Coal mine goaf fire monitoring method and system based on multi-source data fusion

Abstract

The invention discloses a coal mine goaf fire monitoring method and system based on multi-source data fusion, and relates to the technical field of coal mine goaf fire monitoring, and the method comprises the steps: collecting the concentration, concentration and air pressure of each wireless sensing node in a coal mine goaf in real time; according to the concentration of each node, a dual-parameter dynamic coupling gradient field traceability algorithm of fusion / gas expansion and chemical reaction mechanism is adopted to calculate a fire source coordinate; performing anomaly detection according to the concentration of each node, and positioning an air leakage point by adopting a three-dimensional oxygen concentration gradient field positioning algorithm and combining air pressure compensation for an oxygen concentration anomaly point; based on goaf point cloud data, a goaf three-dimensional grid model is constructed, and a fire source spreading path, an air leakage channel and gas concentration distribution are dynamically and visually displayed in combination with multi-source monitoring data and fire source and air leakage point positioning results. The goaf fire prevention and control system can achieve all-around prevention and control of goaf fire.

Description

Technical Field

[0001] This invention relates to the field of coal mine goaf fire monitoring technology, and in particular to a coal mine goaf fire monitoring method and system based on multi-source data fusion. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] A coal mine goaf, also known as an "old pit" or "mined area," refers to a cavity or subsidence area formed after coal mining, when the overlying strata previously supported by coal lose their support and collapse, fracture, or subside. Goafs are characterized by their high degree of concealment, complex structure, tendency to accumulate harmful gases such as methane, and high risk of spontaneous combustion of residual coal. They can also trigger roof collapses, surface subsidence, and water inrush accidents, making them a key area for safety management in coal mines. Spontaneous combustion in goafs is one of the major safety hazards in coal mines. Existing regulations and detailed rules, such as the "Coal Mine Safety Regulations" and "Coal Mine Fire Prevention and Extinguishing Rules," clearly stipulate requirements for fire monitoring in goafs, emphasizing the need to establish a comprehensive monitoring and early warning system to achieve early identification and precise control of hidden fire sources through real-time data collection and analysis.

[0004] Currently, fire monitoring in goaf areas has the following significant characteristics: the fire source is concealed, making it difficult to pinpoint the exact location; the mine uses negative pressure ventilation, which easily allows toxic and harmful gases to enter the roadways, making the fire difficult to control once it spreads, and potentially leading to a gas explosion, causing significant casualties and economic losses. Therefore, timely and accurate detection of fire hazards in goaf areas and prompt fire control are important research directions for mine safety production. However, due to the special circumstances of goaf areas, many current monitoring technologies and methods are difficult to implement there, such as: (1) Bundle tube monitoring system: The principle is to lay bundle tubes in the goaf to extract gas and send the gas to a gas analysis device for gas composition analysis. The gas index is used to determine whether an fire has occurred. However, this method has a high rate of physical damage. Periodic pressure on the roof of the goaf will induce rockfall and collapse. After being impacted, the bundle tubes will be crushed (when the tube diameter deformation is >50%, the gas flow rate drops to less than 30% of the normal value), the joints will come loose (leading to gas leakage and failure), and rock powder will block the tubes (requiring manual maintenance). The failure rate of monitoring points due to damage is usually high. The gas in the deep goaf cannot be extracted, and it cannot accurately and effectively monitor the fire situation in the goaf.

[0005] (2) The principle of the distributed fiber optic temperature measurement system is to lay temperature-sensing optical cables along the goaf and locate the temperature point through the time-domain reflection of the optical signal. However, this method relies solely on temperature as a single parameter for monitoring, and since gas itself is not a good heat-conducting medium, simply relying on temperature to locate the fire source will lead to a large error. In addition, the movement of the roof rock layer will squeeze the optical cable, resulting in micro-bending loss of the optical cable (when the radius of curvature of the optical fiber is <5cm, the scattered signal will be misidentified as a temperature jump >30℃), local stress concentration (the optical cable is subjected to shear force and produces a chirping effect, and the system misjudges the fire point coordinates to be offset by more than 20m), and environmental interference (there are false alarms from non-fire heat sources such as blasting vibration and equipment heat dissipation). Therefore, it cannot accurately and effectively monitor the fire situation in the goaf.

[0006] (3) Lack of goaf air leakage monitoring technology. Most safety hazards in goaf areas are caused by air leakage. Air leakage in goaf areas will disrupt the balance of the ventilation system, leading to local gas accumulation and providing oxygen to fuel combustion (for example, when the oxygen concentration is >8%, the probability of spontaneous combustion increases by 6 times). At present, goaf fire monitoring is only carried out manually through tracer gas detection, which is time-consuming, labor-intensive, and ineffective, and does not take into account the impact of air leakage on fire.

[0007] (4) Monitoring data cannot intuitively reflect the complex situation inside the goaf. Currently, fire monitoring and management in goaf areas mostly rely on two-dimensional drawings (such as CAD plan and geological profile), which cannot intuitively display the complex three-dimensional spatial structure of the goaf, cannot be linked to the dynamic monitoring of fire in real time, and have poor interactivity. Summary of the Invention

[0008] To address the shortcomings of the existing technologies, this invention provides a method and system for monitoring coal mine goaf fires based on multi-source data fusion, through the design of fusion... / A dual-parameter dynamic coupled gradient field tracing algorithm based on gas diffusion and chemical reaction mechanisms is used to accurately locate the fire source. A goaf leakage analysis algorithm based on gas concentration anomalies combined with spatial concentration and pressure gradient analysis is designed to accurately locate the leakage point. Then, combined with a three-dimensional model of the goaf, the fire situation is visualized and the automatic fire extinguishing trigger is achieved. Through a closed-loop prevention and control method of goaf fire perception, fire point location, leakage detection, leakage point location, and fire handling, comprehensive prevention and control of goaf fires can be achieved.

[0009] In a first aspect, the present invention provides a method for monitoring fires in coal mine goaf areas based on multi-source data fusion.

[0010] A method for monitoring coal mine goaf fires based on multi-source data fusion, comprising: Based on wireless sensor nodes uniformly deployed in the coal mine goaf, real-time data is collected from each node. concentration, Concentration and air pressure; According to each node concentration, Concentration, using fusion / A dual-parameter dynamic coupling gradient field source tracing algorithm based on gas expansion and chemical reaction mechanisms was used to calculate the coordinates of the fire source. According to each node Anomaly detection is performed on the concentration. For points with abnormal oxygen concentration, a three-dimensional oxygen concentration gradient field localization algorithm combined with air pressure compensation is used to locate the air leakage point. Based on point cloud data of the goaf, a three-dimensional grid model of the goaf is constructed. Combined with multi-source monitoring data and the location results of fire sources and air leakage points, the fire spread path, air leakage channels and gas concentration distribution are dynamically and visually displayed.

[0011] Further technical solutions also include: Based on the fire source location results, the fire extinguishing capsule closest to the fire source is activated; among them, several fire extinguishing capsules are evenly deployed in the coal mine goaf, and each fire extinguishing capsule is equipped with a temperature sensor and a micro controller. After the fire extinguishing capsule is activated, it releases the internal nanogel to form a suffocation layer to control the fire through self-triggering or wireless triggering. The self-triggering method involves real-time monitoring of the ambient temperature, and automatic release when the ambient temperature exceeds the set value. The wireless triggering method involves release when a remote control signal is received.

[0012] Further technical solutions will be developed based on the characteristics of each node. concentration, Concentration, using fusion / A dual-parameter dynamically coupled gradient field source tracing algorithm based on gas expansion and chemical reaction mechanisms calculates the fire source coordinates as follows: According to each node concentration, The concentration is calculated using a dual-parameter dynamic coupled gradient field calculation formula to calculate the gradient of each node; wherein, the gradient is a three-dimensional vector, the gradient direction represents the fire source propagation direction at the current spatial location, and the gradient magnitude represents the current fire intensity; The source of the fire is traced back in reverse according to the gradient direction of each node until the node whose gradient magnitude is a local maximum is determined. This node is the fire source node, and the fire source location is completed.

[0013] A further technical solution is that the calculation formula for the dual-parameter dynamically coupled gradient field is as follows: ; in, Indicates carbon monoxide concentration. Indicates oxygen concentration. Represents the traditional concentration gradient. This is the air intake direction vector. Represents protection against zero disturbance term, 、 、 For dynamic weights, it is represented as: ; ; .

[0014] Further technical solutions will be developed based on the characteristics of each node. Anomaly detection is performed on the concentration. For points of abnormal oxygen concentration, a three-dimensional oxygen concentration gradient field localization algorithm combined with air pressure compensation is used to locate the air leak point, which is as follows: According to each node Concentration anomaly detection and screening. Abnormal sensing nodes with concentrations exceeding the set value; For abnormal sensor nodes, according to The concentration and pressure are used to calculate the gradient of the node using the formula for the three-dimensional oxygen concentration gradient field. The leakage point is located by identifying the node whose gradient direction is consistent with the air pressure gradient direction and whose gradient magnitude is a local maximum.

[0015] A further technical solution is that the calculation formula for the three-dimensional oxygen concentration gradient field is: ; in, Indicates oxygen concentration. Represents the traditional concentration gradient. 、 、 For direction weights, This represents the air pressure compensation coefficient. Represents the air pressure gradient vector. This is a pressure-time coupling term. is the coupling coefficient.

[0016] Secondly, this invention provides a coal mine goaf fire monitoring system based on multi-source data fusion. A coal mine goaf fire monitoring system based on multi-source data fusion includes: The data acquisition module includes wireless sensor nodes evenly deployed in the coal mine goaf area, used to collect data from each node in real time. concentration, Concentration and air pressure; The data processing module is used to process data from each node. concentration, Concentration, using fusion / A dual-parameter dynamically coupled gradient field source tracing algorithm based on gas expansion and chemical reaction mechanisms is used to calculate the coordinates of the fire source; based on the nodes... Anomaly detection is performed on the concentration. For points with abnormal oxygen concentration, a three-dimensional oxygen concentration gradient field localization algorithm combined with air pressure compensation is used to locate the air leakage point. The visualization and interaction module is used to construct a three-dimensional grid model of the goaf based on the point cloud data of the goaf. Combined with multi-source monitoring data and the location results of fire sources and air leakage points, it dynamically and visually displays the fire source spread path, air leakage channels and gas concentration distribution.

[0017] Thirdly, the present invention also provides an electronic device, comprising: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the above-mentioned method for monitoring coal mine goaf fires based on multi-source data fusion.

[0018] Fourthly, the present invention also provides a computer-readable storage medium storing executable instructions for causing a processor to execute the executable instructions to implement the above-described method for monitoring coal mine goaf fires based on multi-source data fusion.

[0019] Fifthly, the present invention also provides a computer program product, which includes executable instructions stored in a computer-readable storage medium; wherein, when the processor of the electronic device reads the executable instructions from the computer-readable storage medium and executes the executable instructions, the above-mentioned method for monitoring coal mine goaf fires based on multi-source data fusion is implemented.

[0020] The above one or more technical solutions have the following beneficial effects: 1. This invention proposes a method and system for monitoring coal mine goaf fires based on multi-source data fusion, and designs a dual-parameter dynamic coupled gradient field source tracing algorithm, which integrates... / Chemical reaction mechanism (CO generation and By incorporating the coupling relationship of consumption and a gas diffusion model, and introducing dynamic weights, the system can "precisely locate" hidden fire sources in goaf areas, providing accurate coordinates for subsequent firefighting operations and avoiding the waste of firefighting resources or delays in prevention and control caused by inaccurate positioning in traditional technologies. Considering that air leakage is the core cause of spontaneous combustion in goaf areas, an air leakage analysis algorithm based on abnormal gas concentration combined with spatial concentration and pressure gradient analysis was also designed. This algorithm can achieve intelligent and automated detection of air leakage in goaf areas, with the accuracy of air leakage point positioning coordinated with fire source positioning. It can identify spontaneous combustion risk areas (such as areas with abnormally high oxygen concentration) in advance, reducing fire hazards from the root.

[0021] 2. In this invention, a dual-parameter dynamically coupled gradient field source tracing algorithm is used to integrate the CO concentration used to reflect the generation of the fire source and the oxygen consumption. By combining the chemical reaction mechanism and gas diffusion law between concentrations with dynamic weights and pressure compensation correction, more accurate fire source location and dynamic tracking are achieved. Simultaneously, a three-dimensional oxygen concentration gradient field localization algorithm with a combined pressure-time coupling term is used to screen suspicious nodes based on oxygen concentration anomaly detection. The leak point is located by combining the pressure gradient direction (air leakage causing pressure reduction) and the maximum gradient modulus, achieving precise location of the leak point. Finally, by combining goaf point cloud data and multi-source monitoring data, a dynamic three-dimensional mesh model is constructed. This model dynamically displays the fire source spread path, air leakage channels, and gas concentration distribution, effectively improving the intelligence and precision of fire prevention and control in coal mine goaf areas.

[0022] 3. In this invention, a three-dimensional model of the goaf is also used to visualize the fire situation and trigger automated fire extinguishing. Abstract monitoring data is transformed into a three-dimensional dynamic scene that can be intuitively perceived, providing managers with immersive decision support. It can simulate scenarios such as fire spread and air leakage evolution, and formulate prevention and control strategies in advance.

[0023] 4. This invention achieves comprehensive prevention and control of goaf fires through a closed-loop prevention and control method that includes goaf fire detection, fire point location, air leakage detection, air leakage point location, and fire handling, thereby effectively improving emergency response efficiency.

[0024] Advantages of additional aspects of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0025] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0026] Figure 1This is an overall flowchart of the coal mine goaf fire monitoring method based on multi-source data fusion as described in this embodiment of the invention; Figure 2 This is a schematic diagram of the deployment of wireless sensor nodes in the goaf area in an embodiment of the present invention; Figure 3 This is a flowchart of the air leakage analysis and location in the goaf in an embodiment of the present invention; Figure 4 This is a flowchart illustrating the activation and triggering of the fire extinguishing capsule in an embodiment of the present invention. Detailed Implementation

[0027] It should be noted that the following detailed descriptions are exemplary and are intended only to describe specific embodiments and to provide further explanation of the invention, and are not intended to limit the scope of exemplary embodiments of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0028] Example 1 This embodiment provides a method for monitoring coal mine goaf fires based on multi-source data fusion, breaking through the limitations of traditional two-dimensional management of goaf fires and realizing multi-source data fusion and dynamic fire monitoring. Figure 1 As shown in this embodiment, the method for monitoring fires in coal mine goaf areas specifically includes the following steps: Step S1: Based on wireless sensor nodes uniformly deployed in the coal mine goaf, collect data from each node in real time. concentration, Concentration and air pressure.

[0029] In this embodiment, as Figure 2 As shown in the diagram (green icons indicate online devices, gray icons indicate offline devices), several LoRa wireless sensor nodes are first evenly deployed in the coal mine goaf area to form a sensor network. This wireless sensor network is then used to collect data from each node in real time. concentration, The system also includes a concentration sensor and a MEMS pressure sensor to collect real-time air pressure data at various nodes within the goaf, monitoring pressure gradient changes (accuracy ±0.1 kPa). Additionally, a fiber optic grating sensor can be installed to monitor roof strata movement (resolution 0.1 mm), facilitating the construction and optimization of a subsequent 3D model of the goaf. Preferably, a temperature sensor can also be installed to collect real-time temperature data at various nodes within the goaf. The data collected in real-time by these sensors can be wirelessly transmitted to a cloud server or processor for subsequent data analysis and processing.

[0030] As one implementation method, after on-site installation, the installation is analyzed and modified according to the specific situation. For example, the suspension height is increased from 1.5 meters to about 3 meters; additional fixing measures are added to the internal antenna interface of the equipment to ensure that it will not fall off under severe vibration; and the impact of not adding a protective cover on the signal is tested. After the modifications, the stability of the air intake and return air equipment can be further improved.

[0031] Preferably, for this data transmission, this embodiment uses the LoRaWAN protocol to achieve low-power wide-area coverage, with a single base station coverage radius of ≥3km and support for 500 concurrent access nodes; in addition, data is encrypted during transmission (AES-128) and tamper-proofed by verification (SHA-256) to ensure the security of data transmission.

[0032] Furthermore, after acquiring the aforementioned multi-source monitoring data, a data preprocessing process is also included, namely: cleaning the collected data, eliminating sensor noise such as CO concentration fluctuations caused by dust interference through Kalman filtering; and then filling in missing data using the sliding window mean method, the window size of which is adjustable and can be set to a maximum of 5 minutes.

[0033] Step S2: Based on each node's concentration, Concentration, using fusion / A dual-parameter dynamic coupling gradient field source tracing algorithm based on gas expansion and chemical reaction mechanisms was used to calculate the coordinates of the fire source.

[0034] Specifically, for locating ignition sources in coal mine goaf areas, compared to traditional gradient fields that use a single parameter (such as only based on...),... (Concentration) Generates a concentration gradient vector field: This embodiment proposes a dual-parameter dynamic coupled gradient field source tracing algorithm for fire source localization. Based on the actual conditions of coal mine goaf areas, it introduces oxygen and carbon monoxide as dual parameters, integrating the algorithm with gas diffusion in the goaf area and... Generation and The chemical reaction coupling mechanism is deeply coupled, and the coupling characteristics are integrated into the spatial gradient calculation. The fire situation is accurately located and predicted by coupling gradient field, thereby realizing high-precision fire source location in coal mine goaf.

[0035] In this embodiment, firstly, based on the nodes... concentration, The concentration is calculated using a two-parameter dynamically coupled gradient field formula to determine the gradient at each node. The calculated gradient is a three-dimensional vector (including direction and magnitude). The gradient direction represents the direction of fire source propagation at the current spatial location (arrows point to the fire source core), and the gradient magnitude represents the current fire intensity (the larger the magnitude, the closer to the fire source). The formula for calculating this two-parameter dynamically coupled gradient field is: ; in, express Growth direction express Attenuation direction, For coupling correction term, This indicates the concentration of carbon monoxide (ppm, parts per million). Indicates oxygen concentration (%); Represents a traditional concentration gradient; This is the air intake direction vector, which is manually input, and each goaf has a clearly defined air intake direction; Represents prevention of zero disturbance items; 、 、 For dynamic weights, it is represented as: Among them, the faster the change in CO, the higher the weight. ;in, The weight increases significantly when the price drops sharply; Coupling is enhanced under hypoxic conditions.

[0036] Secondly, the source is traced back in reverse according to the gradient direction of each node until the node whose gradient magnitude is a local maximum is determined. This node is the source node, and the source location is completed.

[0037] Using the above algorithm, through dynamic weights ( 、 、 Coupled with CO concentration gradient (reflecting ignition source generation characteristics), Concentration gradient (reflecting oxygen consumption characteristics) and inlet air direction vector (Reflecting the dominant direction of gas diffusion) to locate the fire source.

[0038] As one implementation method, considering the calculation of the aforementioned dual-parameter dynamically coupled gradient field, only the inlet air direction vector is used. The ventilation status was indirectly related, but the pressure gradient inside the goaf was not taken into account. The influence of air pressure, a core physical factor affecting gas diffusion rate and concentration gradient distribution (e.g., the greater the pressure difference, the faster the gas diffusion rate, and the closer the peak of the concentration gradient is to the leak point), leads to insufficient adaptability of the above calculation method to complex ventilation environments. Therefore, this embodiment also introduces air pressure data within the goaf area; specifically, the air pressure gradient within the goaf... Directly affects the direction and rate of gas diffusion: when When the direction is consistent with the air intake direction, the gas diffusion rate increases, and the peak concentration gradient shifts towards the direction of decreasing air pressure; while when When the direction of airflow is opposite to the direction of air intake (such as when air pressure is abnormal due to localized air leakage), "reverse diffusion" will occur. Relying solely on the air intake direction vector will lead to gradient direction deviation. Therefore, this embodiment introduces an air pressure compensation term. It can correct the direction and magnitude of the concentration gradient, making the gradient field more consistent with the actual gas diffusion law.

[0039] The pressure compensation coefficient δ is defined as follows: ; In the above formula, The pressure gradient modulus is expressed in kPa / m. The average air pressure in the goaf is expressed in kPa; the value of δ ranges from 0.1 to 0.5 to ensure that the compensation term does not excessively interfere with the original gradient.

[0040] The corrected concentration gradient can then be expressed as: ; ; In the above formula, sign is the sign function. When the concentration gradient is in the same direction as the pressure gradient, sign=1, which enhances the gradient; otherwise, sign=-1, which corrects the effect of reverse diffusion.

[0041] Furthermore, the revised formula for calculating the two-parameter dynamically coupled gradient field is as follows: ; The above methods can adapt to complex ventilation environments, such as abnormal air pressure caused by local air leakage in the goaf (i.e., (sudden increase), through The correction ensures the gradient direction still points to the fire source, avoiding positioning offsets caused by air leakage. Compared to the unimproved calculation method, which resulted in over 20m offsets due to misjudging fire point coordinates caused by air leakage, the improved method controls the offset within 3m. Furthermore, it enables the inversion of the air leakage-fire source correlation, i.e., through... The coupling relationship with the concentration gradient can indirectly determine the degree of impact of air leakage on the fire source (e.g., the larger δ is, the stronger the air leakage and the higher the risk of fire spread), providing a basis for subsequent fire extinguishing decisions.

[0042] As another implementation, considering that the calculation of the above-mentioned two-parameter dynamically coupled gradient field is based only on a certain moment ( The gradient is calculated using static concentration data, which does not take into account... The decay pattern of CO concentration over time, such as the natural decay caused by gas diffusion in the goaf and the rate of increase in concentration caused by changes in the intensity of combustion from ignition sources, is prone to positioning errors due to instantaneous data fluctuations. Therefore, in this embodiment, the analysis of CO concentration changes in the goaf over time includes both "ignition source generation increment" and "natural diffusion decay," requiring the use of a time decay coefficient to further correct the static concentration gradient and eliminate concentration fluctuations caused by non-ignition source factors, such as temporary increases in CO concentration due to temporary ventilation disturbances. Specifically, a time decay factor τ is introduced, based on... Time series data of concentration, calculate the rate of change of concentration ( The decay trend of ) is used to correct the dynamic weights. 、 This ensures that the weights better reflect the continuous combustion characteristics of the fire source rather than instantaneous interference. The time decay factor is defined as: ; In the above formula, The data collection time interval is matched with the sliding window mean method mentioned above; k is the attenuation coefficient, which can be dynamically adjusted according to the permeability of the goaf. The higher the permeability, the larger k is, and the faster the attenuation. Its value range is 0.01-0.1.

[0043] The corrected dynamic weights can then be expressed as: ; ; The above methods can eliminate transient interference, such as a temporary increase in CO concentration caused by blasting vibrations. / Sudden increase but Small, α will not increase abnormally), to avoid misjudging the location of the fire source. In addition, if the fire source weakens ( / decline, (increase) 、 Synchronous adjustments ensure that the gradient direction always points to the core of the fire source, improving the accuracy of dynamic tracking and positioning of the fire source.

[0044] Step S3: Based on each node's... Anomaly detection is performed on the concentration. For points with abnormal oxygen concentration, a three-dimensional oxygen concentration gradient field localization algorithm combined with air pressure compensation is used to locate the air leakage point.

[0045] Specifically, such as Figure 3 As shown, firstly, based on a distributed sensor network, according to the data monitored by each node... Concentration abnormalities were detected (normal concentration should be <12%), and screening was conducted. Abnormal sensor nodes with concentrations exceeding a set value; secondly, considering that the oxygen concentration near the air leak point will form an "oxygen chimney effect," exhibiting a significant gradient change, this embodiment also targets the selected abnormal sensor nodes according to... For both concentration and pressure, the gradient at this node is calculated using the formula for a three-dimensional oxygen concentration gradient field, with the gradient direction aligned with the pressure gradient direction and the gradient magnitude being a local maximum (i.e., ...). The nodes are identified as air leakage points, enabling precise location of these points.

[0046] Furthermore, the formula for calculating the above three-dimensional oxygen concentration gradient field is as follows: ; in, Indicates oxygen concentration, used for locating air leaks; Represents a traditional concentration gradient; 、 、 The directional weights are used to adjust the contribution of concentration gradients in different spatial directions to the final result. This represents the pressure compensation coefficient, which is dynamically optimized in the range of 0.8-1.2. This represents the pressure gradient vector, used to compensate for the effect of air pressure on oxygen concentration distribution; This is a pressure-time coupling term. The coupling coefficient reflects the acceleration of air pressure changes. When the air pressure changes from positive to negative, the second derivative changes from positive to negative. The coupling term adjusts the weight of the oxygen concentration gradient calculation simultaneously to ensure that the gradient direction is always consistent with the air pressure trend caused by the air leakage, thereby improving the accuracy of the air leakage point location.

[0047] Furthermore, 、 、 The weights are directional and can be adaptively adjusted based on the fracture topology. For example, if a certain direction (such as the x-direction) is the main fracture orientation, the fracture distribution can be obtained through geological exploration data or LiDAR point cloud modeling. Considering that air leakage is more likely to propagate along the main fractures, the concentration gradient in this direction can better reflect the true path of air leakage. Therefore, the weight of this direction is adjusted accordingly. Increase, for example =0.5, to enhance the contribution of the oxygen concentration gradient in this direction to the location of air leakage; correspondingly, if there are fewer fractures in a certain direction (such as the z-direction), such as a small vertical height of the goaf and good rock strata integrity, then the weight of this direction is increased. Decrease, for example =0.1, to avoid the concentration gradient in the secondary direction interfering with the location of the air leak.

[0048] In the above process, considering the breathing effect in the goaf and the periodic switching between positive and negative pressure, the air pressure level cannot be used as a basis for judging air leakage. However, air pressure changes can lead to inaccurate judgments, and stable air pressure is a prerequisite. Data during the switching period needs to be ignored. Therefore, this embodiment introduces air pressure field verification to confirm that the gradient direction is consistent with the air pressure decreasing trend, thereby accurately locating the air leakage point. Verification shows that this air pressure compensation mechanism can effectively suppress false alarms caused by ventilation disturbances, with a false alarm rate of <0.3% / month, while the false alarm rate of traditional systems is 22% / month.

[0049] By deeply integrating the fire source location algorithm (dual-parameter gradient field coupling) and the air leakage analysis algorithm (oxygen concentration gradient-pressure compensation model) described above, scientific decision support can be provided for coal mine fire early warning, air leakage control, and emergency rescue. This embodiment simultaneously considers the coupling effect of combustion dynamics (CO generation) and fluid dynamics (air leakage diffusion), achieving a fire source location error of <1.5m. Field measurements have verified that, compared to the 15-20m location error of traditional bundled tube systems, this embodiment achieves higher accuracy in fire source location.

[0050] Step S4: Based on the point cloud data of the goaf, construct a three-dimensional grid model of the goaf, and combine multi-source monitoring data with the location results of fire source and air leakage point to dynamically and visually display the fire source spread path, air leakage channel and gas concentration distribution.

[0051] Specifically, in order to visualize the monitored fire situation, this embodiment first uses a LiDAR scanner to acquire point cloud data of the goaf (accuracy ±2cm), and then uses existing modeling methods or software to construct a three-dimensional mesh model of the goaf based on the point cloud data of the goaf.

[0052] Preferably, considering that the roof of the goaf may be subjected to periodic pressure and move and deform, which may cause the simulated mesh model to deviate from the actual situation, this embodiment also updates the subsidence of the goaf roof in real time based on displacement monitoring data (update frequency ≤ 1 time / hour) to optimize the three-dimensional mesh model. At the same time, geological exploration data (fault, joint distribution, etc.) are also imported and combined with FLAC3D software to simulate the movement and deformation of rock strata, so as to ensure the accuracy of the subsequent model display.

[0053] Furthermore, based on the aforementioned 3D mesh model of the goaf, and combined with multi-source monitoring data and the location results of fire sources and air leaks, the fire spread path, air leak channels, and gas concentration distribution are dynamically visualized, and the displayed content is rendered in 3D. Preferably, this 3D rendering engine is developed based on Unity 3D, supports WebGL deployment, and has a rendering frame rate of ≥60fps, thereby rendering disaster scenes such as flame effects (based on particle systems), smoke diffusion (volume fog technology), and rock fracture animation.

[0054] In this embodiment, dynamic visualization functionality can be implemented, including: (1) Fire spread path: The arrows indicate the direction of the oxygen-carbon monoxide concentration coupling gradient. Different colors indicate the fire intensity or temperature. For example, red indicates a temperature > 800℃, yellow indicates a temperature of 400-800℃, and green indicates a temperature < 400℃.

[0055] (2) Air leakage channel marking: Blue streamlined display Along the concentration gradient, flow rate and leakage are positively correlated, with flow rate > 1 m / s marked as high risk.

[0056] (3) Gas concentration thermogram: Overlay a semi-transparent color level layer and display in real time. , Concentration distribution. For the color scale, red indicates >200 ppm, yellow indicates 50-200 ppm, and blue indicates <50 ppm.

[0057] Using the above methods, the dynamic evolution of fire-air leakage in coal mine goaf areas can be carried out based on multi-source data fusion and numerical simulation technology. Real-time dynamic simulation of fire source spread path, air leakage channel distribution and gas concentration changes can be achieved through three-dimensional visualization technology.

[0058] As one implementation method, such as Figure 4 As shown, in step S5, based on the fire source location result, the fire extinguishing capsule closest to the fire source is activated. After activation, the fire extinguishing capsule releases its internal nanogel through self-triggering or wireless triggering to form an asphyxiation layer to control the fire. The outer shell of the fire extinguishing capsule is made of a low-temperature brittle material (self-shattering at -10℃), releasing nanogel to cover the fire source. The gel expands 50 times upon contact with air, forming an asphyxiation layer (oxygen concentration <5%), thus extinguishing the fire.

[0059] Preferably, the composite shell material of the fire extinguishing capsule can be: a composite of polylactic acid (PLA) and nano-hydroxyapatite (nHA) in a mass ratio of 9:1, with a tensile strength ≤1MPa at 80℃ and a compressive strength ≥15MPa at room temperature; the fire extinguishing gel formula of the fire extinguishing capsule can be: containing sodium percarbonate, ammonium nitrate and a nano-SiO2 coating layer, wherein the mass ratio of sodium percarbonate to ammonium nitrate is 6:3, the nano-SiO2 coating layer accounts for 10%, and the volume expansion rate after exposure to air is ≥5000%.

[0060] Specifically, in this embodiment, when deploying the wireless sensing device, a suspension point is added to suspend the capsule below the wireless sensing device. As a result, several fire extinguishing capsules are evenly deployed in the coal mine goaf, and each fire extinguishing capsule is equipped with a temperature sensor and a micro controller.

[0061] In addition, the fire extinguishing capsule has two triggering methods: a self-triggering method, which monitors the ambient temperature in real time and automatically triggers release when the ambient temperature exceeds a set value (e.g., >80℃); and an unlimited triggering method, which triggers release when a remote control signal is received (response time <1 second). This dual triggering design ensures the reliability and timeliness of fire extinguishing.

[0062] Example 2 This embodiment provides a coal mine goaf fire monitoring system based on multi-source data fusion, including: The data acquisition module includes wireless sensor nodes evenly deployed in the coal mine goaf area, used to collect data from each node in real time. concentration, Concentration and air pressure; The data processing module is used to process data from each node. concentration, Concentration, using fusion / A dual-parameter dynamically coupled gradient field source tracing algorithm based on gas expansion and chemical reaction mechanisms is used to calculate the coordinates of the fire source; based on the nodes... Anomaly detection is performed on the concentration. For points with abnormal oxygen concentration, a three-dimensional oxygen concentration gradient field localization algorithm combined with air pressure compensation is used to locate the air leakage point. The visualization and interaction module is used to construct a three-dimensional grid model of the goaf based on the point cloud data of the goaf. Combined with multi-source monitoring data and the location results of fire sources and air leakage points, it dynamically and visually displays the fire source spread path, air leakage channels and gas concentration distribution.

[0063] Furthermore, it also includes a fire control module. This fire control module comprises several fire extinguishing capsules evenly deployed in the coal mine goaf. Based on the fire source location, it activates the fire extinguishing capsule closest to the fire source. After activation, the fire extinguishing capsule releases its internal nanogel to form an asphyxiation layer to control the fire, either automatically or wirelessly. Each fire extinguishing capsule has a built-in temperature sensor and microcontroller. The automatic triggering method involves real-time monitoring of the ambient temperature; when the ambient temperature exceeds a set value, it automatically triggers release. The wireless triggering method involves triggering release upon receiving a remote control signal.

[0064] Example 3 This embodiment provides an electronic device, including: a memory for storing executable instructions; and a processor for executing the executable instructions stored in the memory to implement the method provided in this embodiment.

[0065] Example 4 This embodiment also provides a computer-readable storage medium storing executable instructions, which, when executed by a processor, will cause the processor to execute the method described above in this embodiment.

[0066] Example 5 This embodiment provides a computer program product including executable instructions, which are computer instructions; the executable instructions are stored in a computer-readable storage medium. When the processor of an electronic device reads the executable instructions from the computer-readable storage medium and executes the executable instructions, the electronic device performs the method described in this embodiment.

[0067] The steps and methods involved in Embodiments 2 to 5 above correspond to those in Embodiment 1. For specific implementation details, please refer to the relevant description section of Embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood as including any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and enabling the processor to perform any of the methods in this invention.

[0068] Those skilled in the art will understand that the modules or steps of the present invention described above can be implemented using general-purpose computer devices. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage device for execution by a computer device, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. The present invention is not limited to any particular combination of hardware and software.

[0069] The above description is only a preferred embodiment of the present invention. Although the specific implementation of the present invention has been described in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that, based on the technical solution of the present invention, various modifications or variations that can be made by those skilled in the art without creative effort are still within the scope of protection of the present invention.

Claims

1. A coal mine goaf fire monitoring method based on multi-source data fusion, characterized in that, The method comprises the following steps: Based on the uniform deployment of wireless sensor nodes in coal mine goaf, real-time collection of each node's concentration, concentration and air pressure; According to the nodes of Concentration, Concentration, using fusion / Gas expansion and chemical reaction mechanism of two-parameter dynamic coupling gradient field source algorithm, the fire source coordinates are calculated; According to the concentration of each node The abnormality detection is performed according to the concentration of each node, for the oxygen concentration abnormal point, a three-dimensional oxygen concentration gradient field positioning algorithm is adopted and combined with air pressure compensation to position the air leakage point. Based on the point cloud data of the goaf, a three-dimensional grid model of the goaf is constructed, and the fire spread path, the air leakage channel and the gas concentration distribution are dynamically visualized and displayed in combination with the multi-source monitoring data and the fire source and air leakage point positioning results.

2. The coal mine goaf fire monitoring method based on multi-source data fusion according to claim 1, characterized in that, The method further comprises the following steps: Based on the fire source positioning result, the fire extinguishing capsule closest to the fire source is started, wherein a plurality of fire extinguishing capsules are uniformly arranged in the goaf of the coal mine, and each fire extinguishing capsule is internally provided with a temperature sensor and a microcontroller; After the fire extinguishing capsule is started, the internal nanogel is triggered to release to form a suffocation layer to control the fire through a self-triggering or wireless triggering mode; wherein the self-triggering mode is to monitor the environmental temperature in real time, and when it is monitored that the environmental temperature exceeds a set value, the release is automatically triggered; the wireless triggering mode is to trigger the release when a remote control signal is received.

3. The coal mine goaf fire monitoring method based on multi-source data fusion according to claim 1, characterized in that, According to the nodes of Concentration, Concentration, using fusion / Gas expansion and chemical reaction mechanism of two-parameter dynamic coupling gradient field source algorithm, the fire source coordinates are calculated as: According to the concentration of each node Concentration, Concentration, the gradient of each node is calculated by using the calculation formula of the dynamic coupling gradient field of two parameters; wherein, the gradient is a three-dimensional vector, the gradient direction represents the fire source propagation direction of the current space position, and the gradient module length represents the current fire intensity; The gradient direction of each node is reversely traced until the node with the local maximum gradient module length is determined, and the node is the fire source node, thereby completing the fire source positioning.

4. The coal mine goaf fire monitoring method based on multi-source data fusion according to claim 3, characterized in that, The calculation formula of the dual-parameter dynamic coupling gradient field is ； wherein, denotes the carbon monoxide concentration, denotes the oxygen concentration, denotes the conventional concentration gradient, is the wind direction vector, denotes the prevention of zero disturbance terms, 、 、 is the dynamic weight, denoted as: ； ； 。 5. The coal mine goaf fire monitoring method based on multi-source data fusion according to claim 1, characterized in that, According to the concentration of each node For the oxygen concentration anomaly point, the three-dimensional oxygen concentration gradient field positioning algorithm is adopted combined with air pressure compensation to locate the air leakage point. Anomaly detection is performed on the concentrations of each node and the abnormal sensor nodes with concentrations greater than a set value are screened out and the abnormal sensor nodes with concentrations greater than a set value are screened out For the abnormal sensor node, according to Concentration and air pressure, using the calculation formula of three-dimensional oxygen concentration gradient field, the gradient of the node is calculated; The node with the gradient direction consistent with the air pressure gradient direction and the local maximum gradient module length is taken as the air leakage point, thereby completing the air leakage point positioning.

6. The coal mine goaf fire monitoring method based on multi-source data fusion according to claim 5, characterized in that, The calculation formula of the three-dimensional oxygen concentration gradient field is: ； wherein denotes the oxygen concentration, denotes the conventional concentration gradient, 、 、 is the direction weight, denotes the barometric compensation factor, denotes the barometric gradient vector; is the barometric-time coupling term, denotes the coupling factor.

7. A coal mine goaf fire monitoring system based on multi-source data fusion, characterized in that, The method comprises the following steps: The data acquisition module comprises wireless sensor nodes uniformly arranged in the coal mine goaf, and is used for collecting the concentration, concentration and air pressure in real time data processing module, configured to perform anomaly detection according to the concentration of each node concentration, concentration, adopt a fusion / gas expansion and chemical reaction mechanism dual-parameter dynamic coupling gradient field tracing algorithm to calculate the fire source coordinates; perform anomaly detection according to the concentration of each node concentration, adopt a three-dimensional oxygen concentration gradient field positioning algorithm combined with air pressure compensation to locate the air leakage point; The visual interactive module is used to construct a three-dimensional grid model of the goaf based on the point cloud data of the goaf, and dynamically visualize and display the fire spread path, the air leakage channel and the gas concentration distribution in combination with the multi-source monitoring data and the fire source and air leakage point positioning results.

8. An electronic device, comprising: The method comprises the following steps: The memory is used to store executable instructions; The processor is used to execute the executable instructions stored in the memory, thereby implementing the coal mine goaf fire monitoring method based on multi-source data fusion according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The executable instructions are stored in the memory, and are used to cause the processor to execute the executable instructions, thereby implementing the coal mine goaf fire monitoring method based on multi-source data fusion according to any one of claims 1-6.

10. A computer program product, characterised in that, The computer program product comprises executable instructions stored in a computer readable storage medium; When the processor of the electronic device reads the executable instructions from the computer readable storage medium and executes the executable instructions, the coal mine goaf fire monitoring method based on multi-source data fusion according to any one of claims 1-6 is implemented.

Images