Intelligent safety monitoring method and system for coal mining

Abstract

The invention discloses an intelligent safety monitoring method and system for coal mine excavation, and belongs to the technical field of coal mine safety, and the method specifically comprises the steps: deploying a plurality of adaptive sensing nodes in an excavation area, each node comprising an environment state collection unit and an equipment vibration collection unit, and collecting environment data and mechanical vibration waveforms respectively; then, according to the position of the mining machine and the tunnel structure change, node physical connection is adjusted to form a dynamic topology network, and when mechanical movement exceeds a threshold value, adjacent nodes are subjected to high-density monitoring; then, synchronously generating a coupling data stream according to the collected data timestamps, inputting the coupling data stream into a pre-constructed model, and predicting the overlap ratio of mechanical vibration abnormity and a target gas accumulation area space; when the overlap ratio exceeds a critical value, an emergency response is triggered, the speed is reduced firstly, inert gas is released, and power is cut off if set duration is continued; and finally, reversely correcting surrounding rock stress parameters in the model according to emergency data. According to the invention, the accuracy and response timeliness of early warning of coal mine disasters are improved.

Description

Technical Field

[0001] The present invention relates to the field of coal mine safety technology, and in particular to an intelligent safety monitoring method and system for coal mine excavation. Background Art

[0002] Coal mining operations are complex and volatile, with hazards such as gas outbursts and rock bursts posing a serious threat to production safety. Currently, monitoring relies primarily on fixed sensor networks. However, the constant movement of mining machinery leads to frequent blind spots, making it difficult to achieve full, real-time monitoring of the entire working surface. With the advancement of intelligent mining, traditional static monitoring methods are no longer sufficient to meet the demand for accurate early warning.

[0003] Existing safety monitoring systems often utilize independently deployed environmental sensors and vibration detection devices, collecting data through wired or wireless networking. While these systems can capture basic safety parameters, they analyze each sensor data in isolation and lack a comprehensive assessment of the interaction between mechanical operations and the geological environment. Common threshold alarm mechanisms only provide simple early warnings based on a single indicator and are unable to identify complex disaster risks.

[0004] Existing fixed sensor networks are unable to obtain timely monitoring data during mining operations, resulting in the operating front becoming a monitoring blind spot. Furthermore, the separate analysis of environmental parameters and equipment status data makes it difficult to accurately warn of secondary disasters induced by mechanical vibration. These problems have seriously restricted the development of intelligent coal mine safety monitoring and urgently require innovative solutions. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent safety monitoring method and system for coal mining, which solves the following technical problems: Existing fixed sensor networks have monitoring blind spots; and environmental parameters and equipment status data are analyzed separately.

[0006] The purpose of the present invention can be achieved through the following technical solutions: An intelligent safety monitoring method for coal mining, comprising the following steps: Multiple adaptive sensing nodes are deployed in the mining area. These nodes include an environmental status acquisition unit and an equipment vibration acquisition unit. The environmental status acquisition unit acquires gas concentration, temperature and humidity data, and surrounding rock stress data at the location. The equipment vibration acquisition unit acquires the vibration waveform of the mining machinery. Based on the real-time location of the mining machinery and changes in the tunnel structure, the physical connection between adaptive sensing nodes is adjusted to form a dynamic topology network. When the mining machinery moves beyond a preset distance threshold, the node closest to the mining machinery switches to high-density monitoring mode, and other nodes enter a low-power standby state. Gas concentration, temperature, and humidity data are time-stamped and synchronized with mechanical vibration waveforms to generate a coupled data stream. This coupled data stream is then fed into a pre-built 3D geological-equipment dynamic model. This model overlays historical mining trajectories with current surrounding rock stress distribution to predict the spatial overlap between areas of mechanical vibration anomalies and target gas accumulation zones. When the spatial overlap exceeds a preset critical value, an emergency response is triggered, controlling the mining machinery to slow down and simultaneously releasing inert gas to the area with the highest overlap to suppress the risk. If this continues for a set period of time, the machinery power is cut off and emergency ventilation is activated. The surrounding rock stress parameters in the 3D geological-equipment dynamic model are reversely corrected based on the emergency response data.

[0007] As a further solution of the present invention: the deployment mode of the adaptive sensing node is: The adaptive sensing nodes are divided into first-class nodes and second-class nodes. The first-class nodes include an environmental status acquisition unit and an equipment vibration acquisition unit, while the second-class nodes only include an environmental status acquisition unit. The first-class nodes are fixed on the surface of the mining machine, and the second-class nodes are suspended at set intervals along the tunnel roof. A physical link is formed between the first-class nodes and the second-class nodes via a flexible connection device. When the mining machine moves forward, the first-class nodes extend the flexible connection device as the machine moves, and the second-class nodes automatically adjust their spatial posture according to the change in tension of the connection device. The extension direction of the flexible connection device is parallel to the propulsion direction of the mining machinery, and it automatically shrinks to its initial length when the machinery stops; data transmission between nodes is realized through the communication medium embedded in the flexible connection device, and at the same time, the link tension is detected in real time during the movement of the machinery, and the signal gain is dynamically adjusted to offset the impact of physical deformation on communication quality.

[0008] As a further solution of the present invention: the vibration sensing unit of the equipment is arranged on the surface of the rotating cutting part, power transmission part and walking support part of the mining machinery, and each vibration sensing unit is tightly fitted to the surface of the machine through a detachable fixing device, and the installation position avoids the mechanical movable joints and high-temperature areas; the waveform data collected by the vibration sensing unit includes vibration amplitude, frequency distribution characteristics and energy attenuation trend, and is transmitted to adjacent nodes in real time through the physical link between the adaptive sensing nodes; the vibration sensing unit continuously monitors the fit during the operation of the machine, and when a gap is detected on the fitting surface, the position calibration mechanism is automatically triggered, and the angle of the sensing unit is adjusted through the built-in fine-tuning mechanism until the preset fit pressure threshold is restored, and the data collected during the calibration process is marked as low-confidence data.

[0009] As a further solution of the present invention: the adjustment method of the dynamic topology network also includes: When any adaptive sensing node detects that the dust concentration or electromagnetic interference intensity in the environment exceeds the safety threshold, it analyzes the spectrum distribution, intensity fluctuation period and spatial attenuation characteristics of the current interference signal in real time to generate interference feature code; According to the interference feature code, the optimal anti-interference mode is selected from a preset communication mode library. If the interference is broadband random noise, the system switches to a time-sharing frequency band hopping mode, switching the transmission frequency band according to a pseudo-random sequence within each communication cycle. If the interference is a narrow-band periodic pulse, the system adopts a spectrum avoidance mode, concentrating the transmission of high-priority data within the interval window of the interference signal. Controlling the adaptive sensing node to disconnect the physical link connection with the adjacent node, activating the directional wireless transmission unit, and switching to the wireless transmission mode, wherein the directional wireless transmission unit adjusts the antenna beam direction according to the interference feature code so that the signal transmission path avoids the area with the highest interference intensity; After switching to wireless transmission mode, the current adaptive sensing node broadcasts status information including its own coordinates, interference signature code, and remaining battery power to the network. Adjacent nodes calculate the optimal link path to avoid the interference area based on the received status information and the current network topology map. During path optimization, adjacent nodes prioritize nodes that are farther away from interference sources than a preset safety value as relay points, and dynamically distribute the data transmission load of each node through a negotiation mechanism to avoid local network congestion. In wireless communication mode, the current adaptive perception node continuously monitors changes in environmental interference intensity. When it detects that the interference intensity falls below the safety threshold and the stability duration reaches the preset conditions, it automatically initiates a physical link reconstruction request, gradually restores the cable connection and synchronously updates the network topology configuration.

[0010] As a further solution of the present invention: when any adaptive sensing node loses its working ability due to environmental anomalies or hardware failure, the adjacent nodes in its physical link automatically start a redundant takeover program. The takeover node expands its own sensor detection range to the original coverage area of the failed node and adjusts the link length through the deformation compensation mechanism of the flexible connection device. At the same time, the takeover node inherits the weight distribution of the failed node in the three-dimensional geological-equipment dynamic model and temporarily increases its own data sampling frequency to cover the monitoring blind spot. The takeover state continues until the failed node recovers, during which time all monitoring data are marked with redundancy.

[0011] As a further solution of the present invention: the method for generating the coupled data stream is: The gas concentration data and the mechanical vibration waveform in the same time window are aligned in the time domain, and the corresponding gas concentration change rate when the amplitude increase in the vibration waveform exceeds the set threshold is extracted, and the correlation strength between the two in the spatial dimension is calculated; the correlation strength is obtained by the vector superposition of the vibration energy distribution and the gas diffusion trend, and the direction of the superimposed vector indicates the expansion path of the potential risk area; when the angle between the main axis direction of the vibration energy distribution and the vector direction of the gas diffusion trend is less than the preset angle, it is determined that the damage to the surrounding rock structure by mechanical vibration accelerates gas escape, and the angle value is used as the core weight factor of the spatial coincidence.

[0012] As a further solution of the present invention, the specific process of predicting the spatial overlap between the abnormal mechanical vibration area and the target gas accumulation area is as follows: Extracting the past operation trajectory data of the current mining machine from the historical mining database. The operation trajectory data includes the spatial coordinates of the mining path, the peak position of the mechanical vibration energy, and the corresponding timestamp. Based on the rock layer inclination and fault distribution characteristics of the current roadway, the historical trajectory data is spatially corrected to generate a corrected trajectory that matches the current geological conditions. Real-time collection of surrounding rock stress data within the mining area. Three-dimensional stress vectors are acquired at each monitoring point through a stress sensor array. These vectors are decomposed into vertical and horizontal shear stress components. Based on elastic mechanics theory, the stress diffusion boundary is calculated to generate a three-dimensional cloud map of the current geological stress distribution. The corrected trajectory is spatially superimposed with the three-dimensional stress cloud map to identify the intersection of the historical vibration energy peak area and the current high stress area as the initial prediction area of the mechanical vibration abnormal area; In the initial prediction area, the angle between the gas diffusion direction and the vibration energy propagation direction is calculated in combination with the real-time collected gas concentration gradient data. If the angle is less than a preset threshold, it is determined that the gas is accelerating along the cracks formed by the vibration damage, and the area is marked as a high-risk overlap area. Based on the real-time position and advancement speed of the mining machinery, the spatial expansion trend of the high-risk overlap area within a set time window in the future is predicted. The trend is quantified by multiplying the cumulative effect of mechanical vibration on surrounding rock damage in the historical trajectory and the current stress release rate, and finally a dynamic evolution map of spatial overlap is output.

[0013] As a further solution of the present invention: when the spatial overlap exceeds a preset critical value, the specific response mode is: When the spatial overlap is first detected to exceed a critical value, the cutting components of the mining machine are controlled to reduce their rotational speed to a preset safe range, and an inert gas is released into the target area through a gas suppression unit integrated in the adaptive sensing node; the release rate of the inert gas is positively correlated with the growth rate of the spatial overlap, and the release direction is in the inverse vector direction of the vibration energy distribution; If the spatial overlap does not drop below the critical value within the continuous monitoring period, the power output of the mining machinery will be cut off and the independently deployed emergency ventilation system in the tunnel will be activated. The emergency ventilation system adjusts the outlet angle according to the risk path predicted by the three-dimensional geological-equipment dynamic model to form a targeted gas dispersion channel.

[0014] As a further solution of the present invention: the specific process of the reverse correction is: The surrounding rock stress data collected during the emergency response phase is extracted and compared with historical data to identify the correspondence between stress release characteristics and mechanical vibration waveforms. If the corrected stress value distribution shows that the rock stability in a specific area has decreased by more than a set value, a dynamic attenuation coefficient is applied to this area in the three-dimensional geological-equipment dynamic model to reduce its vibration energy bearing threshold. The value of the attenuation coefficient is related to the stress release rate and the gas concentration recovery time, and is continuously iteratively updated in subsequent monitoring until the stress data in this area returns to a safe fluctuation range.

[0015] The present invention also includes an intelligent safety monitoring system for coal mining, which is used to implement the above-mentioned intelligent safety monitoring method for coal mining, comprising: Adaptive sensing nodes include an environmental status acquisition unit and an equipment vibration acquisition unit. The environmental status acquisition unit acquires gas concentration, temperature and humidity data, and surrounding rock stress data at the location, while the equipment vibration acquisition unit acquires the vibration waveform of the mining machinery. The node control module is used to adjust the physical connection between adaptive sensing nodes based on the real-time position of the mining machinery and changes in the tunnel structure, forming a dynamic topology network. When the mining machinery moves beyond a preset distance threshold, the node closest to the mining machinery switches to high-density monitoring mode, and other nodes enter a low-power standby state. An anomaly monitoring module is used to synchronize gas concentration, temperature and humidity data with mechanical vibration waveforms through time stamps to generate a coupled data stream. This coupled data stream is then input into a pre-built 3D geological-equipment dynamic model. The model predicts the spatial overlap between mechanical vibration anomaly areas and target gas accumulation areas by superimposing historical mining trajectories with current surrounding rock stress distribution. An emergency response module is used to trigger an emergency response when the spatial overlap exceeds a preset critical value, control the mining machinery to slow down, and simultaneously release inert gas to the area with the highest overlap to suppress the risk. If the overlap lasts for a set period of time, the mechanical power is cut off and emergency ventilation is activated; The model optimization module is used to reversely correct the surrounding rock stress parameters in the three-dimensional geological-equipment dynamic model based on emergency response data.

[0016] Beneficial effects of the present invention: The present invention solves the monitoring blind spot problem of traditional fixed sensor networks through dynamic topology network technology. Its core lies in deploying adaptive sensing nodes that move with mining machinery, and forming a scalable physical communication link through a flexible connection device, so that the first type of node can follow the movement of the machinery and maintain a stable connection with the second type of node, ensuring that the operation front is always within the monitoring coverage range; to solve the false alarm problem caused by isolated data analysis, an innovative multimodal data coupling analysis method is adopted to associate the mechanical vibration waveform with the gas concentration data through time stamp synchronization, and use the three-dimensional geological-equipment dynamic model to calculate the spatial overlap between the vibration abnormality area and the gas accumulation area, realizing the "mechanical vibration-surrounding rock damage-gas escape" disaster chain. Accurate prediction; in terms of response mechanism, it breaks through the limitations of the traditional single alarm mode and establishes a phased dynamic response strategy based on spatial overlap classification. It intelligently adjusts the operating status of mining machinery according to the risk level and implements targeted intervention, which not only ensures safety but also avoids excessive downtime; the specially designed model closed-loop optimization mechanism reversely corrects the surrounding rock stress parameters to make the prediction model continuously fit the actual situation on site, solving the problem of prediction deviation caused by the solidification of traditional model parameters; in addition, the innovative anti-interference communication mode and redundant fault-tolerant mechanism ensure the reliable operation of the system in harsh environments. When the node detects that the interference exceeds the standard, it can autonomously switch to directional wireless transmission and trigger network reorganization. Adjacent nodes can also adjust and take over the monitoring tasks of the failed node through flexible connection devices. The synergistic effect of these key technical features has realized the overall all-round and intelligent safety monitoring of the coal mining process, significantly improving the accuracy of disaster warning and the timeliness of response. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The present invention will be further described below with reference to the accompanying drawings.

[0018] Figure 1 This is a flow chart of an intelligent safety monitoring method for coal mining according to the present invention; Figure 2 This is a module schematic diagram of an intelligent safety monitoring system for coal mining according to the present invention. DETAILED DESCRIPTION

[0019] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0020] See also Figure 1 As shown, the present invention is an intelligent safety monitoring method for coal mining, comprising the following steps: First, multiple adaptive sensing nodes are deployed within the mining area. These nodes have two core units: an environmental status acquisition unit and an equipment vibration acquisition unit. The environmental status acquisition unit is responsible for acquiring key environmental data at the location, including gas concentration, which enables real-time monitoring of hazardous gases such as methane; temperature and humidity data, which accurately assesses the temperature and humidity conditions of the operating environment; and surrounding rock stress data, which provides information on the stresses acting on the surrounding rock and provides a basis for determining geological stability. The equipment vibration acquisition unit focuses on acquiring the vibration waveforms of mining machinery, analyzing these waveforms to provide insight into the operating status of the mining machinery.

[0021] Next, the physical connections between adaptive sensing nodes are adjusted based on the real-time location of the mining machinery and changes in the tunnel structure, thus forming a dynamic topological network. During the mining process, when the mining machinery moves beyond a preset distance threshold, the node closest to the mining machinery switches to high-density monitoring mode, allowing for more detailed and comprehensive monitoring of the surrounding environment and equipment operating status. At the same time, other nodes enter a low-power standby state, ensuring overall monitoring coverage while reducing energy consumption.

[0022] Next, the gas concentration, temperature, and humidity data acquired by the environmental status acquisition unit are time-stamped and synchronized with the mechanical vibration waveforms captured by the equipment vibration acquisition unit. These different types of data are integrated to create a coupled data stream. This coupled data stream is then input into a pre-built three-dimensional geological-equipment dynamic model. This model overlays historical mining trajectories with the current surrounding rock stress distribution to predict the spatial overlap between areas of abnormal mechanical vibration and target gas accumulation areas. Historical mining trajectories reflect the impact of past operations on the geological structure, while the current surrounding rock stress distribution reflects the current geological conditions. The combination of these two provides strong support for accurate prediction.

[0023] When the predicted spatial overlap exceeds a preset threshold, an emergency response mechanism is immediately triggered. This slows down mining machinery to reduce the likelihood of risk. Furthermore, inert gas is released in a targeted manner toward the area with the highest overlap, suppressing potential hazards, such as reducing flammable gas concentrations and preventing explosions. If this high-risk state persists for a set period of time, the system takes further action, shutting off mechanical power to prevent further serious consequences from equipment operation. It also initiates emergency ventilation to improve air quality in the work area and dilute hazardous gas concentrations.

[0024] Finally, based on the data generated during the emergency response, the surrounding rock stress parameters in the 3D geological-equipment dynamic model were reversed. Using data from actual emergency situations, the model parameters were optimized to better reflect the actual geological conditions and improve the accuracy of subsequent predictions.

[0025] In a preferred embodiment of the present invention, the deployment mode of the adaptive sensing node is: Adaptive sensing nodes are categorized into Type I and Type II nodes. Type I nodes are more comprehensive, integrating both environmental status and equipment vibration collection units, enabling simultaneous monitoring of both environmental conditions and the vibration of mining machinery. Type II nodes, on the other hand, are equipped solely with the environmental status collection unit and focus solely on collecting environmental data. In actual deployment, Type I nodes are securely fixed to the surface of the mining machinery, enabling them to closely track its movement and accurately acquire machine-related data. Type II nodes are suspended at predetermined intervals along the tunnel roof to achieve comprehensive monitoring coverage of the tunnel environment. A flexible connector establishes a physical link between Type I and Type II nodes, playing a critical role in the overall monitoring system. As the mining machinery advances, the Type I node moves with it, pulling the flexible connector to extend. During this process, the Type II node automatically adjusts its spatial posture based on the tension changes in the connector, ensuring consistent monitoring of its surroundings and maintaining a good connection with other nodes. The flexible connector has a unique characteristic: its extension direction is strictly parallel to the mining machinery's propulsion direction. This ensures that the extension of the connector during operation does not interfere with other equipment or operations. When the machine stops operating, the flexible connector automatically retracts to its initial length, facilitating subsequent operations and equipment maintenance. Data transmission between nodes relies on the communication medium embedded in the flexible connector. Furthermore, during machine movement, the system monitors link tension in real time and dynamically adjusts signal gain based on tension changes. This offsets the negative impact of physical deformation on communication quality, ensuring stable and accurate data transmission.

[0026] In a preferred embodiment of this embodiment, equipment vibration sensing units are mounted on the surfaces of the rotating cutting component, power transmission component, and travel support component of the mining machine—areas critical for reflecting the machine's operating status. Each vibration sensing unit adheres tightly to the machine surface using removable fasteners. During installation, locations are carefully avoided around movable joints and high-temperature areas to prevent the complex environments of these areas from affecting the unit's proper operation. The waveform data collected by the vibration sensing unit covers several key aspects, including vibration amplitude, which provides a visual indicator of vibration intensity; frequency distribution, which helps analyze vibration frequency during machine operation; and energy attenuation trends, which provide a basis for assessing issues such as wear on machine components. This waveform data is transmitted in real time to adjacent nodes via physical links between adaptive sensing nodes for subsequent comprehensive analysis. Furthermore, the vibration sensing unit continuously monitors its contact with the machine surface during operation. If a gap is detected in the contact surface, a position calibration mechanism is automatically triggered. A built-in fine-tuning mechanism adjusts the sensing unit angle until it returns to a preset contact pressure threshold. The data collected during this calibration process may have data deviations due to unstable fitting status, so they are marked as low-confidence data so that they can be distinguished and processed in subsequent data analysis.

[0027] In another preferred embodiment of the present invention, the adjustment method of the dynamic topology network further includes: When any adaptive sensing node detects that the dust concentration or electromagnetic interference intensity in its environment exceeds a safety threshold, it immediately conducts a detailed analysis of the interference signal. Specifically, it analyzes the interference signal's spectral distribution, intensity fluctuation period, and spatial attenuation characteristics in real time, integrating this critical information to generate an interference signature code. This code accurately describes the current interference characteristics and provides a basis for selecting the appropriate anti-interference mode.

[0028] Based on the generated interference signature code, the system selects the optimal anti-interference mode from a preset communication mode library. If the interference manifests as broadband random noise, the node switches to time-sharing frequency hopping mode. In this mode, the node switches transmission frequency bands according to a pseudo-random sequence within each communication cycle, preventing the signal from being in the interfered frequency band for a long time and improving communication stability. If the interference manifests as narrowband periodic pulses, the node adopts spectrum avoidance mode, concentrating transmission of high-priority data within the intervening window of the interference signal, ensuring the timely and accurate transmission of important data.

[0029] To further minimize the impact of interference on communications, the system controls the adaptive sensing node to disconnect its physical link with neighboring nodes and activate the directional wireless transmission unit, switching to wireless transmission mode. The directional wireless transmission unit adjusts the antenna beam direction based on the interference signature code, directing the signal transmission path away from areas with the highest interference intensity, thereby reducing interference.

[0030] After switching to wireless transmission mode, the current adaptive sensing node broadcasts status information to the network, including its coordinates, interference signature code, and remaining battery life. Neighboring nodes receive this status information and, combined with the current network topology map, calculate the optimal link path to circumvent the interference area. During path optimization, neighboring nodes prioritize nodes with a spatial distance from the interference source greater than a preset safety threshold as relay points. A negotiation mechanism dynamically distributes data transmission load among nodes, preventing local network congestion and ensuring overall network communication efficiency.

[0031] In wireless communication mode, the current adaptive sensing node continuously monitors changes in environmental interference intensity. When it detects that the interference intensity has fallen below a safety threshold and that the stability duration meets a preset condition, the node automatically initiates a physical link reestablishment request, gradually restoring the cable connection and simultaneously updating the network topology configuration, restoring the network to normal physical link communication.

[0032] In a preferred scenario of this embodiment, when any adaptive sensing node loses its ability to work due to an environmental anomaly or hardware failure, the adjacent nodes in the physical link to which it belongs will automatically start a redundant takeover program. The takeover node will expand the detection range of its own sensor to cover the original coverage area of the failed node, and at the same time adjust the link length through the deformation compensation mechanism of the flexible connection device to ensure the continuity of monitoring. In addition, the takeover node will inherit the weight distribution of the failed node in the three-dimensional geological-equipment dynamic model and temporarily increase its own data sampling frequency to fill the monitoring blind spot. This takeover state will continue until the failed node resumes normal operation. During this period, all monitoring data will be attached with redundant tags to facilitate subsequent differentiation and processing of the data.

[0033] In another preferred embodiment of the present invention, the method for generating the coupled data stream is: First, the gas concentration data and the mechanical vibration waveform within the same time window are aligned in the time domain to ensure that the two correspond in the time dimension. Next, the gas concentration change rate corresponding to the increase in the vibration waveform amplitude exceeding a set threshold is extracted to explore the connection between the change in mechanical vibration amplitude and the change in gas concentration. The spatial correlation strength between the two is then calculated. This correlation strength is obtained by superimposing the vectors of the vibration energy distribution and the gas diffusion trend. The direction of the superimposed vector can indicate the expansion path of the potential risk area, providing a basis for early risk prediction. When the angle between the main axis direction of the vibration energy distribution and the vector direction of the gas diffusion trend is less than a preset angle, it means that the mechanical vibration damage to the surrounding rock structure has accelerated gas escape. At this time, the angle value is used as the core weighting factor for the spatial overlap and is used in the subsequent assessment of the spatial overlap between the abnormal mechanical vibration area and the target gas accumulation area.

[0034] In a preferred embodiment of the present invention, the specific process of predicting the spatial overlap between the abnormal mechanical vibration region and the target gas accumulation region is as follows: First, the historical trajectory data of the current mining machine is extracted from a historical mining database. This data includes the spatial coordinates of the mining path, the peak location of the mechanical vibration energy, and the corresponding timestamp. Taking into account the impact of different geological conditions on mining operations, the historical trajectory data is spatially corrected based on the rock layer inclination and fault distribution characteristics of the current roadway, thereby generating a corrected trajectory that matches the current geological conditions.

[0035] Secondly, real-time rock stress data within the mining area is collected, and a stress sensor array is used to obtain three-dimensional stress vectors at each monitoring point. These vectors are decomposed into vertical and horizontal shear stress components. The stress diffusion boundary is then calculated based on elastic mechanics theory, ultimately generating a three-dimensional cloud map of the current geological stress distribution, visually presenting the stress status of the current mining area.

[0036] Then, the corrected trajectory is spatially superimposed with the three-dimensional stress cloud map to identify the intersection range of the historical vibration energy peak area and the current high stress area, and this intersection range is used as the initial prediction area of the mechanical vibration abnormal area.

[0037] Next, within the initial prediction area, the angle between the gas diffusion direction and the vibration energy propagation direction is calculated using real-time gas concentration gradient data. If this angle is less than a preset threshold, it indicates that gas is rapidly accumulating along the cracks formed by vibration damage, and the area is marked as a high-risk overlap zone.

[0038] Finally, based on the real-time position and advancement speed of the mining machinery, the spatial expansion trend of the high-risk overlap zone within a set time window is predicted. This trend is quantified by multiplying the cumulative effect of mechanical vibration on surrounding rock damage in the historical trajectory with the current stress release rate, ultimately outputting a dynamic evolutionary map of spatial overlap.

[0039] In another preferred embodiment of the present invention, when the spatial overlap exceeds a preset critical value, the specific response mode is: When it is first detected that the spatial overlap exceeds the critical value, the system will immediately control the cutting components of the mining machinery to reduce the speed so that it is within the preset safe speed range, thereby reducing the degree of damage to the surrounding rock caused by mechanical vibration and reducing the possibility of gas escape. At the same time, inert gas is released to the target area through the gas suppression unit integrated in the adaptive sensing node. The release rate of inert gas is positively correlated with the growth rate of spatial overlap, that is, the faster the spatial overlap grows, the higher the inert gas release rate. In addition, the release direction advances along the inverse vector direction of the vibration energy distribution, which can more effectively suppress the accumulation and diffusion of gas in the target area.

[0040] If the spatial overlap fails to fall below the critical value within the continuous monitoring period, current countermeasures are failing to effectively mitigate the risk. In this case, the system will shut down the mining machinery, halting operations to prevent further deterioration. Simultaneously, the independently deployed emergency ventilation system within the tunnel will be activated. This system adjusts the outlet angle based on the risk path predicted by the 3D geological-equipment dynamic model, creating a targeted gas dispersal channel and accelerating the discharge of hazardous gases from the target area into the tunnel.

[0041] In another preferred embodiment of the present invention, the specific process of the reverse correction is: First, the surrounding rock stress data collected during the emergency response phase is extracted and compared with historical data to identify the correspondence between stress release characteristics and mechanical vibration waveforms. If the corrected stress value distribution shows that the rock stability in a specific area has decreased by more than the set value, it means that the geological conditions in the area have changed significantly and there is a high safety risk. At this time, a dynamic attenuation coefficient is applied to the area in the three-dimensional geological-equipment dynamic model to reduce its vibration energy bearing threshold. The value of this attenuation coefficient is related to the stress release rate and the gas concentration recovery time. The faster the stress release rate and the longer the gas concentration recovery time, the larger the attenuation coefficient value. In the subsequent monitoring process, the attenuation coefficient will be continuously iterated and updated until the stress data in the area returns to the safe fluctuation range, so that the model can more accurately reflect the actual geological conditions.

[0042] See also Figure 2As shown, the present invention also includes an intelligent safety monitoring system for coal mining, which is used to implement the above-mentioned intelligent safety monitoring method for coal mining, including: Adaptive sensing nodes include an environmental status acquisition unit and an equipment vibration acquisition unit. The environmental status acquisition unit acquires gas concentration, temperature and humidity data, and surrounding rock stress data at the location, while the equipment vibration acquisition unit acquires the vibration waveform of the mining machinery. The node control module is used to adjust the physical connection between adaptive sensing nodes based on the real-time position of the mining machinery and changes in the tunnel structure, forming a dynamic topology network. When the mining machinery moves beyond a preset distance threshold, the node closest to the mining machinery switches to high-density monitoring mode, and other nodes enter a low-power standby state. An anomaly monitoring module is used to synchronize gas concentration, temperature and humidity data with mechanical vibration waveforms through time stamps to generate a coupled data stream. This coupled data stream is then input into a pre-built 3D geological-equipment dynamic model. The model predicts the spatial overlap between mechanical vibration anomaly areas and target gas accumulation areas by superimposing historical mining trajectories with current surrounding rock stress distribution. An emergency response module is used to trigger an emergency response when the spatial overlap exceeds a preset critical value, control the mining machinery to slow down, and simultaneously release inert gas to the area with the highest overlap to suppress the risk. If the overlap lasts for a set period of time, the mechanical power is cut off and emergency ventilation is activated; The model optimization module is used to reversely correct the surrounding rock stress parameters in the three-dimensional geological-equipment dynamic model based on emergency response data.

[0043] The above is a detailed description of an embodiment of the present invention. However, the content described is only a preferred embodiment of the present invention and should not be considered to limit the scope of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. An intelligent safety monitoring method for coal mining, characterized in that: The following steps are involved: Multiple adaptive sensing nodes are deployed in the mining area. These nodes include an environmental status acquisition unit and an equipment vibration acquisition unit. The environmental status acquisition unit acquires gas concentration, temperature and humidity data, and surrounding rock stress data at the location. The equipment vibration acquisition unit acquires the vibration waveform of the mining machinery. Based on the real-time location of the mining machinery and changes in the tunnel structure, the physical connection between adaptive sensing nodes is adjusted to form a dynamic topology network. When the mining machinery moves beyond a preset distance threshold, the node closest to the mining machinery switches to high-density monitoring mode, and other nodes enter a low-power standby state. Gas concentration, temperature, and humidity data are time-stamped and synchronized with mechanical vibration waveforms to generate a coupled data stream. This coupled data stream is then fed into a pre-built 3D geological-equipment dynamic model. This model overlays historical mining trajectories with current surrounding rock stress distribution to predict the spatial overlap between areas of mechanical vibration anomalies and target gas accumulation zones. When the spatial overlap exceeds a preset critical value, an emergency response is triggered, controlling the mining machinery to slow down and simultaneously releasing inert gas to the area with the highest overlap to suppress the risk. If this continues for a set period of time, the machinery power is cut off and emergency ventilation is activated. The surrounding rock stress parameters in the 3D geological-equipment dynamic model are reversely corrected based on the emergency response data.

2. The intelligent safety monitoring method for coal mining according to claim 1, characterized in that: The deployment mode of the adaptive sensing node is as follows: The adaptive sensing nodes are divided into first-class nodes and second-class nodes. The first-class nodes include an environmental status acquisition unit and an equipment vibration acquisition unit, while the second-class nodes only include an environmental status acquisition unit. The first-class nodes are fixed on the surface of the mining machine, and the second-class nodes are suspended at set intervals along the tunnel roof. A physical link is formed between the first-class nodes and the second-class nodes via a flexible connection device. When the mining machine moves forward, the first-class nodes extend the flexible connection device as the machine moves, and the second-class nodes automatically adjust their spatial posture according to the change in tension of the connection device. The extension direction of the flexible connection device is parallel to the propulsion direction of the mining machine, and automatically shrinks to the initial length when the machine stops; Data transmission between nodes is achieved through the communication medium embedded in the flexible connection device. At the same time, the link tension is detected in real time during mechanical movement, and the signal gain is dynamically adjusted to offset the impact of physical deformation on communication quality.

3. The intelligent safety monitoring method for coal mining according to claim 2, characterized in that: The vibration sensing unit of the equipment is arranged on the surface of the rotating cutting part, power transmission part and walking support part of the mining machinery. Each vibration sensing unit is tightly fitted to the surface of the machine through a detachable fixing device, and the installation position avoids the mechanical movable joints and high-temperature areas; the waveform data collected by the vibration sensing unit includes vibration amplitude, frequency distribution characteristics and energy attenuation trend, and is transmitted to adjacent nodes in real time through the physical link between the adaptive sensing nodes; the vibration sensing unit continuously monitors the fit during the operation of the machine, and when a gap is detected on the fitting surface, the position calibration mechanism is automatically triggered, and the angle of the sensing unit is adjusted through the built-in fine-tuning mechanism until the preset fit pressure threshold is restored, and the data collected during the calibration process is marked as low-confidence data.

4. The intelligent safety monitoring method for coal mining according to claim 1, characterized in that: The adjustment method of the dynamic topology network also includes: When any adaptive sensing node detects that the dust concentration or electromagnetic interference intensity in the environment exceeds the safety threshold, it analyzes the spectrum distribution, intensity fluctuation period and spatial attenuation characteristics of the current interference signal in real time to generate interference feature code; According to the interference feature code, the optimal anti-interference mode is selected from a preset communication mode library. If the interference is broadband random noise, the system switches to a time-sharing frequency band hopping mode, switching the transmission frequency band according to a pseudo-random sequence within each communication cycle. If the interference is a narrow-band periodic pulse, the system adopts a spectrum avoidance mode, concentrating the transmission of high-priority data within the interval window of the interference signal. Controlling the adaptive sensing node to disconnect the physical link connection with the adjacent node, activating the directional wireless transmission unit, and switching to the wireless transmission mode, wherein the directional wireless transmission unit adjusts the antenna beam direction according to the interference feature code so that the signal transmission path avoids the area with the highest interference intensity; After switching to wireless transmission mode, the current adaptive sensing node broadcasts status information including its own coordinates, interference signature code, and remaining battery power to the network. Adjacent nodes calculate the optimal link path to avoid the interference area based on the received status information and the current network topology map. During path optimization, adjacent nodes prioritize nodes that are farther away from interference sources than a preset safety value as relay points, and dynamically distribute the data transmission load of each node through a negotiation mechanism to avoid local network congestion. In wireless communication mode, the current adaptive perception node continuously monitors changes in environmental interference intensity. When it detects that the interference intensity falls below the safety threshold and the stability duration reaches the preset conditions, it automatically initiates a physical link reconstruction request, gradually restores the cable connection and synchronously updates the network topology configuration.

5. The intelligent safety monitoring method for coal mining according to claim 4, characterized in that: When any adaptive sensing node loses its operating capability due to an environmental anomaly or hardware failure, the adjacent nodes in its physical link automatically initiate a redundant takeover procedure. The takeover node extends its own sensor detection range to the area originally covered by the failed node and adjusts the link length through the deformation compensation mechanism of the flexible connection device. At the same time, the takeover node inherits the weight distribution of the failed node in the 3D geological-equipment dynamic model and temporarily increases its own data sampling frequency to cover the monitoring blind spot. The takeover state lasts until the failed node recovers, during which time all monitoring data are marked as redundant.

6. The intelligent safety monitoring method for coal mining according to claim 1, characterized in that: The method for generating the coupled data stream is: The gas concentration data and the mechanical vibration waveform in the same time window are aligned in the time domain, and the corresponding gas concentration change rate when the amplitude increase in the vibration waveform exceeds the set threshold is extracted, and the correlation strength between the two in the spatial dimension is calculated; the correlation strength is obtained by the vector superposition of the vibration energy distribution and the gas diffusion trend, and the direction of the superimposed vector indicates the expansion path of the potential risk area; when the angle between the main axis direction of the vibration energy distribution and the vector direction of the gas diffusion trend is less than the preset angle, it is determined that the damage to the surrounding rock structure by mechanical vibration accelerates gas escape, and the angle value is used as the core weight factor of the spatial coincidence.

7. The intelligent safety monitoring method for coal mining according to claim 6, characterized in that: The specific process of predicting the spatial overlap between the abnormal mechanical vibration area and the target gas accumulation area is as follows: Extracting the past operation trajectory data of the current mining machine from the historical mining database. The operation trajectory data includes the spatial coordinates of the mining path, the peak position of the mechanical vibration energy, and the corresponding timestamp. Based on the rock layer inclination and fault distribution characteristics of the current roadway, the historical trajectory data is spatially corrected to generate a corrected trajectory that matches the current geological conditions. Real-time collection of surrounding rock stress data within the mining area. Three-dimensional stress vectors are acquired at each monitoring point through a stress sensor array. These vectors are decomposed into vertical and horizontal shear stress components. Based on elastic mechanics theory, the stress diffusion boundary is calculated to generate a three-dimensional cloud map of the current geological stress distribution. The corrected trajectory is spatially superimposed with the three-dimensional stress cloud map to identify the intersection of the historical vibration energy peak area and the current high stress area as the initial prediction area of the mechanical vibration abnormal area; In the initial prediction area, the angle between the gas diffusion direction and the vibration energy propagation direction is calculated in combination with the real-time collected gas concentration gradient data. If the angle is less than a preset threshold, it is determined that the gas is accelerating along the cracks formed by the vibration damage, and the area is marked as a high-risk overlap area. Based on the real-time position and advancement speed of the mining machinery, the spatial expansion trend of the high-risk overlap area within a set time window in the future is predicted. The trend is quantified by multiplying the cumulative effect of mechanical vibration on surrounding rock damage in the historical trajectory and the current stress release rate, and finally a dynamic evolution map of spatial overlap is output.

8. The intelligent safety monitoring method for coal mining according to claim 1, characterized in that: When the spatial overlap exceeds a preset critical value, the specific response is: When the spatial overlap is first detected to exceed a critical value, the cutting components of the mining machine are controlled to reduce their rotational speed to a preset safe range, and an inert gas is released into the target area through a gas suppression unit integrated in the adaptive sensing node; the release rate of the inert gas is positively correlated with the growth rate of the spatial overlap, and the release direction is in the inverse vector direction of the vibration energy distribution; If the spatial overlap does not drop below the critical value within the continuous monitoring period, the power output of the mining machinery will be cut off and the independently deployed emergency ventilation system in the tunnel will be activated. The emergency ventilation system adjusts the outlet angle according to the risk path predicted by the three-dimensional geological-equipment dynamic model to form a targeted gas dispersion channel.

9. The intelligent safety monitoring method for coal mining according to claim 1, characterized in that: The specific process of the reverse correction is: The surrounding rock stress data collected during the emergency response phase is extracted and compared with historical data to identify the correspondence between stress release characteristics and mechanical vibration waveforms. If the corrected stress value distribution shows that the rock stability in a specific area has decreased by more than a set value, a dynamic attenuation coefficient is applied to this area in the three-dimensional geological-equipment dynamic model to reduce its vibration energy bearing threshold. The value of the attenuation coefficient is related to the stress release rate and the gas concentration recovery time, and is continuously iteratively updated in subsequent monitoring until the stress data in this area returns to a safe fluctuation range.

10. An intelligent safety monitoring system for coal mining, used to implement the intelligent safety monitoring method for coal mining according to any one of claims 1 to 9, characterized in that: include: Adaptive sensing nodes include an environmental status acquisition unit and an equipment vibration acquisition unit. The environmental status acquisition unit acquires gas concentration, temperature and humidity data, and surrounding rock stress data at the location, while the equipment vibration acquisition unit acquires the vibration waveform of the mining machinery. The node control module is used to adjust the physical connection between adaptive sensing nodes based on the real-time position of the mining machinery and changes in the tunnel structure, forming a dynamic topology network. When the mining machinery moves beyond a preset distance threshold, the node closest to the mining machinery switches to high-density monitoring mode, and other nodes enter a low-power standby state. An anomaly monitoring module is used to synchronize gas concentration, temperature and humidity data with mechanical vibration waveforms through time stamps to generate a coupled data stream. This coupled data stream is then input into a pre-built 3D geological-equipment dynamic model. The model predicts the spatial overlap between mechanical vibration anomaly areas and target gas accumulation areas by superimposing historical mining trajectories with current surrounding rock stress distribution. An emergency response module is used to trigger an emergency response when the spatial overlap exceeds a preset critical value, control the mining machinery to slow down, and simultaneously release inert gas to the area with the highest overlap to suppress the risk. If the overlap lasts for a set period of time, the mechanical power is cut off and emergency ventilation is activated; The model optimization module is used to reversely correct the surrounding rock stress parameters in the three-dimensional geological-equipment dynamic model based on emergency response data.

Images