A coal mine fire early warning control method and system

Abstract

The present application relates to the technical field of coal mine safety monitoring and fire prevention, in particular to a coal mine fire early warning control method and system, the present application is based on D-S evidence theory, fuses multiple sensor data, based on the fusion result of D-S evidence theory and the coal type correction parameter, constructs a fire risk scoring model, judges the risk level according to the fire risk scoring model, adopts different control schemes according to different risk levels, through the collaborative operation of multi-dimensional gas monitoring network, intelligent early warning algorithm model and dynamic gas control strategy, the problem of coal mine fire early warning not timely and control not accurate in the prior art is solved, the fire discovery time is advanced to the initial gas anomaly stage, the fire accident rate is effectively reduced, and the safety production and personnel life and property safety of coal mine are ensured.

Description

Technical Field

[0001] This invention relates to the field of coal mine safety monitoring and fire prevention technology, specifically to a coal mine fire early warning and control method and system. Background Technology

[0002] Fire prevention and extinguishing are crucial aspects of comprehensive coal mine management, impacting mine property safety and the lives of all employees. Currently, coal mine fires are categorized into internal and external causes. The spontaneous combustion tendency of coal, coupled with the oxygen storage environment, is a key factor contributing to internal coal mine fires. Different coal types exhibit significant differences in spontaneous combustion tendency; lignite, due to its high volatile matter and low carbonization, has a much higher spontaneous combustion tendency than anthracite. Regarding the oxygen storage environment, residual coal in goaf areas is prone to low-temperature oxidation in the oxidation zone with an oxygen concentration of 15%-5%. During this process, carbon and hydrogen elements in the coal react chemically with oxygen to produce characteristic gases such as carbon monoxide (CO) and ethylene (C2H4). When the temperature threshold exceeds 60℃, the coal oxidation reaction rate increases dramatically, significantly raising the risk of spontaneous combustion. Studies show that under suitable oxygen and temperature conditions, the cycle from initial oxidation to spontaneous combustion of coal can be as short as several weeks, posing a serious threat to mine safety. Equipment failure, gas explosions, and management negligence are the main factors triggering external coal mine fires. Equipment malfunctions (such as conveyor belt friction and cable overheating) account for 32% of fires. During long-term operation, uneven tension and idler roller malfunctions can cause friction and heat generation between the conveyor belt and the frame or other components. When the heat accumulates to a certain level, it can ignite surrounding combustibles. Cable overheating is usually caused by overload or short circuits, which damage the cable insulation and lead to a fire. Gas explosions account for 25% of fires. When gas reaches a certain concentration in the mine, it will explode upon encountering an ignition source. The high temperature and shock wave generated by the explosion can easily ignite fires in the surrounding area. Management negligence leading to open flame ignition accounts for 18%. Illegal use of open flames and smoking underground can both become the trigger for fires. These multiple sources of fire hazards are intertwined, increasing the difficulty of coal mine fire prevention and control.

[0003] However, existing technologies for the protection and control of coal mine fires have certain shortcomings, mainly in the following aspects:

[0004] 1. Limitations of Single Sensors: Traditional electrochemical sensors are susceptible to interference from temperature and humidity, resulting in a false alarm rate of 15%-20%. In the high-humidity environment of mines, the electrolyte in the sensor is prone to dilution or leakage, leading to abnormal electrode reactions and thus false alarms. Ionization smoke sensors have a false alarm rate exceeding 10% in high-dust environments. High concentrations of dust in mines adsorb onto the sensor's sensing element, hindering the contact between smoke particles and the sensing element, preventing the sensor from detecting smoke in a timely manner, and making it unsuitable for the complex aerosol environment of mines.

[0005] 2. Early Warning Delay: Monitoring methods relying on manual inspections have a delay of over 30 minutes. Manual inspections are limited by factors such as inspection routes, time intervals, and personnel experience, making it difficult to detect subtle changes in the early stages of a fire in a timely manner. Existing bundled tube detection systems have long sampling cycles (10-15 minutes), making it difficult to capture sudden changes in initial gas concentrations in a timely manner. In the early stages of a fire, gas concentrations change rapidly, and a 10-15 minute sampling cycle may cause the best early warning opportunity to be missed, failing to buy valuable time for fire prevention and control.

[0006] 3. Inefficient control strategies: Traditional ventilation systems have a response time exceeding 5 minutes, making it impossible to effectively remove toxic and harmful gases and reduce temperature in a timely manner during a fire, leading to the spread of the fire. Grouting fire extinguishing technology has a slurry retention rate of only 60%-70%, failing to achieve precise fire suppression. Some slurry cannot reach the fire source, resulting in low fire extinguishing efficiency and potential secondary pollution to the mine environment. Summary of the Invention

[0007] To address the shortcomings of existing technologies, this invention proposes a coal mine fire early warning and control method and system, aiming to construct a coal mine fire prevention and control system integrating "real-time monitoring, intelligent early warning, and precise control." Through the coordinated operation of a multi-dimensional gas monitoring network, an intelligent early warning algorithm model, and a dynamic gas control strategy, it solves the problems of untimely early warning and inaccurate control in existing coal mine fire technologies. It enables early detection of fires at the initial gas anomaly stage, specifically when the carbon monoxide (CO) concentration is ≥5 ppm, while significantly improving fire extinguishing efficiency. Compared to traditional methods, fire extinguishing efficiency is increased by more than 40%, effectively reducing the incidence of fire accidents and ensuring safe coal mine production and the safety of personnel and property.

[0008] A method for early warning and control of coal mine fires includes the following steps:

[0009] Data on CO, C2H4 gas concentrations, methane concentrations, oxygen concentrations, temperature, humidity, pressure, and carbon monoxide concentrations in coal mine goafs, roadways, and electromechanical chambers; data on equipment temperature in electromechanical chambers; and data on smoke concentrations in goafs, roadways with conveyor belts, and electromechanical chambers.

[0010] Based on the DS evidence theory, a fire risk scoring model is constructed by integrating CO, C2H4 gas concentration data, methane concentration, oxygen concentration, temperature and humidity, pressure, carbon monoxide concentration data, equipment temperature data of electromechanical chambers, and smoke concentration data of goaf, roadway with conveyor belt, and electromechanical chamber. Based on the fusion results of the DS evidence theory and coal type correction parameters, a fire risk scoring model is constructed.

[0011] The risk level is determined based on the fire risk scoring model, and different control schemes are adopted according to different risk levels.

[0012] Furthermore, the expression for the fire risk scoring model is:

[0013] R = K×(0.4C + 0.3T + 0.2S + 0.1H);

[0014] In the formula, R is the risk coefficient; K is the coal type correction parameter, representing the coal type activity, with a value range of 0.8 to 1.2; C is the comprehensive index of the marker gas; T is the temperature anomaly coefficient; S is the smoke scattering rate; and H is the humidity correction parameter.

[0015] C = 0.6C1 + 0.4C2, where C1 is the standardized value of CO concentration; C2 is the standardized value of the rate of change of C2H4 concentration.

[0016] T = |T t – T1| / (T2 – T1), where T t To monitor the real-time temperature of the area, T1 is the reference value for normal underground temperature, and T2 is the critical value for fire warning temperature.

[0017] H = 1 - (RH 实 - RH 常 ) / (RH 临高 – RH 临低 ), where RH 实 The relative humidity (RH) is the measured value of the ambient humidity. 常 This is the baseline value for normal downhole humidity; RH 临高 The humidity warning threshold is RH. 临高 The set value is 95%RH; RH 临低 This is the critical low value for humidity warning, RH 临低 The set value is 30%RH;

[0018] S = S1 / S2, where S1 is the smoke concentration in the detection area and S2 is the baseline value of normal smoke concentration in the well.

[0019] Furthermore, based on the fire risk scoring model, the risk level is determined, and the specific methods for adopting different control schemes according to different risk levels are as follows:

[0020] When the coal type is highly reactive, R < 0.25 indicates a safe level, 0.25 ≤ R < 0.42 indicates a low-risk level, 0.42 ≤ R < 0.58 indicates a medium-risk level, and R ≥ 0.58 indicates a high-risk level.

[0021] When the coal type is medium-reactivity coal, R < 0.3 is the safety level, 0.3 ≤ R < 0.5 is the low-risk level, 0.5 ≤ R < 0.7 is the medium-risk level, and R ≥ 0.7 is the high-risk level.

[0022] When the coal type is low-activity coal, R < 0.38 is the safety level, 0.38 ≤ R < 0.63 is the low safety level, 0.63 ≤ R < 0.88 is the medium risk level, and R ≥ 0.88 is the high risk level.

[0023] When each area is at a safe level, all sensors in each area monitor normally. When an area is at a low safety level, an audible and visual alarm is activated in that area, and the sampling frequency of the corresponding sensors in that area is increased. When an area is at a medium risk level, an audible and visual alarm is activated in that area, a warning message is sent to the management personnel, the local smoke exhaust system is activated for ventilation, and the fire extinguishing equipment is put into standby mode. When an area is at a high risk level, an audible and visual alarm is activated throughout the entire tunnel, an emergency broadcast is activated, a warning message is sent to the management personnel, the power supply to the monitored area and surrounding areas is cut off, and all smoke exhaust systems and fire extinguishing devices are activated.

[0024] A coal mine fire early warning system, comprising:

[0025] The sensor layer is responsible for collecting various data. Specifically, laser spectral sensors are evenly distributed every 50m in the goaf, roadways, and electromechanical chambers to acquire CO and C2H4 gas concentration data; multi-parameter sensors are evenly distributed every 30m in the goaf, roadways, and electromechanical chambers, integrating five monitoring modules for methane, oxygen, temperature and humidity, pressure, and carbon monoxide to acquire methane, oxygen, temperature and humidity, pressure, and carbon monoxide concentration data; infrared thermal imagers are installed inside the electromechanical chambers to monitor equipment surface temperature in real time; and smoke concentration sensors are evenly distributed every ten meters in the goaf, roadways with conveyor belts, and electromechanical chambers to acquire smoke concentration data in these areas.

[0026] The transmission layer is responsible for transmitting all the data collected by the sensors to the processing layer. It adopts an industrial Ethernet-based transmission method with ZigBee networking as a supplement. ZigBee is used for short-distance transmission of sensor data in areas where wiring is difficult.

[0027] The processing layer is used to process all sensor data. Edge computing nodes perform preliminary processing on the data from each sensor, and the remote server constructs a fire risk scoring model and determines the risk level based on the data from each sensor and the coal type correction parameters.

[0028] The application layer provides a visual interface, alarm push notifications, and decision support. The visual interface is represented by an early warning platform, which displays monitoring data and early warning information in real time. Alarm push notifications and decision support are remotely controlled via a control terminal, which is used to remotely control audible and visual alarms, emergency broadcasts, send early warning information, and cut off power, smoke extraction systems, and fire extinguishing devices in and around the monitored area.

[0029] The beneficial effects of this invention are as follows:

[0030] 1. This invention overcomes the limitations of traditional single sensors by employing a collaborative approach involving laser spectroscopy, multi-parameter sensors, and smoke sensors to achieve multi-dimensional sensing of various key parameters such as gas concentration, temperature, and smoke. This multi-modal fusion monitoring technology, through the complementary advantages of different sensor types, significantly improves the accuracy and reliability of monitoring. In the complex mine environment, the laser spectroscopy sensor can accurately detect the concentrations of key gases such as CO and C2H4, while the multi-parameter sensor can monitor environmental parameters such as methane, oxygen, temperature, and humidity in real time, and the smoke sensor can monitor smoke data in real time. Experimental data shows that after adopting multi-modal fusion monitoring, the system's false alarm rate is reduced to below 5%, effectively avoiding unnecessary emergency responses caused by false alarms and improving the efficiency and safety of coal mine production.

[0031] 2. This invention, based on the DS evidence theory, integrates data on CO and C2H4 gas concentrations, methane concentrations, oxygen concentrations, temperature and humidity, pressure, and carbon monoxide concentrations from goaf areas, roadways, and electromechanical chambers, as well as equipment temperature data from electromechanical chambers and smoke concentration data from goaf areas, conveyor belt roadways, and electromechanical chambers. Based on the fusion results of the DS evidence theory and coal type correction parameters, a fire risk scoring model is constructed. Compared with traditional fixed threshold early warning methods, this invention's technology can better adapt to the complex and variable environment of underground coal mines, improving the accuracy of early warnings. Studies have shown that using this invention significantly improves early warning accuracy, enabling more timely and accurate detection of fire hazards, and providing strong support for the prevention and control of coal mine fires.

[0032] 3. When a risk is detected, this invention combines local ventilation with JTF-I fire extinguishing liquid gel infusion technology, which effectively improves the fire extinguishing efficiency and thus effectively enhances the ability to prevent and control coal mine fires. Attached Figure Description

[0033] Figure 1 This is a flowchart of a coal mine fire early warning and control method according to the present invention. Detailed Implementation

[0034] The present invention will now be described in detail with reference to the accompanying drawings.

[0035] like Figure 1 As shown, a coal mine fire early warning and control method includes the following steps:

[0036] Data on CO, C2H4 gas concentrations, methane concentrations, oxygen concentrations, temperature, humidity, pressure, and carbon monoxide concentrations in coal mine goafs, roadways, and electromechanical chambers; data on equipment temperature in electromechanical chambers; and data on smoke concentrations in goafs, roadways with conveyor belts, and electromechanical chambers.

[0037] Based on the DS evidence theory, a fire risk scoring model is constructed by integrating CO, C2H4 gas concentration data, methane concentration, oxygen concentration, temperature and humidity, pressure, carbon monoxide concentration data, equipment temperature data of electromechanical chambers, and smoke concentration data of goaf, roadway with conveyor belt, and electromechanical chamber. Based on the fusion results of the DS evidence theory and coal type correction parameters, a fire risk scoring model is constructed.

[0038] The risk level is determined based on the fire risk scoring model, and different control schemes are adopted according to different risk levels.

[0039] In this embodiment, the coal type parameter K is used as a fixed correction coefficient, which works synergistically with the sensor data (C, T, S, H) after being fused by DS evidence theory: First, multi-source sensor data are fused by DS evidence theory to obtain a preliminary risk score; then, the K value is determined in combination with the coal type on site, and the preliminary score and warning threshold are adjusted to finally output an accurate risk level.

[0040] Furthermore, the expression for the fire risk scoring model is:

[0041] R = K×(0.4C + 0.3T + 0.2S + 0.1H);

[0042] In the formula, R is the risk coefficient; K is the coal type correction parameter, representing the coal type activity, with a value range of 0.8 to 1.2; C is the comprehensive index of the marker gas; T is the temperature anomaly coefficient; S is the smoke scattering rate; and H is the humidity correction parameter.

[0043] C = 0.6C1 + 0.4C2, where C1 is the standardized value of CO concentration, C1 = measured value of CO concentration / fire threshold value; C2 is the standardized value of C2H4 concentration change rate, C2 is the increase value of C2H4 concentration per unit time;

[0044] T = |T t – T1| / (T2 – T1), where T t To monitor the real-time temperature of the area, T1 is the normal underground temperature baseline value, and T2 is the fire warning temperature threshold value, T2=70℃; when the temperature rises abnormally, the T value approaches 1, which better reflects the potential signs of a fire.

[0045] H = 1 - (RH 实 - RH常 ) / (RH 临高 – RH 临低 ), where RH 实 The relative humidity (RH) is the measured value of the ambient humidity. 常 This is the baseline value for normal downhole humidity; RH 临高 The humidity warning threshold is RH. 临高 The set value is 95%RH; RH 临低 This is the critical low value for humidity warning, RH 临低 The set value is 30%RH; the lower the humidity, the easier it is for coal to oxidize, and the higher the H value. This parameter is used to correct other indicators and improve the accuracy of the model.

[0046] S = S1 / S2, where S1 is the smoke concentration in the detection area and S2 is the baseline value of normal smoke concentration in the well. The higher the smoke content, the larger the S value, which is a clear indicator of fire development. When there is no smoke, S = 0, and when the smoke scattering rate reaches the fire critical value, S = 1.

[0047] In this implementation, the values ​​of C, T, S, and H range from 0 to 1; the value of the coal type correction parameter K ranges from 0.8 to 1.2, determined based on the oxidation activity of the coal type. For high-activity coal (lignite, long-flame coal): volatile matter V > 37%, fast oxidation rate, early release of indicator gases, K = 1.2 (lowering the warning threshold and improving warning sensitivity); for medium-activity coal (gas coal, fat coal, coking coal): 20% < V ≤ 37%, medium oxidation activity, K = 1.0 (adopting the baseline warning threshold); for low-activity coal (lean coal, anthracite): V ≤ 20%, slow oxidation rate, delayed release of indicator gases, K = 0.8 (increasing the warning threshold and avoiding false warnings).

[0048] Furthermore, based on the fire risk scoring model, the risk level is determined, and the specific methods for adopting different control schemes according to different risk levels are as follows:

[0049] When the coal type is a highly reactive coal type, k=1.2, R<0.25 is the safety level, 0.25≤R<0.42 is the low risk level, 0.42≤R<0.58 is the medium risk level, and R≥0.58 is the high risk level.

[0050] When the coal type is medium-reactivity coal, k=1.0, R<0.3 is the safety level, 0.3≤R<0.5 is the low-risk level, 0.5≤R<0.7 is the medium-risk level, and R≥0.7 is the high-risk level.

[0051] When the coal type is low-activity coal, k=0.8, R<0.38 is the safety level, 0.38≤R<0.63 is the low safety level, 0.63≤R<0.88 is the medium risk level, and R≥0.88 is the high risk level.

[0052] When all areas are at a safe level, all sensors in each area are monitoring normally, and coal oxidation is in its initial stage. When an area is at a low safety level, there is a minor fire hazard, with a slight increase in the concentration of indicator gases or a slight temperature anomaly. An audible and visual alarm is activated in this area, and the sampling frequency of the corresponding sensors is increased to once every 30 seconds to strengthen the monitoring of changes in indicator gas concentration and temperature. No active gas control measures are required. When an area is at a medium risk level, the fire hazard is more significant, with a continuous increase in the concentration of indicator gases and an abnormal temperature rise. An audible and visual alarm is activated in this area, and a warning message is sent to management personnel. Simultaneously, the local smoke exhaust system is activated for ventilation, and fire extinguishers are put into standby mode. When an area is at a high risk level, there is a serious fire risk, and an initial fire may have already occurred. An audible and visual alarm is activated throughout the entire roadway, an emergency broadcast is activated, and a warning message is sent to management personnel. Simultaneously, the power supply to the monitored area and surrounding areas is cut off, all smoke exhaust systems and fire extinguishing devices are activated, and local fans are started via the PLC control system. This system can quickly respond and accurately control the fan operating parameters, increasing the wind speed in the target area to 1.2-1.5 m / s. Within this wind speed range, air circulation is effectively promoted, achieving a gas dilution rate of ≥30% within 3 minutes, reducing the likelihood of fire and creating favorable conditions for personnel evacuation and subsequent firefighting efforts. The system employs JTF-I fire extinguishing liquid gel injection technology, which integrates the fire extinguishing advantages of gel, yellow mud grouting, three-phase foam, nitrogen, and inhibitors. Real-time adjustment of the grouting flow rate via pressure sensor feedback ensures a high-level fire source coverage efficiency of over 90% in the goaf area. The pressure sensor monitors pressure changes in the grouting pipeline in real time and automatically adjusts the grouting pump flow rate based on pressure feedback, ensuring the grout evenly covers the fire source area. Simultaneously, precise control of the gel formulation and formation process maintains the gel's water content between 75% and 85%. This water content range ensures both good water-fixing properties and sufficient fluidity for diffusion within the goaf area, thereby achieving efficient fire extinguishing and preventing reignition.

[0053] A coal mine fire early warning system, including

[0054] The sensor layer, the foundation of the entire system, is responsible for collecting various data. Specifically, laser spectral sensors are evenly distributed every 50m in the goaf, roadways, and electromechanical chambers to acquire CO and C2H4 gas concentration data, with an accuracy of ≤0.1ppm. This ensures the capture of minute gas concentration changes, providing crucial data for early fire warning and ensuring comprehensive and timely monitoring of characteristic gas concentration changes within the goaf. Multi-parameter sensors are evenly distributed every 30m in the goaf, roadways, and electromechanical chambers. These sensors integrate five monitoring modules for methane, oxygen, temperature and humidity, pressure, and carbon monoxide, acquiring data on methane, oxygen, temperature and humidity, pressure, and carbon monoxide concentrations. The response time of these sensors is ≤10 seconds. The system can react quickly to changes in environmental parameters, promptly detect potential fire hazards, and comprehensively monitor environmental parameters. An infrared thermal imager is installed inside the electromechanical chamber to monitor the surface temperature of equipment in real time, with a measurement accuracy of ±0.5℃. This allows for real-time monitoring of equipment surface temperature and early detection of potential fire risks caused by equipment overheating. Smoke concentration sensors are evenly distributed every ten meters in goaf areas, conveyor belt tunnels, and electromechanical chambers. If the tunnel turns or changes slope, the spacing needs to be reduced to 5 meters to obtain smoke concentration data within the goaf areas and tunnels.

[0055] To ensure comprehensive monitoring, the density of sensor deployment can be appropriately increased at key points and in easily overlooked areas, forming a comprehensive, multi-layered sensor monitoring network. This ensures timely detection of fire hazards and provides reliable data support for early warning and prevention of coal mine fires.

[0056] When installing the laser spectral sensor, it is crucial to strictly adhere to the requirement of vertical suspension from the tunnel ceiling, ensuring a distance of ≥0.5m from the sidewall. This prevents interference from the tunnel sidewall and guarantees sampling accuracy. The sampling port should face the airflow direction to allow for smoother gas entry and improve the timeliness of gas detection. The multi-parameter sensor employs a flexible suspension mechanism with a unique design that supports ±15° tilt adjustment. During actual installation, the tilt angle is adjusted according to the specific airflow conditions and equipment layout within the tunnel to ensure unobstructed airflow, enabling the sensor to accurately monitor various gas and environmental parameters.

[0057] The transmission layer is responsible for transmitting all data collected by sensors to the processing layer. Constructing an underground ring network switch cluster is crucial for building a stable data transmission network. During implementation, high-performance ring network switches were selected and rationally deployed according to the mine's roadway layout and monitoring point distribution to ensure fast and stable data transmission from each monitoring point. The transmission network uses an industrial Ethernet transmission network as its backbone, carrying the Modbus TCP application layer protocol to achieve synchronous acquisition of sensor data. A 10Hz high-frequency sampling mechanism was established, coupled with high-precision gas sensors, to capture subtle changes in gas concentration fluctuations ΔC ≥ 1ppm in real time. This provides high-time-density data for intelligent analysis and early warning, avoiding the omission of critical anomaly information due to excessively long sampling periods.

[0058] Industrial Ethernet, with its high speed and high stability, undertakes the backbone data transmission task; ZigBee self-organizing network serves as a supplementary coverage, enabling short-range, low-power wireless transmission of sensors in complex alleyways and areas where wiring is difficult, ensuring data availability across all scenarios.

[0059] At the transmission security level, AES-256 symmetric encryption technology is used to provide end-to-end encryption protection for all data transmitted across the network, preventing data from being stolen or tampered with in the transmission link and improving the security compliance of the transmission layer.

[0060] The processing layer, the core computing unit of the system, processes all sensor data. Edge computing nodes perform preliminary processing on the sensor data, reducing the burden on the cloud server and enabling rapid data analysis and decision-making. The remote server constructs a fire risk scoring model and determines the risk level based on the sensor data and coal type correction parameters. Edge computing nodes are deployed in substations, utilizing the stable power supply and favorable environmental conditions to ensure their stable operation. Through optimized algorithms and hardware configuration, local data preprocessing is achieved, with a data compression rate ≥70%, effectively reducing data transmission volume and improving system efficiency. The cloud server runs a fusion algorithm every 5 seconds, continuously analyzing multi-source data.

[0061] The data processing flow begins with data acquisition. Each sensor collects data in real time using a high-frequency sampling mechanism of 10Hz, ensuring timely detection of subtle changes in gas concentration. The acquired data first enters a noise reduction and filtering stage, employing the Kalman filter algorithm. This algorithm effectively removes noise interference from the data, improving its accuracy and reliability. After noise reduction and filtering, the data enters the feature extraction stage. The PCA dimensionality reduction algorithm is used to extract key features from high-dimensional data, reducing data dimensionality and computational load while retaining the main information. The feature-extracted data then enters the risk assessment module. This module, based on D-S evidence theory, integrates multi-source data such as gas concentration, temperature gradient, and smoke scattering rate to construct a fire risk scoring model for a comprehensive assessment of fire risk. Finally, based on the risk assessment results, early warning decisions are made. When the risk score reaches the corresponding early warning threshold, the system automatically issues an early warning signal and generates a fire source location heatmap in conjunction with the GIS system, providing accurate information support for subsequent firefighting and rescue operations. Throughout the entire data processing process, the data processing delay is ≤2 seconds, ensuring the timeliness and effectiveness of early warnings and saving valuable time for coal mine fire prevention and control.

[0062] The application layer provides users with an intuitive interactive interface, which offers visualization, alarm push notifications, and decision support. The visualization interface is represented by an early warning platform, which displays monitoring data and early warning information in real time, allowing staff to understand the mine's safety status promptly. Alarm push notifications and decision support are remotely controlled via a control terminal, which is used to remotely control audible and visual alarms, emergency broadcasts, send early warning information, and cut off power, smoke extraction systems, and fire extinguishing devices in and around the monitored area, thereby achieving effective prevention and control of mine fires.

[0063] The data flow starts from the sensor layer, passes through the transmission layer to the processing layer, and the processed results are then fed back to the application layer, forming a complete closed loop. The layers are interconnected through data interfaces to ensure efficient system operation. For example, when the processing layer detects a fire risk, it immediately sends an early warning to the application layer. Simultaneously, the control terminal initiates corresponding control measures according to preset strategies, enabling timely response and handling of the fire.

[0064] When using this invention:

[0065] 1. Gas detection accuracy ≤ 0.1ppm. This sub-ppm high-precision detection capability can detect gas in the early stages of a fire when the gas concentration is still at an extremely low level. Compared with traditional detection technologies, it can provide an early warning 2-3 days in advance, buying valuable time for fire prevention and extinguishing measures.

[0066] 2. Warning response time: The system's warning response time is ≤5 seconds. Once a fire risk is detected to reach the warning threshold, it can respond in a very short time, achieving real-time dynamic response, enabling staff to take measures quickly and effectively curb the development of the fire.

[0067] 3. Ventilation adjustment rate: The smoke exhaust system can achieve an adjustable wind speed of 1.2-1.5m / s. In the event of a fire, it can quickly adjust the wind speed to dilute the gas within 3 minutes, reduce the concentration of toxic and harmful gases, improve the working environment, and create favorable conditions for personnel evacuation and fire fighting.

[0068] 4. Gel retention rate: The gel infusion technology used achieves a gel retention rate of ≥90%, allowing it to remain in the fire area for an extended period, providing sustained cooling and isolating oxygen, effectively preventing fire reignition and improving fire extinguishing efficiency.

[0069] Example 1

[0070] In 2025, a laser spectral sensor in a goaf area of ​​a mine in Inner Mongolia played a crucial role, continuously detecting a stable CO concentration of 8-10 ppm for 6 hours, while the C2H4 concentration increased at a rate of 0.2 ppm / hour. This data change indicated that the coal oxidation reaction in the area was gradually intensifying, and the fire risk was continuously increasing. The system immediately initiated a risk assessment program. Based on multi-source data fusion analysis, a risk score of R=0.89 was calculated. Combined with the actual mine setting as medium-reactivity coking coal and matching the coal type correction parameter K=1.0, the risk level was reached as high. This means that the area has a high fire risk and may have entered the early stage of a fire, requiring immediate and effective prevention and control measures to prevent the occurrence and spread of the fire.

[0071] Upon receiving the early warning, the emergency response plan was quickly activated. First, the local ventilation fans were started via the PLC control system. This system can quickly and precisely adjust the operating parameters of the fans, rapidly increasing the wind speed in the target area from 0.8 m / s to 1.3 m / s. The higher wind speed effectively promoted air circulation, allowing toxic and harmful gases to be quickly diluted, reducing the likelihood of a fire. Simultaneously, 300 m³ of Pruitt fire extinguishing liquid was injected into the area. This fire extinguishing liquid combines the advantages of gel, yellow mud grouting, three-phase foam, nitrogen, and inhibitors. During the injection process, the grouting flow rate was adjusted in real time using pressure sensors to ensure that the fire extinguishing liquid could evenly cover every corner of the goaf, especially high-level fire sources, achieving a coverage efficiency of over 90%. After 48 hours of continuous treatment, the gas concentration gradually decreased to a safe threshold, with the CO concentration dropping to ≤5 ppm, successfully curbing the trend of spontaneous combustion of coal and ensuring safe production in the mine.

[0072] Compared with traditional fire prevention and control methods, the technical solution of this invention exhibits significant advantages. The warning time is advanced by 12 hours, enabling staff to detect and take timely measures as soon as fire risks emerge, gaining valuable time for fire prevention and control. The grouting volume is reduced by 25%, which not only lowers firefighting costs but also reduces the impact on the mine environment. The cooling rate of the fire zone is increased by 35%, and the faster cooling rate effectively inhibits coal oxidation, prevents fire reignition, and improves firefighting effectiveness. Analysis of this application case fully verifies the effectiveness and superiority of the technical solution of this invention, providing reliable technical support for coal mine fire prevention and control.

[0073] Example 2

[0074] In a mine in East China, a comprehensive compatibility test was conducted on the fire early warning gas control system of this invention and the existing safety monitoring system (KJ90X). During the test, the stability and accuracy of data transmission, as well as the synchronization of alarm information, were the main focus. Through careful system configuration and debugging, seamless data integration between the two systems was achieved. In terms of data transmission, standardized data interfaces and communication protocols were adopted to ensure that data collected by sensors could be accurately transmitted to the KJ90X system, achieving real-time data sharing. The alarm information synchronization rate reached 100%. When the fire early warning gas control system detected an anomaly and issued an alarm, the KJ90X system could simultaneously receive the alarm information and promptly display and process it, ensuring that staff could obtain accurate early warning information in a timely manner. To further verify the system's stability, a 12-month operational test was conducted. During this period, the system underwent various complex operating conditions, including different gas concentrations, temperatures, humidity, and other environmental conditions, as well as frequent equipment start-ups and shutdowns. After rigorous monitoring and statistical analysis, the system failure rate was ≤0.5%, indicating that the system has extremely high stability and reliability, meeting the needs of long-term, stable operation in coal mines and providing a solid guarantee for safe coal mine production.

[0075] CFD fluid simulation technology can accurately simulate the airflow distribution and temperature field within a mine, providing a scientific basis for optimizing the ventilation system. In the event of a fire, the system can quickly adjust the operating parameters of the ventilation fans and optimize the ventilation path based on the CFD simulation results, ensuring that toxic and harmful gases are discharged promptly and fresh air is quickly replenished. Simultaneously, pressure sensors monitor pressure changes in the grouting pipeline in real time, dynamically adjusting the grouting volume to ensure the grout more accurately covers the fire source area. Experimental results show that after adopting the precise control strategy, the fire extinguishing efficiency is increased by 40% compared to traditional methods, effectively improving the prevention and control capabilities of coal mine fires.

[0076] The technical solution of this invention can significantly reduce the coal mine fire accident rate, by an estimated 60% or more. Through early warning and precise control, it provides underground personnel with at least 15 minutes to evacuate. In the early stages of a fire, the system can promptly issue warning signals, notifying underground personnel to evacuate quickly. Simultaneously, precise fire control measures can effectively curb the spread of the fire, creating a safe environment for personnel evacuation. This 15-minute evacuation time is crucial for ensuring the safety of underground personnel and can greatly reduce the occurrence of casualties.

[0077] This invention enables early warning systems to promptly detect fire hazards and implement effective prevention and control measures, reducing equipment damage and production losses. Based on real-world application case studies, a single mine can save an average of approximately 8 million yuan in disaster prevention costs annually. Precise grouting control strategies increase grouting material utilization by over 20%, reducing material waste and lowering production costs. Early warning systems can also prevent equipment damage and production stoppages caused by fires, reducing equipment repair and replacement costs, as well as economic losses due to production shutdowns.

[0078] This invention promotes the construction of smart mines, improving coal mine safety, reducing accidents, protecting miners' lives and families' well-being, and fostering the sustainable development of the coal industry. Smart mine construction can also improve coal production efficiency and quality, reduce energy consumption and environmental pollution, and contribute to socio-economic development. This invention has broad prospects for industry application and is expected to be widely used in the coal mining industry nationwide and even globally, providing strong technical support for safe coal mine production.

[0079] The coal mine fire early warning gas control method provided by this invention effectively solves the problems of delayed early warning and crude control in existing systems through the innovative combination of multi-dimensional monitoring, intelligent algorithm fusion, and precise control technology. It offers an efficient and reliable technical solution for coal mine fire prevention and control, demonstrating significant technological advancement and engineering practicality. In practical applications, it can detect fire hazards in advance, reduce accident risks, protect personnel safety, and ensure the sustainability of coal mine production. It is expected to become a key technology in the field of coal mine safety, driving the industry towards intelligent and safe development. This invention is applicable to various fire-prone areas in coal mines, including goaf areas, tunneling faces, electromechanical chambers, belt conveyor roadways, and coal bunker perimeters. It is adaptable to different coal types such as lignite, bituminous coal, and anthracite, and can be widely applied in high-gas mines, mines with easily self-igniting coal seams, and other types of coal mines, providing efficient and precise fire early warning and gas control guarantees for safe coal mine production.

Claims

1. A method for early warning and control of coal mine fires, characterized in that, Includes the following steps: Data on CO, C2H4 gas concentrations, methane concentrations, oxygen concentrations, temperature, humidity, pressure, and carbon monoxide concentrations in coal mine goafs, roadways, and electromechanical chambers; data on equipment temperature in electromechanical chambers; and data on smoke concentrations in goafs, roadways with conveyor belts, and electromechanical chambers. Based on the DS evidence theory, a fire risk scoring model is constructed by integrating CO, C2H4 gas concentration data, methane concentration, oxygen concentration, temperature and humidity, pressure, carbon monoxide concentration data, equipment temperature data of electromechanical chambers, and smoke concentration data of goaf, roadway with conveyor belt, and electromechanical chamber. Based on the fusion results of the DS evidence theory and coal type correction parameters, a fire risk scoring model is constructed. The risk level is determined based on the fire risk scoring model, and different control schemes are adopted according to different risk levels.

2. The coal mine fire early warning and control method according to claim 1, characterized in that, The expression for the fire risk scoring model is: R = K×(0.4C + 0.3T + 0.2S + 0.1H); In the formula, R is the risk coefficient; K is the coal type correction parameter, representing the coal type activity, with a value range of 0.8 to 1.2; C is the comprehensive index of the marker gas; T is the temperature anomaly coefficient; S is the smoke scattering rate; and H is the humidity correction parameter. C = 0.6C1 + 0.4C2, where C1 is the standardized value of CO concentration; C2 is the standardized value of the rate of change of C2H4 concentration. T = |T t – T1| / (T2 – T1), where T t To monitor the real-time temperature of the area, T1 is the reference value for normal underground temperature, and T2 is the critical value for fire warning temperature. H = 1 - (RH 实 - RH 常 ) / (RH 临高 – RH 临低 ), where RH 实 The relative humidity (RH) is the measured value of the ambient humidity. 常 This is the baseline value for normal downhole humidity; RH 临高 The humidity warning threshold is RH. 临高 The set value is 95%RH; RH 临低 This is the critical low value for humidity warning, RH 临低 The set value is 30%RH; S = S1 / S2, where S1 is the smoke concentration in the detection area and S2 is the baseline value of normal smoke concentration in the well.

3. The coal mine fire early warning and control method according to claim 1, characterized in that, The specific methods for determining the risk level based on the fire risk scoring model and adopting different control schemes according to different risk levels are as follows: When the coal type is highly reactive, R < 0.25 indicates a safe level, 0.25 ≤ R < 0.42 indicates a low-risk level, 0.42 ≤ R < 0.58 indicates a medium-risk level, and R ≥ 0.58 indicates a high-risk level. When the coal type is medium-reactivity coal, R < 0.3 is the safety level, 0.3 ≤ R < 0.5 is the low-risk level, 0.5 ≤ R < 0.7 is the medium-risk level, and R ≥ 0.7 is the high-risk level. When the coal type is low-activity coal, R < 0.38 is the safety level, 0.38 ≤ R < 0.63 is the low safety level, 0.63 ≤ R < 0.88 is the medium risk level, and R ≥ 0.88 is the high risk level. When each area is at a safe level, all sensors in each area monitor normally. When an area is at a low safety level, an audible and visual alarm is activated in that area, and the sampling frequency of the corresponding sensors in that area is increased. When an area is at a medium risk level, an audible and visual alarm is activated in that area, a warning message is sent to the management personnel, the local smoke exhaust system is activated for ventilation, and the fire extinguishing equipment is put into standby mode. When an area is at a high risk level, an audible and visual alarm is activated throughout the entire tunnel, an emergency broadcast is activated, a warning message is sent to the management personnel, the power supply to the monitored area and surrounding areas is cut off, and all smoke exhaust systems and fire extinguishing devices are activated.

4. A coal mine fire early warning system, based on the coal mine fire early warning and control method described in claim 1, characterized in that, include: The sensor layer is responsible for collecting various data. Specifically, laser spectral sensors are evenly distributed every 50m in the goaf, roadways, and electromechanical chambers to acquire CO and C2H4 gas concentration data; multi-parameter sensors are evenly distributed every 30m in the goaf, roadways, and electromechanical chambers, integrating five monitoring modules for methane, oxygen, temperature and humidity, pressure, and carbon monoxide to acquire methane, oxygen, temperature and humidity, pressure, and carbon monoxide concentration data; infrared thermal imagers are installed inside the electromechanical chambers to monitor equipment surface temperature in real time; and smoke concentration sensors are evenly distributed every ten meters in the goaf, roadways with conveyor belts, and electromechanical chambers to acquire smoke concentration data in these areas. The transmission layer is responsible for transmitting all the data collected by the sensors to the processing layer. It adopts an industrial Ethernet-based transmission method supplemented by ZigBee networking. ZigBee is used for short-distance transmission of sensor data in areas where wiring is difficult. The processing layer is used to process all sensor data. Edge computing nodes perform preliminary processing on the data from each sensor, and the remote server constructs a fire risk scoring model and determines the risk level based on the data from each sensor and the coal type correction parameters. The application layer provides a visual interface, alarm push notifications, and decision support. The visual interface is represented by an early warning platform, which displays monitoring data and early warning information in real time. Alarm push notifications and decision support are remotely controlled via a control terminal, which is used to remotely control audible and visual alarms, emergency broadcasts, send early warning information, and cut off power, smoke extraction systems, and fire extinguishing devices in and around the monitored area.

Images