrieve the multi-physics field monitoring data of the potential mine risk area set and the upstream area of ​​the risk.

[0041] Step S300 further includes dividing the real-time three-dimensional situation map of the mine according to a preset grid size to obtain a three-dimensional grid situation map of the mine; using the risk rule base to perform traversal grid scanning and risk area prediction on the three-dimensional grid situation map of the mine to obtain a set of mine risk grid areas; and aggregating and marking adjacent areas of the set of mine risk grid areas to obtain a set of potential mine risk areas.

[0042] Preferably, according to a preset grid size, such as dividing the entire mine space into three-dimensional cubic grids with side lengths of 1 meter, 5 meters, or 10 meters, the target three-dimensional space of the mine covered by the real-time three-dimensional situation map is regularly divided and discretized into multiple grid units with regular geometric shapes and unique spatial indices. Each grid unit contains various real-time data corresponding to its spatial location, such as average gas concentration, stress value, equipment status, etc., to obtain a three-dimensional grid situation map of the mine. A risk rule base is used to traverse and scan each grid unit in the three-dimensional grid situation map of the mine, extracting the condition parameters required for the risk rules from all its real-time data, such as gas concentration, stress value, microseismic events, etc. within the grid, and then substituting these parameters into the risk rules. Each risk grid in the database undergoes logical judgment and calculation. If the triggering condition of any risk rule is met, the grid is determined to be a risk grid, and all risk grids are combined to form a set of mine risk grid regions. Then, based on the spatial adjacency relationship of the grids, such as face adjacency and edge adjacency, a region growth algorithm based on connectivity is used to cluster and merge multiple spatially adjacent or close risk grids into a connected, larger three-dimensional spatial region. Each aggregated connected region is assigned a unique identifier, and its spatial characteristics, such as three-dimensional boundary, center position, total volume, and main risk types contained therein, are calculated. Finally, a set of potential mine risk regions is obtained, which contains multiple spatially continuous three-dimensional risk region volumes with clear boundaries and comprehensive risk characteristics.

[0043] Preferably, based on the causes and propagation mechanisms of mine disasters such as gas outbursts, rock bursts, and water inrushes, the upstream risk areas of each potential risk zone are identified. These areas include at least the areas along the geological structure, stress transmission direction, or groundwater flow direction; the areas in the direction of origin of the current working face or disturbance source, such as the coal and rock mass above the working face, coal pillars on the side of the goaf, etc.; and areas located on the windward side or upstream supply direction of the potential risk zone according to the airflow or gas seepage path. Potential mine risk areas are simultaneously retrieved from multiphysics monitoring data stored in real-time or historical databases. The data collected includes multiphysics monitoring data from the current and past several hours to tens of hours in the upstream area of ​​the risk, which may include, but are not limited to, microseismic monitoring data such as event location, energy, magnitude, and frequency, used to invert rock mass fracturing and stress release; geosonic monitoring data such as high-frequency rock fracturing acoustic emission signals, used to identify the activity of local rock mass damage; electromagnetic radiation data such as the intensity and frequency of electromagnetic pulse signals generated during the loading and fracturing process of coal mine rock mass; internal stress and strain monitoring data of rock mass obtained through borehole stress gauges, fiber optic sensors, etc.; and geothermal field and seepage field data under specific conditions.

[0044] Step S400: Based on the multi-physics field monitoring data, verify and review the potential mine risk area set, determine the target mine risk area set, and conduct risk quantification assessment and collaborative dynamic management of the target mine risk area set.

[0045] Preferably, multiphysics monitoring data is used to verify and review the set of potential mine risk areas, including cross-validation and precursor consistency checks. Specifically, it checks whether the monitoring data of different physical fields within each potential risk area show mutually corroborating and co-evolving risk precursor patterns. For example, it determines whether the risk area simultaneously experiences microseismic events that change from scattered to concentrated with increasing energy, significant increases in electromagnetic radiation signal intensity and pulse number, and rapid stress changes in borehole stress gauge data. Then, a spatiotemporal evolution trend analysis is performed to assess whether the multiphysics parameters exhibit unstable trends such as acceleration and abrupt changes in time series, and whether they have a clear coupling relationship with the boundaries and centroids of the potential risk areas in space. Rock mechanics and gas dynamics models are then invoked, utilizing multiphysics... Using physical field data as input, risk assessment parameters such as stress concentration coefficient, energy accumulation index, and outburst hazard index are calculated within the region. Then, based on the test results, each potential risk area in the potential risk area set is adjudicated. If the multi-physics field data of the area shows a significant, consistent, and disaster-compliant abnormal precursor pattern, and reaches the danger threshold, it is confirmed as a real risk area. If the multi-physics field data of the area does not show a clear anomaly, or the anomaly pattern is inconsistent and cannot be mutually corroborated, or is only an isolated, transient signal, the area is determined to be a regular false alarm or a low-risk / observation area, and is excluded from or downgraded from the risk list that requires immediate action. Finally, a target mine risk area set with higher reliability and lower false alarm rate is obtained.

[0046] Furthermore, step S400 also includes extracting and evaluating risk factors from the target mine risk area set to obtain a mine risk area factor set and a corresponding risk area factor degree set; performing risk level assessment and control strategy analysis based on the mine risk area factor set and the corresponding risk area factor degree set to determine a mine area hierarchical control strategy; and using the mine area hierarchical control strategy to perform collaborative dynamic control of the target mine risk area set.

[0047] Preferably, risk factors causing the risk are identified and extracted from the multi-data associated with the target mine risk area set. These are specific measurable physical quantities or states, such as gas factors, stress / energy factors, and triggering factors. The severity of the current state of each extracted risk factor is quantified. For example, the severity of the gas factor is assessed as "1.8% concentration with accelerated outburst rate", and the severity of the stress factor is assessed as "energy index reaches twice the warning value and shows an upward trend". The mine risk area factor set is then output, which is the list of core risk factors corresponding to each risk area, and the corresponding risk area factor severity set, which is the quantified severity description. Risk level assessment is conducted based on a set of risk area factors and corresponding risk area factor severity sets. Taking into account the type, quantity, severity, and interactions of factors, a quantitative risk level is assigned to each risk area, such as Level I (Red / Extremely Serious), Level II (Orange / Serious), Level III (Yellow / Relatively Serious), and Level IV (Blue / Moderate). Simultaneously, a pre-built mine management strategy library defines standardized, tiered response and handling instructions for different combinations of risk factors and risk levels. Based on the factor types and assessed risk levels of the current risk area, the system automatically matches, parses, and combines specific, executable instruction sets from the strategy library, forming a tiered management strategy for the mine area. This tiered management strategy is then translated into actual management actions and distributed to corresponding execution terminals via an industrial IoT network. This enables collaborative and dynamic management of the target mine risk area set, continuously monitoring changes in relevant factors in the risk areas, dynamically evaluating the management effectiveness based on feedback data, and automatically adjusting the intensity or content of the management strategy to achieve precise, tiered, and collaborative response and handling of identified risks.

[0048] Furthermore, step S400 also includes: performing risk weighting assessment and level classification based on the mine risk area factor set and the corresponding risk area factor degree set according to the risk level classification system to obtain a risk area level information set; constructing a mine management and control strategy library; and performing management and control strategy matching and analysis on the mine risk area factor set and the risk area level information set based on the mine management and control strategy library to determine the mine area hierarchical management and control strategy.

[0049] Preferably, a pre-defined risk level classification system is used to comprehensively assess each risk area. The risk level classification system is used to define and assign different weight coefficients to different types of risk factors and different severity ranges. By weighted summing of the severity values ​​of each factor with their weight coefficients, a comprehensive risk value is calculated, and the comprehensive risk value range corresponding to different risk levels (such as red / orange / yellow / blue) is defined. Then, according to the risk level classification system, the risk area factor set and the corresponding risk area factor severity set are risk-weighted and classified, and a specific risk level is assigned to each risk area, thus forming a risk area level information set.

[0050] Preferably, a structured mine management strategy library is pre-configured based on historical mine management data. The trigger condition is a combination of risk factor type and risk level, and the response content is a specific management action instruction. For example, for a gas factor, Level I (red), the power supply and equipment in the area are immediately and automatically cut off, emergency extraction is initiated, a full evacuation order is issued, and the rescue team is notified to stand by. For a roof stress factor, Level II (orange), an audible and visual alarm is issued, non-essential personnel are restricted from entering, the support monitoring frequency is automatically increased to once per minute, and technical personnel are notified to conduct on-site verification. For a general environmental factor, Level III (yellow), a warning alarm is issued, prompting increased ventilation, and inspection personnel are notified to pay close attention. Then, the target mine risk area set is traversed. Based on the mine risk area factor set and risk area level information set corresponding to each area, a query and match is performed in the mine management strategy library to determine the strategy that best matches the trigger condition with the factor type and level of the current area. The matched strategy content is parsed and combined, and fine-tuned according to the actual risk level, ultimately generating a mine area hierarchical management strategy for that risk area.

[0051] In the above text, refer to Figure 1 This paper describes in detail a dynamic management method for mine risk areas based on the Industrial Internet of Things (IIoT) according to embodiments of the present invention. Next, reference will be made to... Figure 2 This invention describes a dynamic management system for mine risk areas based on the Industrial Internet of Things (IIoT) according to an embodiment of the present invention.

[0052] The dynamic management system for mine risk areas based on the Industrial Internet of Things (IIoT) according to embodiments of the present invention addresses the technical problems of lagging mine risk identification, severe information silos, and lack of coordination in risk control measures in existing technologies. It achieves comprehensive real-time perception of mine risk areas, dynamic and proactive prediction of risk situations, and improves the accuracy and reliability of risk identification and control. Figure 2 As shown, the dynamic management system for mine risk areas based on the Industrial Internet of Things includes: a mine data acquisition module 10, a data modeling and analysis module 20, a risk prediction and marking module 30, and a risk assessment and control module 40.

[0053] The mine data acquisition module 10 is used to identify key areas and deploy sensors in the target mine, construct a sensing device array network, acquire multi-source mine area data streams through the sensing device array network, and transmit the multi-source mine area data streams to the cloud processing center through an industrial-grade wireless network; the data modeling and analysis module 20 is used to perform data fusion modeling and situation analysis on the multi-source mine area data streams based on the cloud processing center, construct a three-dimensional real-time mine situation map, and simultaneously pre-set a risk rule base; the risk prediction and marking module 30 is used to predict and mark risk areas on the three-dimensional real-time mine situation map using the risk rule base, obtain a set of potential mine risk areas, and simultaneously retrieve multi-physics field monitoring data of the potential mine risk area set and the upstream risk area; the risk assessment and control module 40 is used to verify and review the set of potential mine risk areas based on the multi-physics field monitoring data, determine the target mine risk area set, and perform risk quantification assessment and collaborative dynamic control of the target mine risk area set.

[0054] The specific configuration of the mine data acquisition module 10 will be described in detail below. The mine data acquisition module 10 further includes: acquiring geological exploration data, historical accident data, and work plan data of the target mine; identifying key areas in the geological exploration data, historical accident data, and work plan data to obtain a set of geological risk areas, a set of accident-prone areas, and a set of intensive work areas; taking the union of the geological risk area set, the accident-prone area set, and the intensive work area set as the mine's key area set, and obtaining the monitoring requirement parameters for the mine's key area set; designing and optimizing a sensor deployment scheme for the mine's key area set according to the monitoring requirement parameters, and constructing a sensing device array network.

[0055] The specific configuration of the data modeling and analysis module 20 will be described in detail below. The data modeling and analysis module 20 further includes: calling a data cleaning program based on the cloud processing center; the data cleaning program includes outlier removal, missing value filling, spatiotemporal alignment, and format standardization; cleaning the multi-source mine area data stream according to the data cleaning program to obtain a usable multi-source mine area data stream; performing data fusion modeling by combining the geological exploration data of the target mine and the usable multi-source mine area data stream to generate a three-dimensional spatial model of the mine; and performing situation analysis and visualization rendering on the three-dimensional spatial model of the mine to construct a real-time three-dimensional situation map of the mine.

[0056] The specific configuration of the data modeling and analysis module 20 will be described in detail below. The data modeling and analysis module 20 further includes: using the geological exploration data of the target mine as a static base for three-dimensional modeling to obtain a three-dimensional model of the mine base; performing entity relationship mapping on the available multi-source mine area data stream to obtain a mine area entity relationship mapping diagram, specifically a four-level entity relationship diagram of mine-area-equipment-sensor; weighting and fusion of the available multi-source mine area data stream according to the accuracy information of the sensor type to obtain a fused mine area data stream; and dynamically modeling the fused mine area data stream by overlaying it onto the mine base three-dimensional model based on the mine area entity relationship mapping diagram to generate a three-dimensional spatial model of the mine.

[0057] The specific configuration of the data modeling and analysis module 20 will be described in detail below. The data modeling and analysis module 20 further includes: constructing a situation assessment index system, which includes environmental safety indicators, equipment health indicators, and operational safety indicators; performing situation assessment and prediction on the fused data stream of the mine area based on the situation assessment index system to obtain mine situation indicator assessment parameters; performing color coding analysis on each situation indicator in the situation assessment index system according to mine safety standards to construct mine situation indicator color coding rules; and performing visualization rendering on the mine's three-dimensional spatial model based on the mine situation indicator color coding rules and the mine situation indicator assessment parameters to construct a real-time three-dimensional situation map of the mine.

[0058] The following section will continue to describe the specific configuration of the data modeling and analysis module 20 in detail. The data modeling and analysis module 20 further includes: pre-setting static rules for mine safety thresholds based on industry experience standards; performing safety rule mining and confidence level filtering on historical accident data of the target mine to obtain dynamic correlation rules for mine safety; editing and combining the static rules for mine safety thresholds and the dynamic correlation rules for mine safety according to a pre-set rule structure to construct an initial risk rule set; and conducting simulation tests and adjustments on the initial risk rule set in conjunction with actual mine risk cases to pre-set a risk rule library.

[0059] The specific configuration of the risk prediction and marking module 30 will be described in detail below. The risk prediction and marking module 30 further includes: dividing the real-time three-dimensional mine situation map according to a preset grid size to obtain a three-dimensional mine grid situation map; using the risk rule base to perform traversal grid scanning and risk area prediction on the three-dimensional mine grid situation map to obtain a set of mine risk grid areas; aggregating and marking adjacent areas in the set of mine risk grid areas to obtain a set of potential mine risk areas.

[0060] The specific configuration of the risk assessment and control module 40 will be described in detail below. The risk assessment and control module 40 further includes: extracting and assessing risk factors from the target mine risk area set to obtain a mine risk area factor set and a corresponding risk area factor degree set; performing risk level assessment and control strategy analysis based on the mine risk area factor set and the corresponding risk area factor degree set to determine a mine area hierarchical control strategy; and using the mine area hierarchical control strategy to perform collaborative dynamic control of the target mine risk area set.

[0061] The specific configuration of the risk assessment and control module 40 will be described in detail below. The risk assessment and control module 40 further includes: performing risk weighted assessment and level classification based on the mine risk area factor set and the corresponding risk area factor degree set according to the risk level classification system, to obtain a risk area level information set; constructing a mine control strategy library; and performing control strategy matching and analysis on the mine risk area factor set and the risk area level information set based on the mine control strategy library to determine the mine area hierarchical control strategy.

[0062] The mine risk area dynamic management system based on industrial Internet of Things provided in this embodiment of the invention can execute the mine risk area dynamic management method based on industrial Internet of Things provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0063] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A dynamic management method for mine risk areas based on the Industrial Internet of Things, characterized in that, The method includes: Key areas of the target mine are identified and sensors are deployed to construct a network of sensing devices. Multi-source mine area data streams are collected through the sensing device network and transmitted to the cloud processing center through an industrial-grade wireless network. Based on the cloud processing center, data fusion modeling and situation analysis are performed on the multi-source mine area data stream to construct a three-dimensional real-time situation map of the mine, while a risk rule base is pre-set. The risk rule base is used to predict and mark the risk areas of the three-dimensional real-time situation map of the mine, and a set of potential mine risk areas is obtained. Simultaneously, the multi-physics field monitoring data of the potential mine risk area set and the upstream area of ​​the risk are retrieved. The potential mine risk area set is verified and reviewed based on the multi-physics field monitoring data to determine the target mine risk area set, and the target mine risk area set is subjected to risk quantification assessment and collaborative dynamic management.

2. The method for dynamic management of mine risk areas based on the Industrial Internet of Things as described in claim 1, characterized in that, Constructing a network of sensing device arrays includes: Collect geological exploration data, historical accident data, and work plan data of the target mine; Key areas are identified from the geological exploration data, historical accident data, and work plan data to obtain sets of geological risk areas, sets of accident-prone areas, and sets of intensive work areas. The union of the geological risk area set, the accident high-incidence area set, and the intensive operation area set is taken as the key area set of the mine, and the monitoring requirement parameters of the key area set of the mine are obtained. Based on the monitoring requirements parameters, a sensor deployment scheme is designed and optimized for the key areas of the mine, and a sensing device array network is constructed.

3. The dynamic management method for mine risk areas based on the Industrial Internet of Things as described in claim 2, characterized in that, Constructing a 3D real-time situation map of the mine, including: According to the cloud processing center, a data cleaning program is invoked, which includes outlier removal, missing value filling, spatiotemporal alignment, and format standardization. The data cleaning procedure is used to clean the multi-source mine area data stream to obtain a usable multi-source mine area data stream. By combining the geological exploration data of the target mine and the available multi-source mine area data stream, a data fusion model is generated to create a three-dimensional spatial model of the mine. The situation analysis and visualization rendering of the three-dimensional spatial model of the mine are performed to construct a real-time three-dimensional situation map of the mine.

4. The dynamic management method for mine risk areas based on the Industrial Internet of Things as described in claim 3, characterized in that, Generate a 3D spatial model of the mine, including: Using the geological exploration data of the target mine as a static base, a three-dimensional model of the mine base is obtained. The available multi-source mine area data stream is mapped and identified by entity relationships to obtain a mine area entity relationship mapping diagram. Specifically, the mine area entity relationship mapping diagram is a four-level entity relationship diagram of mine-area-equipment-sensor. The available multi-source mine area data streams are weighted according to the accuracy information of the sensor types and then fused by weighted decision to obtain the mine area fused data stream. Based on the entity relationship mapping diagram of the mine area, the fused data stream of the mine area is superimposed on the three-dimensional model of the mine base to perform dynamic modeling and generate a three-dimensional spatial model of the mine.

5. The dynamic management method for mine risk areas based on the Industrial Internet of Things as described in claim 4, characterized in that, The situation analysis and visualization rendering of the three-dimensional spatial model of the mine are performed to construct a real-time three-dimensional situation map of the mine, including: A situation assessment index system is constructed, which includes environmental safety indicators, equipment health indicators, and operational safety indicators. Based on the aforementioned situation assessment index system, the situation assessment and prediction of the fused data stream of the mine area are performed to obtain the mine situation index assessment parameters. According to the mine safety standards, each situation indicator in the situation assessment indicator system is color-coded and analyzed to construct a color coding rule for mine situation indicators. Based on the color coding rules of the mine situation indicators and the evaluation parameters of the mine situation indicators, the three-dimensional spatial model of the mine is visualized and rendered to construct a real-time three-dimensional situation map of the mine.

6. The method for dynamic management of mine risk areas based on the Industrial Internet of Things as described in claim 1, characterized in that, A pre-built risk rule base includes: Based on industry experience and standards, static rules for mine safety thresholds are preset. Safety rule mining and confidence screening are performed on the historical accident data of the target mine to obtain dynamic correlation rules for mine safety. The static rules for mine safety thresholds and the dynamic rules for mine safety association are edited and combined according to a preset rule structure to construct an initial risk rule set; The initial risk rule set was simulated and adjusted based on actual mine risk cases, and a pre-set risk rule library was established.

7. The method for dynamic management of mine risk areas based on the Industrial Internet of Things as described in claim 1, characterized in that, The set of potential mine risk areas is obtained, including: The three-dimensional real-time situation map of the mine is divided according to a preset grid size to obtain a three-dimensional grid situation map of the mine; The risk rule base is used to perform grid scanning and risk area prediction on the three-dimensional grid situation map of the mine to obtain the set of mine risk grid areas. The adjacent regions of the aforementioned mine risk grid area set are aggregated and labeled to obtain the potential mine risk area set.

8. The method for dynamic management of mine risk areas based on the Industrial Internet of Things as described in claim 1, characterized in that, The risk quantification assessment and collaborative dynamic management of the target mine risk area set include: Risk factors are extracted and evaluated from the target mine risk area set to obtain the mine risk area factor set and the corresponding risk area factor degree set. Based on the set of risk area factors and the corresponding set of risk area factor degrees, risk level assessment and control strategy analysis are performed to determine the graded control strategy for the mine area. The aforementioned mine area hierarchical management and control strategy is used to conduct collaborative and dynamic management and control of the target mine risk area set.

9. The dynamic management method for mine risk areas based on the Industrial Internet of Things as described in claim 8, characterized in that, Determine the hierarchical management and control strategy for mine areas, including: Based on the risk level classification system, risk weighting assessment and level classification are carried out according to the set of risk area factors and the corresponding set of risk area factor degrees of the mine, to obtain the risk area level information set. A mine management and control strategy library is constructed. Based on the mine management and control strategy library, management and control strategies are matched and analyzed for the mine risk area factor set and the risk area level information set to determine the mine area hierarchical management and control strategy.

10. A dynamic management system for mine risk areas based on the Industrial Internet of Things, characterized in that, The system is used to implement the dynamic management method for mine risk areas based on the Industrial Internet of Things as described in any one of claims 1 to 9, and the system includes: The mine data acquisition module is used to identify key areas and deploy sensors in the target mine, build a sensing device array network, collect multi-source mine area data streams through the sensing device array network, and transmit the multi-source mine area data streams to the cloud processing center through an industrial-grade wireless network. The data modeling and analysis module is used to perform data fusion modeling and situation analysis on the multi-source mine area data stream based on the cloud processing center, construct a three-dimensional real-time mine situation map, and pre-set a risk rule library. The risk prediction and marking module is used to predict and mark the risk areas of the three-dimensional real-time situation map of the mine using the risk rule base, to obtain a set of potential mine risk areas, and to simultaneously retrieve the multi-physics field monitoring data of the set of potential mine risk areas and the upstream area of ​​the risk. The risk assessment and control module is used to verify and review the potential mine risk area set based on the multi-physics field monitoring data, determine the target mine risk area set, and perform risk quantification assessment and collaborative dynamic control of the target mine risk area set.

Images