Mine underground mining intelligent monitoring system

Abstract

The invention belongs to the field of mine underground mining, and particularly relates to a mine underground mining intelligent monitoring system which comprises a sensing layer, a network layer and a data processing layer, the sensing layer comprises a monitoring terminal for detecting operation conditions and an environment sensor for detecting environment conditions, and the network layer comprises a mixed transmission network and relay equipment. And the data processing layer comprises an edge computing node module, a data credibility evaluation module, an assessment result correction module and a maintenance demand judgment module. Through data comprehensive credibility calculation and a score correction formula, the influence of environmental factors such as dust shielding and noise interference on monitoring data is effectively counteracted, the goodness of fit between assessment scores and manual assessment of experts is greatly improved, the system is remarkably superior to a traditional system, and misjudgment caused by data distortion is avoided.

Description

Technical Field

[0001] This invention belongs to the field of underground mining, and specifically relates to an intelligent monitoring system for underground mining. Background Technology

[0002] In the production operations of non-coal underground mines such as metal mines, chemical mines, and gypsum mines, assessing and scoring the level of standardization of underground workers' work is an important means to ensure production safety and improve production efficiency. By monitoring workers' work processes, attitudes, and language, non-standard operations can be identified in a timely manner, workers can be urged to comply with operating procedures, and the occurrence of safety accidents can be reduced.

[0003] However, the underground environment in non-coal mines is extremely harsh, generally characterized by high humidity, high dust levels, strong corrosiveness, and high stress. These problems cause sensors and transmission equipment used for monitoring to age easily, short-circuit, or be physically damaged. For example, the acidic water produced by the oxidation of sulfide minerals in metal mines can corrode cables and sensor interfaces, while the high dust levels in limestone mines can clog sensor pores, leading to data distortion.

[0004] Meanwhile, underground tunnels are winding, fault-prone, and have dense rock masses, causing wireless signals such as 4G / 5G and WiFi to attenuate extremely quickly or even become completely blocked; wired transmissions such as fiber optic cables and electrical cables are easily broken due to tunnel deformation, blasting vibrations, and other mining activities, resulting in "disconnection" of monitoring data. Especially in deep mining or complex goaf areas, signal coverage blind spots are common.

[0005] In existing technologies, monitoring systems for assessing the performance of underground workers often struggle to adapt to the extreme environments described above, resulting in problems such as high equipment maintenance frequency, distorted monitoring data, and inaccurate assessment scores. More importantly, existing systems cannot quantify the impact of data distortion caused by environmental factors on assessment results, making it difficult to scientifically determine whether equipment requires maintenance, often leading to either "over-maintenance" or "delayed maintenance." Summary of the Invention

[0006] The present invention aims to overcome the shortcomings of the prior art and provide an intelligent monitoring system for underground mining.

[0007] To achieve the above objectives, the technical solution adopted by this invention is: an intelligent monitoring system for underground mining, comprising a perception layer, a network layer, and a data processing layer. The perception layer includes monitoring terminals for detecting operational conditions and environmental sensors for detecting environmental conditions. The network layer includes a hybrid transmission network and relay equipment. The data processing layer includes an edge computing node module, a data reliability assessment module, an assessment result correction module, and a maintenance requirement judgment module. The edge computing node module preprocesses the raw data collected by the monitoring terminal to obtain a video data quality index Q1, an audio data quality index Q2, a motion data quality index Q3, and a positioning data quality index Q4, and calculates the original data reliability.

[0008] RC=α×Q1+β×Q2+γ×Q3+δ×Q4

[0009] Q1, Q2, Q3 and Q4 all take values ​​in the range of 0-1, and α, β, γ and δ are weighting coefficients that satisfy α+β+γ+δ=1;

[0010] The edge computing node module also preprocesses the raw data collected by environmental sensor parameters to obtain a reliability attenuation coefficient:

[0011]

[0012] P i Let P be the measured value of the i-th type of environmental parameter. i,max For the tolerance threshold of this parameter, w i To influence the weights and satisfy ∑w i =1, where n represents the types of environmental parameters;

[0013] The data credibility assessment module derives the overall data credibility based on the original data credibility and the credibility decay coefficient.

[0014] CC = RC × K;

[0015] The assessment result correction module adjusts the initial assessment score based on the overall reliability of the data.

[0016] S2=S1×(1+θ×(1−CC))

[0017] S2 is the corrected score, S1 is the initial score, and θ is the correction coefficient with a value range of 0.1-0.3.

[0018] The maintenance demand assessment module determines whether the sensing layer equipment needs maintenance based on the Maintenance Demand Index (MR).

[0019] MR=λ(1−CC)+μF+ν(S2−S1) / S1

[0020] F represents the failure probability of the sensing layer device, and λ, μ, and ν are weighting coefficients that satisfy λ+μ+ν=1. When MR is greater than the set value, it is determined that maintenance is required.

[0021] Furthermore, the environmental sensors include a humidity sensor, a dust concentration detector, and an H2S gas concentration detector.

[0022] Furthermore, when MR ≥ 0.6, it is determined that repair is required.

[0023] Furthermore, it also includes an application layer, which includes an assessment and scoring terminal and an equipment management terminal. The assessment and scoring terminal is used to display assessment and scoring results and replay of the operation process, while the equipment management terminal is used to display equipment status, maintenance warnings, and maintenance scheduling information.

[0024] Furthermore, the monitoring terminal includes a camera, a microphone, a motion sensor, and a positioning module.

[0025] Furthermore, the camera uses an anti-fog and dustproof lens and is coated with a hydrophobic and oleophobic coating; the positioning module uses a combination of inertial navigation and UWB positioning.

[0026] Furthermore, the hybrid transmission network includes a wireless mesh network and a wired transmission network.

[0027] Furthermore, the environmental sensor is provided with a housing using a gradient functional coating, which includes: a bottom layer using a graphene / polytetrafluoroethylene composite anti-corrosion layer; a middle layer using a biomimetic micro-nano structure hydrophobic layer; and a surface layer using a wear-resistant layer with embedded silicon carbide nanoparticles.

[0028] Furthermore, the environmental sensor is equipped with a self-cleaning device connected to it. When the data detected by the environmental sensor reaches a set value, the self-cleaning device automatically starts cleaning the environmental sensor.

[0029] Furthermore, the self-cleaning device includes a miniature air pump and an annular nozzle; the relay equipment is powered by a combination of downhole power and power collected by photovoltaic panels.

[0030] Compared to existing technologies, this invention offers the following advantages: By using a comprehensive data reliability calculation and scoring correction formula, it effectively offsets the impact of environmental factors such as dust obstruction and noise interference on monitoring data, significantly improving the consistency between assessment scores and expert manual evaluations. This is significantly superior to traditional systems and avoids misjudgments caused by data distortion. Based on a maintenance demand index-based early warning mechanism, this invention significantly improves the accuracy of maintenance response through a three-dimensional assessment of comprehensive reliability, equipment failure probability, and scoring deviation. It can effectively warn of potential faults while reducing unnecessary manual inspections, significantly lowering maintenance costs. Furthermore, this invention integrates multi-dimensional data from video, audio, motion, positioning, and environmental sensors, effectively controlling scoring errors in key assessment items and enhancing the objectivity and credibility of assessment results. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the system architecture of the present invention; Detailed Implementation

[0032] See Figure 1 This embodiment takes the underground working face of a large iron mine as the application scenario. The average humidity in the underground iron mine is 85%-92%, the instantaneous dust concentration after blasting can reach 150mg / m³, and the H2S concentration occasionally reaches 15ppm. The iron mine also has a complex tunnel structure, with a maximum burial depth of 1200 meters and typical extreme environmental characteristics such as multiple faults.

[0033] like Figure 1 As shown, the system provided in this embodiment is applied to underground operations in the iron ore mine. This embodiment includes a sensing layer, a network layer, a data processing layer, and an application layer.

[0034] The sensing layer uses monitoring equipment and sensors to collect technical data on downhole operations.

[0035] The perception layer specifically includes a monitoring terminal and environmental sensors. The monitoring terminal includes a camera, a microphone, a motion sensor, and a positioning module. In this embodiment, the environmental sensors include a humidity sensor, a dust concentration detector, and an H2S gas concentration detector. In other embodiments, vibration, methane gas, and other detection sensors may also be incorporated into the environmental sensors.

[0036] In this embodiment, each underground worker is equipped with one monitoring terminal. The monitoring terminal adopts an IP68 waterproof rating shell design and is fixed to the front of the safety helmet. The monitoring terminal contains:

[0037] The camera uses a 2-megapixel anti-fog and dustproof lens. The lens surface is coated with a polydimethylsiloxane hydrophobic and oleophobic coating, with a measured water contact angle of 155°, which can effectively prevent mine water and dust from adhering.

[0038] The microphone has a built-in noise reduction algorithm that uses adaptive filtering to filter high-frequency noise generated by equipment such as rock drills and crushers, while preserving the human voice frequency range.

[0039] The motion sensor uses a triaxial accelerometer to capture the movements of drilling, support and other operations.

[0040] The positioning module integrates MEMS inertial navigation and UWB positioning. One UWB base station is deployed every 50 meters along the roadway. In the goaf area without UWB signal, the positioning accuracy can be maintained within ±0.5 meters for 10 minutes through inertial navigation.

[0041] In this embodiment, 10 sets of environmental sensors are deployed in the working face and a 500-meter-long tunnel around it, with each set of environmental sensors spaced 50 meters apart. Each set of environmental sensors has a stainless steel housing with a gradient functional coating on its surface, which includes a base layer, a middle layer, and a top layer.

[0042] The bottom layer is a graphene / PTFE composite anti-corrosion layer with a thickness of 30μm, which has been verified by salt spray test to withstand acidic solution corrosion with pH=2; the middle layer is a biomimetic micro-nano structure hydrophobic layer with a thickness of 20μm, which uses nano-sized SiO2 particles to construct a micron column array; the surface layer is a wear-resistant layer with a thickness of 30μm and embedded silicon carbide nanoparticles with a particle size of 50nm, and a hardness of HV1100.

[0043] In this embodiment, after the worker enters the work area, the monitoring terminal automatically turns on, the camera captures video of the work process, the microphone collects the work language, the motion sensor records body movements, and the positioning module outputs the position coordinates in real time.

[0044] The environmental sensor collects environmental data every 10 seconds. In this embodiment, the environmental sensor collects humidity of 88%, dust concentration of 65 mg / m³, and H2S concentration of 8 ppm.

[0045] The network layer includes hybrid transmission networks and relay devices. The hybrid transmission network includes wireless mesh networks and wired transmission networks.

[0046] The wireless mesh network uses 8 industrial-grade mesh nodes. The node shell integrates a niobium-titanium superconducting shielding layer. The communication module operates in the 2.4GHz frequency band, supports 100 frequency hopping per second, and has a single node coverage radius of 30 meters. It can resist strong electromagnetic interference in magnetite areas.

[0047] The wired transmission network is laid with flexible armored cables along the main roadway. The armor layer is made of galvanized steel strip with aramid fiber, which can withstand tensile force ≥5000N, bending radius ≥200mm, and can adapt to roadway deformation of ±10°.

[0048] The two transmission methods are connected through an intelligent switching module, which automatically switches to wired transmission when the wireless signal strength is less than -85dBm.

[0049] The relay equipment is deployed in signal blind areas, such as the edge of goaf areas and fault zones. The relay equipment uses 20W monocrystalline silicon photovoltaic panels, which convert the light energy collected by the photovoltaic panels into electrical energy and transmit it to the underground to power the relay equipment. At the same time, the underground AC power supply can also power the relay equipment, realizing dual power supply. The relay equipment has a built-in lithium battery as a backup power source, which can maintain operation for a period of time after a power outage.

[0050] In this embodiment, the data collected by the sensing layer is transmitted through a wireless mesh network, and automatically switches to wired transmission in the fault zone area. The relay device enhances the signal strength at the edge of the goaf from -92dBm to -75dBm.

[0051] This embodiment uses a hybrid transmission network combined with relay equipment to effectively solve the signal shielding problem in multi-fault and deep mining scenarios, significantly reduce the data transmission interruption rate, and greatly improve the signal coverage time in blind areas such as goaf areas.

[0052] The data collected by the perception layer is transmitted to the data processing layer via the network layer for data processing.

[0053] The data processing layer includes an edge computing node module, a data credibility assessment module, an assessment result correction module, a maintenance requirement judgment module, and a cloud platform.

[0054] The edge computing node module preprocesses the acquired perception layer data. Specifically, two edge computing servers are deployed in a chamber near the work face. Both edge computing servers adopt a mining explosion-proof and intrinsically safe design and support wide temperature range of -20℃ to 60℃.

[0055] The edge computing node module preprocesses the video data captured by the camera. Specifically, it calculates the video data quality index Q1=0.82 using a sharpness evaluation algorithm based on the Laplacian operator.

[0056] The sharpness algorithm based on the Laplacian operator first converts the image to grayscale. Then, it uses the Laplacian operator to calculate the variance of the response map. Blurry images have smaller variances, while sharp images have larger variances. Next, in OpenCV software, `cv2.Laplacian(image,cv2.CV_64F).var()` is used to perform blur detection, thus effectively determining the image sharpness. In OpenCV, `cv2.Laplacian(image,cv2.CV_64F).var()` is a commonly used image sharpness detection method, mainly used to calculate the Laplacian variance of an image; a larger value generally indicates a sharper image.

[0057] The edge computing node module preprocesses the audio data collected by the microphone. Specifically, it calculates the audio data quality index Q2=0.75 by using the signal-to-noise ratio.

[0058] The edge computing node module preprocesses the motion data collected by the motion sensor. Specifically, by evaluating the integrity of the data, the motion data quality index Q3 is calculated to be 0.90 in this embodiment.

[0059] The edge computing node module preprocesses the positioning data collected by the positioning module. Specifically, it evaluates the positioning error. In this embodiment, the positioning data quality index Q4 is calculated to be 0.85.

[0060] In this embodiment, the video data quality index Q1, audio data quality index Q2, motion data quality index Q3, and positioning data quality index Q4 have all been normalized, with a value range of 0-1. The closer the value is to 1, the higher the data quality is considered, and the closer the value is to 0, the lower the data quality is considered.

[0061] The edge computing node module preprocesses the raw data collected by the monitoring terminal and then calculates the original data reliability.

[0062] RC=α×Q1+β×Q2+γ×Q3+δ×Q4

[0063] Where α, β, γ, and δ are weighting coefficients and satisfy α+β+γ+δ=1.

[0064] In this embodiment, α=0.4, β=0.3, γ=0.2, and δ=0.1, the original data reliability can be calculated as follows:

[0065] RC=0.4×0.82+0.3×0.75+0.2×0.90+0.1×0.85=0.328+0.225+0.18+0.085=0.818.

[0066] The edge computing node module preprocesses the raw data collected by environmental sensor parameters to obtain the reliability attenuation coefficient:

[0067]

[0068] Among them, P i Let P be the measured value of the i-th type of environmental parameter. i,max For the tolerance threshold of this parameter, w i To influence the weights and satisfy ∑w i =1, where n represents the number of environmental parameter types.

[0069] In this embodiment, the tolerance threshold P of the humidity sensor is... 1,max=95%; the tolerance threshold P of the dust concentration detector 2,max =100mg / m 3 The tolerance threshold P of the H2S gas concentration detector 3,max =20ppm.

[0070] The environmental parameter weights are: humidity w1=0.3 (humidity), dust concentration w2=0.4, and H2S concentration w3=0.3.

[0071] In this embodiment, the measured value of the humidity sensor P1 is 88%; the measured value of the dust concentration detector P2 is 65 mg / m³. 3 The measured value of the H2S gas concentration detector was P3 = 8 ppm.

[0072] The reliability attenuation coefficient of environmental parameter calculation is obtained through the above calculation formula:

[0073] K=1-[0.3×(88 / 95)+0.4×(65 / 100)+0.3×(8 / 20)]=1-(0.277+0.26+0.12)=1-0.657=0.343.

[0074] The data credibility assessment module derives the overall data credibility based on the original data credibility and the credibility decay coefficient.

[0075] CC = RC × K = 0.818 × 0.343 ≈ 0.281

[0076] Subsequently, the assessment result correction module adjusts the initial exam score based on the overall data credibility obtained from the data credibility assessment module. The correction formula is as follows:

[0077] S2 = S1 × (1 + θ × (1 − CC)), where S2 is the corrected score, S1 is the initial score, and θ is the correction coefficient with a value range of 0.1-0.3.

[0078] In other embodiments, environmental parameter weights, correction coefficients, etc., can be flexibly configured according to the characteristics of the mineral, such as the strong corrosiveness of chemical minerals and the high dust content of gypsum minerals.

[0079] An initial score for a worker's support operation is given using a human or cloud platform based on a deep learning model. In this embodiment, the initial score S1 = 85.

[0080] The correction coefficient θ is set according to the assessment items for safe operation and routine operation. For safe operation assessments, such as blasting warning and support operations, the value of θ is 0.3, while for routine operation assessments such as ore handling, the value of θ is 0.1.

[0081] This example illustrates a support operation, with a value of θ = 0.3. The score is calculated after correction.

[0082] S2 = 85 × (1 + 0.3 × (1 − 0.281)) = 85 × (1 + 0.3 × 0.719) = 85 × 1.2157 ≈ 103.3 points, rounded to 103 points.

[0083] The maintenance demand assessment module determines whether equipment requires maintenance based on the maintenance demand index. The formula for calculating the maintenance demand index (MR) is as follows:

[0084] MR=λ×(1−CC)+μ×F+ν×∣S2−S1∣ / S1

[0085] Where F is the probability of equipment failure, and λ, μ, and ν are weighting coefficients satisfying λ + μ + ν = 1. In this embodiment, the weighting coefficients are λ = 0.4, μ = 0.3, and ν = 0.3. The maintenance warning threshold is set to MR = 0.6. When MR ≥ 0.6, maintenance is deemed necessary. In other embodiments, MR can be set to other values, such as 0.7, 0.8, etc.

[0086] In this embodiment, the historical fault data displayed by the monitoring terminal worn by the worker shows a fault probability F=0.15.

[0087] Scoring deviation: |S2−S1| / S1=|103−85| / 85≈0.2118, thus obtaining the maintenance demand index:

[0088] MR=0.4×(1−0.281)+0.3×0.15+0.3×0.2118

[0089] =0.4×0.719+0.045+0.0635=

[0090] 0.2876+0.045+0.0635≈0.396<0.6, therefore it is determined that no repair is required.

[0091] The assessment and scoring terminal showed that the worker's final score was 103 points, with the correction reason noted: environmental factors, humidity 88%, dust 65mg / m³. The equipment management terminal showed that all equipment was in normal condition, with no maintenance warnings.

[0092] The cloud platform is used for cloud communication. Specifically, the ground data center deploys cloud platform servers, which carry an operation assessment model based on the Transformer architecture and communicate with the underground edge nodes through 10Gbps optical fiber.

[0093] The application layer includes assessment and scoring terminals and equipment management terminals.

[0094] The assessment and scoring terminal displays the assessment and scoring results as well as a playback of the work process. Specifically, three monitors are configured in the ground monitoring center, and one intrinsically safe touch terminal is deployed in the rest room at the underground work face. All of them support the display of the original score, correction coefficient, final score, and playback of the work video. The touch terminal can display the corrected score and abnormal operation prompts in real time, which can significantly reduce the incidence of worker violations and greatly improve the response speed of safety hazard investigation.

[0095] The equipment management terminal displays equipment status, maintenance warnings, and maintenance scheduling information. Specifically, it uses an industrial-grade tablet computer that supports IP65 protection and has an equipment management APP installed, which can display equipment status, maintenance warnings, and maintenance route planning in real time.

[0096] Example 2

[0097] The difference between this embodiment and Embodiment 1 is that in this embodiment, 10 minutes after the blasting operation, the environmental sensor detected a sudden increase in dust concentration to 120 mg / m³, humidity of 90%, and H2S concentration of 10 ppm. The calculations yielded:

[0098] K=1−[0.3×(90 / 95)+0.4×(120 / 100)+0.3×(10 / 20)]

[0099] =1−(0.284+0.48+0.15)=1−0.914=0.086.

[0100] Due to dust obstruction, the video data quality index Q1 = 0.3. The calculation yields:

[0101] RC=0.4×0.3+0.3×0.6+0.2×0.8+0.1×0.7

[0102] =0.12+0.18+0.16+0.07=0.53,

[0103] CC = 0.53 × 0.086 ≈ 0.0456;

[0104] Repair Demand Index:

[0105] MR = 0.4 × (1 − 0.0456) + 0.3 × 0.2 (increased failure probability) + 0.3 × 0.4 (scoring deviation) ≈ 0.3818 + 0.06 + 0.12 = 0.5618 ≈ 0.56 < 0.6, so no maintenance is needed, but a device cleaning reminder is triggered.

[0106] In this embodiment, a self-cleaning device is installed on the environmental sensor. The self-cleaning device includes a miniature air pump and an annular nozzle. The self-cleaning device is connected to a dust concentration detector. When the dust concentration is detected to be >70mg / m³, the self-cleaning device automatically starts a 0.3MPa pulse airflow. Each cleaning lasts for 0.5 seconds, with an interval of 30 seconds, and is repeated 3 times.

[0107] In this embodiment, the environmental sensor uses a gradient functional coating and a self-cleaning device, which can work stably in extreme environments such as high humidity, high dust and acidic sulfides, significantly extending the mean time between failures (MTBF) of the equipment and greatly reducing the hardware failure rate.

Claims

1. An intelligent monitoring system for underground mining, characterized in that: The system comprises a perception layer, a network layer, and a data processing layer. The perception layer includes monitoring terminals for detecting operational status and environmental sensors for detecting environmental conditions. The network layer includes a hybrid transmission network and relay equipment. The data processing layer includes an edge computing node module, a data reliability assessment module, an assessment result correction module, and a maintenance requirement judgment module. The edge computing node module preprocesses the raw data collected by the monitoring terminal to obtain video data quality index Q1, audio data quality index Q2, motion data quality index Q3, and positioning data quality index Q4, and calculates the original data reliability. RC=α×Q1+β×Q2+γ×Q3+δ×Q4 Q1, Q2, Q3 and Q4 all take values ​​in the range of 0-1, and α, β, γ and δ are weighting coefficients that satisfy α+β+γ+δ=1; The edge computing node module also preprocesses the raw data collected by environmental sensor parameters to obtain a reliability attenuation coefficient: P i Let P be the measured value of the i-th type of environmental parameter. i,max For the tolerance threshold of this parameter, w i To influence the weights and satisfy ∑w i =1, where n represents the types of environmental parameters; The data credibility assessment module derives the overall data credibility based on the original data credibility and the credibility decay coefficient. CC = RC × K; The assessment result correction module adjusts the initial assessment score based on the overall reliability of the data. S2=S1×(1+θ×(1−CC)) S2 is the corrected score, S1 is the initial score, and θ is the correction coefficient with a value range of 0.1-0.

3. The maintenance demand assessment module determines whether the sensing layer equipment needs maintenance based on the Maintenance Demand Index (MR). MR=λ(1−CC)+μF+ν(S2−S1) / S1 F represents the failure probability of the sensing layer device, and λ, μ, and ν are weighting coefficients that satisfy λ+μ+ν=1. When MR is greater than the set value, it is determined that maintenance is required.

2. The intelligent monitoring system for underground mining according to claim 1, characterized in that: The environmental sensors include a humidity sensor, a dust concentration detector, and an H2S gas concentration detector.

3. The intelligent monitoring system for underground mining according to claim 1, characterized in that: When MR≥0.6, it is determined that maintenance is required.

4. The intelligent monitoring system for underground mining according to claim 1, characterized in that: It also includes an application layer, which includes an assessment and scoring terminal and an equipment management terminal. The assessment and scoring terminal is used to display assessment and scoring results and replay of the operation process, while the equipment management terminal is used to display equipment status, maintenance warnings, and maintenance scheduling information.

5. The intelligent monitoring system for underground mining according to claim 1, characterized in that: The monitoring terminal includes a camera, a microphone, a motion sensor, and a positioning module.

6. The intelligent monitoring system for underground mining according to claim 5, characterized in that: The camera uses an anti-fog and dustproof lens and is coated with a hydrophobic and oleophobic coating; the positioning module uses a combination of inertial navigation and UWB positioning.

7. The intelligent monitoring system for underground mining according to claim 1, characterized in that: The hybrid transmission network includes a wireless mesh network and a wired transmission network.

8. The intelligent monitoring system for underground mining according to claim 1, characterized in that: The environmental sensor has a housing with a gradient functional coating, which includes: a bottom layer with a graphene / polytetrafluoroethylene composite anti-corrosion layer; a middle layer with a biomimetic micro-nano structure hydrophobic layer; and a top layer with a wear-resistant layer embedded with silicon carbide nanoparticles.

9. The intelligent monitoring system for underground mining according to claim 1, characterized in that: The environmental sensor is equipped with a self-cleaning device connected to it. When the data detected by the environmental sensor reaches a set value, the self-cleaning device automatically starts cleaning the environmental sensor.

10. The intelligent monitoring system for underground mining according to claim 9, characterized in that: The self-cleaning device includes a miniature air pump and an annular nozzle; the relay equipment is powered by a combination of downhole power and power collected by photovoltaic panels.

Images