Drone-robot dog cooperative multi-frequency sound wave mine heap cavity monitoring system and method

Abstract

The application discloses a UAV-mechanical dog cooperative multi-frequency sound wave ore heap hollow monitoring system and method, and relates to the field of ore heap hollow monitoring. The system comprises the following steps: a plurality of receiving sensors are uniformly arranged on the surface of the ore heap in the stoping space; a UAV platform and a mechanical dog are arranged in the stoping space; a multi-sensor integrated unit comprises a sound wave emission module, an RTK real-time dynamic positioning module, a laser radar module and a visual SLAM module; and a computing and analyzing terminal is connected with the mechanical dog, the multi-sensor integrated unit and the plurality of receiving sensors. The method comprises the following steps: after each ore drawing operation is completed, the mechanical dog is used to arrange a plurality of receiving sensors on the surface of the ore heap at the bottom of the stoping space in the form of a diamond grid; the UAV platform is controlled to sequentially traverse the arrangement points of the receiving sensors, and sound wave signals are emitted; the first wave arrival time, the amplitude attenuation rate and the spectral barycenter offset are obtained through the receiving sensors; and three-level early warning is carried out. The system and method can efficiently and accurately monitor the hollow condition of the ore heap.

Description

Technical Field

[0001] This invention belongs to the field of intelligent mining technology, specifically a drone-mechanical dog collaborative multi-frequency acoustic wave ore pile cavity monitoring system and method. Background Technology

[0002] As a core technology of traditional open-cut mining, shallow-hole stope mining has been widely used in metal and non-metal mines in my country. During the shallow-hole stope mining process, two-thirds of the mined ore needs to be temporarily retained in the stope as a temporary work platform. However, due to uneven ore size or the hinge effect of gravity, hidden cavities can easily form inside the ore pile. These cavities can lead to localized collapses of the working face, resulting in significant safety hazards such as personnel falling or being pulled into the ore discharge hopper during the discharge process. Traditional monitoring methods typically rely on manual tapping or drilling to detect cavities. These methods suffer from low detection efficiency, limited coverage, and the inability to detect cavities in real time. Furthermore, manual exploration deep into the ore pile carries extremely high risks. While existing technologies have incorporated drones or robots for mine inspection, several shortcomings remain: single-device functionality is limited; drones struggle to accurately deploy sensors on complex ore pile surfaces; and mechanical devices lack efficient mobility and obstacle avoidance capabilities. Acoustic detection technologies primarily use single-frequency signals, failing to consider both ore penetration depth and surface crack resolution. Static data processing algorithms, lacking dynamic threshold adjustments based on ore pile height, result in high false alarm rates. Furthermore, current systems lack tiered response mechanisms, failing to trigger differentiated emergency measures based on cavity risk levels. Recent research has attempted to integrate multi-sensor technologies, but key technical challenges remain, such as insufficient sensor fit, multi-frequency acoustic signal collaborative analysis, and adaptive construction of dynamic benchmark models. For example, fixed sensor networks struggle to adapt to dynamic changes on ore pile surfaces, and single-frequency acoustic technology cannot improve detection accuracy through a complementary mechanism of low-frequency ore penetration and high-frequency surface crack capture. Therefore, developing an intelligent ore pile cavity monitoring system integrating drone-machine dog collaborative operation, multi-frequency acoustic joint scanning, and dynamic threshold adaptive calculation is urgently needed to improve the inherent safety level of shallow-hole ore retention methods. Summary of the Invention

[0003] To address the problems existing in the prior art, this invention provides a UAV-mechanical dog collaborative multi-frequency acoustic ore pile cavity monitoring system and method. This system is highly intelligent and can accurately and efficiently detect cavities inside ore piles through the collaborative operation of UAVs and mechanical dogs, which helps to improve the safety factor of shallow hole ore-keeping operations. This method can efficiently and accurately monitor cavities inside ore piles and effectively solves the defects of low detection efficiency, poor real-time performance, and high risk of missed detection in the prior art.

[0004] To achieve the above objectives, the present invention provides a UAV-mechanical dog collaborative multi-frequency acoustic ore pile cavity monitoring system, including a stope, a mining space located in the stope, interstitial columns distributed on both sides of the stope, a top column located on the top of the stope, a bottom column located at the bottom of the stope, a skylight opened in the interstitial columns, a connecting passage connecting the skylight and the stope, a haulage roadway located at the bottom of the skylight and extending laterally along the stope, a ore discharge hopper opened in the stope chassis and connected to the haulage roadway, receiving sensors, a UAV platform, a mechanical dog, a multi-sensor integrated unit, and a computing and analysis terminal;

[0005] Multiple receiving sensors are evenly arranged on the surface of the ore pile in the mining space;

[0006] The unmanned aerial vehicle (UAV) platform is deployed in the mining space;

[0007] The mechanical dog is positioned in the recovery space;

[0008] The multi-sensor integrated unit includes a mounting platform, an acoustic wave emitting module, an RTK real-time dynamic positioning module, a lidar module, a visual SLAM module, a wireless communication module, and a controller. The controller is connected to the acoustic wave emitting module, the RTK real-time dynamic positioning module, the lidar module, the visual SLAM module, and the wireless communication module, respectively. The mounting platform is fixedly installed at the bottom of the UAV platform, and the acoustic wave emitting module, the RTK real-time dynamic positioning module, the lidar module, and the visual SLAM module are installed on the mounting platform at intervals. The wireless communication module and the controller are installed on the UAV platform. The controller is connected to the acoustic wave emitting module, the RTK real-time dynamic positioning module, the lidar module, the visual SLAM module, the UAV platform, and the wireless communication module, respectively.

[0009] The computing and analysis terminal is connected to the mechanical dog, the multi-sensor integration unit, and multiple receiving sensors.

[0010] As a preferred option, the computing and analysis terminal is an industrial computer.

[0011] As a preferred embodiment, the controller is a PLC controller.

[0012] In this invention, by equipping a multi-sensor integrated unit on a drone platform, it is possible to facilitate the drone's entry into the mining space for scanning and detection operations. This eliminates the need for manual intervention during the scanning and detection phase, effectively saving labor costs, reducing workload, and improving worker safety. By incorporating a sound wave transmitting module into the multi-sensor integrated unit, sound waves can be emitted towards the ore pile below. Receiving sensors positioned on the surface of the ore pile can then receive the echo signals, allowing for the detection of internal conditions within the ore pile and the rapid identification of cavities and surface cracks. The inclusion of an RTK real-time dynamic positioning module in the multi-sensor integrated unit facilitates real-time acquisition of the drone platform's position within the mining space, enabling convenient and efficient management by relevant personnel. Finally, the inclusion of a visual SLAM module in the multi-sensor integrated unit allows for real-time acquisition of surrounding image data during drone flight. This enables management personnel to monitor the entire operation process using image data and provides a clear view of the environmental conditions throughout the mining space. Furthermore, the combination of the RTK real-time dynamic positioning module and the visual SLAM module can significantly improve the UAV's automatic navigation capabilities, enabling the UAV platform to achieve high-precision and high-reliability autonomous flight in complex scenarios. By incorporating a LiDAR module into the multi-sensor integration unit, it is easy to obtain information about obstacles around the UAV platform, thus facilitating autonomous obstacle avoidance path planning during flight. The wireless communication module facilitates convenient communication and interaction between the multi-sensor integration unit and the computing and analysis terminal. Additionally, using the UAV platform as the carrier of the multi-sensor integration unit, and autonomously controlling its actions through a controller, allows for flexible adjustments to the detection methods. This enables unmanned acquisition of waveform data from various receiving sensors in the data acquisition space, significantly improving the intelligence level of the detection process, freeing up manpower, effectively reducing labor costs, improving operator safety, and greatly increasing detection efficiency. By connecting the computing and analysis terminal to multiple receiving sensors and a multi-sensor integration unit, the terminal can receive waveform data from these sensors in real time. This allows for precise identification of internal cavities and surface cracks in the ore pile through joint analysis of acoustic time-domain and frequency-domain characteristics. The inclusion of a robotic sensor facilitates the automated placement and retrieval of several modular receiving sensors on the ore pile surface. This eliminates the need for human intervention during sensor placement and retrieval, significantly improving operational safety.

[0013] The system is highly intelligent and can accurately and efficiently detect cavities inside ore piles through the collaborative operation of inorganic humans and robotic dogs, which helps to improve the safety factor of shallow hole ore deposit operations.

[0014] This invention also provides a method for monitoring cavities in ore piles using a drone-mechanical dog collaborative multi-frequency acoustic wave system, comprising the following steps:

[0015] Step 1: When using the shallow hole ore-holding method in metal and non-metal mines, ore is discharged through a ore-discharging funnel, releasing 1 / 3 of the mined ore each time, leaving 2 / 3 of the ore as a temporary working platform for the next drilling and blasting operation, and forming a mining space above the temporary working platform.

[0016] Step 2: After each ore release operation is completed, the mechanical dog is deployed in the mining space. The mechanical dog is used to arrange several receiving sensors in a diamond grid on the surface of the ore pile at the bottom of the mining space.

[0017] Step 3: Control the UAV platform to sequentially traverse the deployment points of each receiving sensor, and hover above each deployment point for an initial detection cycle. In each initial detection cycle, a 20kHz standard acoustic pulse is emitted downward through the acoustic wave emission module, and the receiving sensor below receives the 20kHz acoustic signal in real time. The signal is then transmitted to the computing and analysis terminal via wireless communication. The computing and analysis terminal records the arrival time and amplitude attenuation rate of the first wave at each receiving sensor, and generates an initial acoustic wave propagation parameter matrix by combining it with the accumulation height of the remaining ore.

[0018] Step 4: Control the UAV platform again to sequentially traverse the deployment points of each receiving sensor, and hover above each deployment point for a formal detection cycle. In each formal detection cycle, the UAV platform will sequentially transmit a 20kHz main frequency detection signal and a 50kHz auxiliary frequency detection signal downwards through the acoustic wave transmission module. At the same time, the receiving sensors below will synchronously receive the 20kHz and 50kHz acoustic wave signals and transmit them to the computing and analysis terminal via wireless communication. The computing and analysis terminal will record the waveform data of each receiving sensor.

[0019] Step 5: The calculation and analysis terminal calculates the value of each detection point based on the 20kHz waveform data received by each receiving sensor. ,in, = (Transmission distance - Reception distance) / First wave arrival time, calculate for each detection point ,in, =20×log10(reference amplitude / measured amplitude) / propagation distance; simultaneously, the spectral centroid change value is obtained based on the 50kHz waveform data received by each receiving sensor. ;

[0020] Step 6: Calculate and analyze the longitudinal wave velocity descent rate based on the 20kHz waveform data received by each receiving sensor and the formula (1). The rate of increase of the attenuation coefficient is calculated according to formula (2). Meanwhile, based on the 50kHz waveform data received by each receiving sensor, the spectral centroid offset is calculated according to formula (3). The fracture density is calculated according to formula (4). ;

[0021] (1);

[0022] (2);

[0023] (3);

[0024] (4);

[0025] In the formula, These are calibration coefficients; , and It was obtained based on a benchmark model of a dense and crack-free ore pile;

[0026] Step 7: When any parameter at a detection point exceeds the standard, a 0.8m radius area is delineated as the key monitoring area centered on that detection point. Based on the degree to which the parameter deviates from the benchmark value, the risk level of the ore pile is determined, and a graded early warning action is taken.

[0027] When 5%≤ <10% and 20% ≤ When <50%, or when When this occurs, a Level 1 risk is identified, and a Level 1 warning is issued;

[0028] When 10%≤ <15% and 50% ≤ When <80%, or when When a level-two risk is identified, a level-two warning is issued.

[0029] When 15% < And 80% < At that time, or when When a level three risk is identified, a level three warning is issued.

[0030] Step 8: After eliminating all safety risks through proper emergency response measures, reposition the mechanical dog in the mining space and retrieve the various receiving sensors on the surface of the ore pile one by one.

[0031] Step 9: Proceed to the next rock drilling and blasting operation.

[0032] As a preferred option, the height of the mining space formed in step one is 2 to 3 meters.

[0033] As a preferred embodiment, in step two, the receiving sensor is a piezoelectric receiving sensor.

[0034] As a preferred embodiment, in step two, a receiving sensor is arranged at each of the four corners, the inner center, and the midpoint of each side of each rhomboid region.

[0035] As a preferred embodiment, in step three, during the flight of the UAV platform in the recovery space, the RTK real-time dynamic positioning module is used to perceive the position information of the UAV platform in real time and send it to the controller. The controller then sends it to the computing and analysis terminal through the wireless communication module. At the same time, the visual SLAM module is used to collect image data around the UAV platform in real time and send it to the controller. The controller then sends it to the computing and analysis terminal through the wireless communication module. Meanwhile, the controller uses the lidar module to obtain the obstacle situation around the UAV platform in real time and dynamically adjusts the flight path according to the surrounding obstacle situation.

[0036] This invention provides a UAV-mechanical dog collaborative multi-frequency acoustic wave method for monitoring cavities in ore piles. The method utilizes a mechanical dog to actively deploy modular piezoelectric sensors on the ore pile surface in a diamond grid, forming a receiving sensor array. The UAV then transmits a composite acoustic wave signal with a primary frequency of 20kHz and an auxiliary frequency of 50kHz along a preset path. The waveform data from the receiving sensors is used to obtain the first wave arrival time, amplitude attenuation rate, and spectral centroid shift. Through the coordinated technology of directional acoustic wave transmission and echo reception, and employing multi-frequency composite acoustic wave scanning technology (primary frequency 20kHz, auxiliary frequency 50kHz), low-frequency signals penetrate the ore body to detect internal cavities, while high-frequency signals capture surface cracks. Then, by combining the time-domain and frequency-domain characteristics of the acoustic waves, and based on real-time calculation of the longitudinal wave velocity and attenuation coefficient, the internal cavities and surface cracks of the ore pile are accurately identified. Finally, a three-level early warning mechanism triggers different levels of risk warnings to facilitate corresponding emergency measures based on the warning level.

[0037] This invention employs multi-frequency acoustic wave collaborative scanning technology, which solves the problem of traditional single-frequency acoustic waves being unable to simultaneously achieve both depth and resolution through a complementary mechanism of low-frequency penetration of the ore body and high-frequency capture of surface fractures. By integrating UAV-mechanical dog collaborative operation, multi-source data fusion, and intelligent response, it overcomes the limitations of low efficiency and poor accuracy of manual exploration, realizing real-time dynamic monitoring and risk classification management of ore pile cavities. It features high precision, high safety, and strong engineering practicality.

[0038] Compared with the prior art, the present invention has the following advantages:

[0039] 1) High-efficiency collaborative detection. This invention utilizes a robotic dog to actively deploy piezoelectric sensors in a diamond grid pattern, combined with a drone emitting multi-frequency composite acoustic signals, to achieve synchronous dynamic scanning of the surface and internal structure of the ore pile. Compared to traditional manual tapping or drilling, the system can complete full-area monitoring within a 30-second cycle, improving detection efficiency by over 80%, and avoiding the operational risks of manually venturing deep into the ore pile.

[0040] 2) Precise Multi-Frequency Acoustic Wave Identification. Employing a collaborative scanning technique using a primary frequency of 20kHz (low-frequency penetration of the ore body) and an auxiliary frequency of 50kHz (high-frequency capture of surface cracks), combined with joint analysis of acoustic time-domain and frequency-domain characteristics, it can simultaneously detect internal cavities and surface cracks in the ore pile, significantly improving detection resolution. Through calculation of the spectral centroid offset, it effectively distinguishes between abnormal ore density and external interference signals, reducing the false positive rate.

[0041] 3) Tiered intelligent response. The system integrates drone-mechanical dog collaborative operation with multi-source data fusion to provide three-level early warning, offering high-precision and real-time safety assurance for shallow hole mining methods. This significantly reduces the risk of personnel fall accidents and has strong engineering practicality.

[0042] This method can efficiently and accurately monitor the cavities inside ore piles, effectively solving the shortcomings of existing technologies such as low detection efficiency, poor real-time performance, and high risk of missed detection. It is especially suitable for the shallow-hole ore-holding method in metal and non-metal mines, enabling dynamic identification and graded early warning of cavities inside ore piles, effectively ensuring the safety of mining personnel. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the shallow-hole ore-holding monitoring system of the present invention;

[0044] Figure 2 This is a schematic diagram of the route of the rhomboid grid in this invention;

[0045] Figure 3 This is a schematic diagram showing the completed state of the receiving sensor arrangement in this invention;

[0046] Figure 4This is a partial schematic diagram of the receiving sensor in the rhomboid grid in this invention;

[0047] Figure 5 This is a top view of the unmanned aerial vehicle platform in this invention;

[0048] Figure 6 This is a front view of the unmanned aerial vehicle platform in this invention;

[0049] Figure 7 This is a schematic diagram of the structure of the multi-sensor integrated unit in this invention.

[0050] Figure 8 This is a block diagram of the control section in this invention.

[0051] In the diagram: 1. Return airway, 2. Top pillar, 3. Shaft, 4. Connecting passage, 5. Interstitial pillar, 6. Ore storage, 7. Bottom pillar, 8. Ore discharge funnel, 9. Along-vein transport roadway, 10. Unmined ore, 11. Mining space, 12. Diamond grid, 13. Receiving sensor, 14. UAV platform, 15. Multi-sensor integrated unit, 15-1. Acoustic wave emission module, 15-2. RTK real-time dynamic positioning module, 15-3. LiDAR module, 15-4. Visual SLAM module, 15-5. Installation platform, 16. Minehouse, 17. Mechanical dog. Detailed Implementation

[0052] The invention will now be further described with reference to the accompanying drawings.

[0053] like Figures 1 to 8 As shown, the present invention provides a UAV-mechanical dog collaborative multi-frequency acoustic ore pile cavity monitoring system, including a stope 16, a mining space 11 located in the stope 16, interstitial columns 5 distributed on both sides of the stope 16, a top column 2 located at the top of the stope 16, a bottom column 7 located at the bottom of the stope 16, a well 3 opened in the interstitial column 5, a connecting passage 4 connecting the well 3 and the stope 16, a haulage roadway 9 located at the bottom of the well 3 and extending laterally along the stope 16, a ore discharge hopper 8 opened in the chassis of the stope 16 and connected to the haulage roadway 9, a receiving sensor 13, a UAV platform 14, a mechanical dog 17, a multi-sensor integration unit 15, and a computing and analysis terminal;

[0054] Among them, the section above the top pillar 2 is the return airway 1, the section above the mining space 11 is the unmined ore 10, and the section below the mining space 11 and between the ore discharge funnel 8 is the stored ore 6.

[0055] Multiple receiving sensors 13 are evenly arranged on the surface of the ore pile in the mining space 11; preferably, the sampling frequency of the receiving sensors 13 is 1MHz.

[0056] The unmanned aerial vehicle platform 14 is arranged in the mining space 11;

[0057] The mechanical dog 17 is arranged in the mining space 11;

[0058] The multi-sensor integrated unit 15 includes a mounting platform 15-5, an acoustic wave emitting module 15-1, an RTK real-time dynamic positioning module 15-2, a lidar module 15-3, a visual SLAM module 15-4, a wireless communication module, and a controller. The controller is connected to the acoustic wave emitting module 15-1, the RTK real-time dynamic positioning module 15-2, the lidar module 15-3, the visual SLAM module 15-4, and the wireless communication module. The mounting platform 15-5 is fixedly mounted on the bottom of the UAV platform 14. The acoustic wave emitting module 15-1, the RTK real-time dynamic positioning module 15-2, the lidar module 15-3, and the visual SLAM module 15-4 are mounted on the mounting platform 15-5 at intervals. The wireless communication module and the controller are mounted on the UAV platform 14. The controller is connected to the acoustic wave emitting module 15-1, the RTK real-time dynamic positioning module 15-2, the lidar module 15-3, the visual SLAM module 15-4, the UAV platform 14, and the wireless communication module.

[0059] As a preferred embodiment, the acoustic wave transmitting module 15-1 has a frequency range of 10-60kHz and a power of 50W. It can transmit acoustic wave signals at a main frequency of 20kHz and an auxiliary frequency of 50kHz. The RTK real-time dynamic positioning module 15-2 and the lidar module 15-3 form a dual-redundant navigation system, which can facilitate the dynamic planning of obstacle avoidance paths. At the same time, it can further facilitate the dynamic adjustment of flight altitude and path by constructing a three-dimensional point cloud model of the mine pile.

[0060] The computing and analysis terminal is connected to the multi-sensor integration unit 15 and multiple receiving sensors 13.

[0061] The mechanical dog 17 is used to arrange the receiving sensor 13 on the surface of the ore pile in a set layout, and also to collect the receiving sensor 13 located on the surface of the ore pile. Preferably, the mechanical dog 17 has a built-in pressure feedback sensing unit, which can ensure that the placed receiving sensor 13 can fully fit the surface of the ore pile. At the same time, it supports a torque adaptive algorithm and can cross obstacles of 30cm.

[0062] As a preferred embodiment, an alarm is also included, which is connected to a computing and analysis terminal and is used to execute alarm actions according to the control of the computing and analysis terminal.

[0063] As a preferred option, UWB base stations can be deployed in the mining space 11, enabling the receiving sensor 13 and the UAV platform 14 to establish a more reliable wireless communication connection with the computing and analysis terminal through the UWB base stations.

[0064] As a preferred option, the computing and analysis terminal is an industrial computer.

[0065] As a preferred embodiment, the controller is a PLC controller.

[0066] In this invention, by equipping a multi-sensor integrated unit on a drone platform, it is possible to facilitate the drone's entry into the mining space for scanning and detection operations. This eliminates the need for manual intervention during the scanning and detection phase, effectively saving labor costs, reducing workload, and improving worker safety. By incorporating a sound wave transmitting module into the multi-sensor integrated unit, sound waves can be emitted towards the ore pile below. Receiving sensors positioned on the surface of the ore pile can then receive the echo signals, allowing for the detection of internal conditions within the ore pile and the rapid identification of cavities and surface cracks. The inclusion of an RTK real-time dynamic positioning module in the multi-sensor integrated unit facilitates real-time acquisition of the drone platform's position within the mining space, enabling convenient and efficient management by relevant personnel. Finally, the inclusion of a visual SLAM module in the multi-sensor integrated unit allows for real-time acquisition of surrounding image data during drone flight. This enables management personnel to monitor the entire operation process using image data and provides a clear view of the environmental conditions throughout the mining space. Furthermore, the combination of the RTK real-time dynamic positioning module and the visual SLAM module can significantly improve the UAV's automatic navigation capabilities, enabling the UAV platform to achieve high-precision and high-reliability autonomous flight in complex scenarios. By incorporating a LiDAR module into the multi-sensor integration unit, it is easy to obtain information about obstacles around the UAV platform, thus facilitating autonomous obstacle avoidance path planning during flight. The wireless communication module facilitates convenient communication and interaction between the multi-sensor integration unit and the computing and analysis terminal. Additionally, using the UAV platform as the carrier of the multi-sensor integration unit, and autonomously controlling its actions through a controller, allows for flexible adjustments to the detection methods. This enables unmanned acquisition of waveform data from various receiving sensors in the data acquisition space, significantly improving the intelligence level of the detection process, freeing up manpower, effectively reducing labor costs, improving operator safety, and greatly increasing detection efficiency. By connecting the computing and analysis terminal to multiple receiving sensors and a multi-sensor integration unit, the terminal can receive waveform data from these sensors in real time. This allows for precise identification of internal cavities and surface cracks in the ore pile through joint analysis of acoustic time-domain and frequency-domain characteristics. The inclusion of a robotic sensor facilitates the automated placement and retrieval of several modular receiving sensors on the ore pile surface. This eliminates the need for human intervention during sensor placement and retrieval, significantly improving operational safety.

[0067] The system is highly intelligent and can accurately and efficiently detect cavities inside ore piles through the collaborative operation of inorganic humans and robotic dogs, which helps to improve the safety factor of shallow hole ore deposit operations.

[0068] This invention also provides a method for monitoring cavities in ore piles using a drone-mechanical dog collaborative multi-frequency acoustic wave system, comprising the following steps:

[0069] Step 1: When using the shallow hole ore-holding method for mining operations in metal and non-metal mines, ore is discharged through the ore discharge funnel 8, releasing 1 / 3 of the ore mined each time, leaving 2 / 3 of the remaining ore 6 as a temporary work platform for the next rock drilling and blasting operation, and forming a mining space 11 above the temporary work platform.

[0070] Step 2: After each ore release operation is completed, the mechanical dog 17 is placed in the mining space 11. The mechanical dog 17 is used to arrange several receiving sensors 13 in a diamond grid on the surface of the ore pile at the bottom of the mining space 11.

[0071] Step 3: Control the UAV platform 14 to sequentially traverse the placement points of each receiving sensor 13, and hover above each placement point for an initial detection cycle. In each initial detection cycle, the sound wave transmitting module 15-1 emits a 20kHz standard sound wave pulse downwards, and the receiving sensor 13 below receives the 20kHz sound wave signal in real time, and sends it to the computing and analysis terminal via wireless communication. The computing and analysis terminal records the arrival time and amplitude attenuation rate of the first wave of each receiving sensor 13, and generates an initial sound wave propagation parameter matrix by combining it with the accumulation height of the stored ore 6.

[0072] As a preferred method, the initial detection period is 10ms. The standard acoustic pulse is a sine wave.

[0073] Step 4: Control the UAV platform 14 again to sequentially traverse the deployment points of each receiving sensor 13, and hover above each deployment point for a formal detection cycle. In each formal detection cycle, the sound wave transmitting module 15-1 sequentially transmits a 20kHz main frequency detection signal and a 50kHz auxiliary frequency detection signal downwards. At the same time, the receiving sensor 13 below synchronously receives the 20kHz and 50kHz sound wave signals and transmits them to the computing and analysis terminal via wireless communication. The computing and analysis terminal records the waveform data of each receiving sensor 13.

[0074] As a preferred option, the formal detection period is 10ms.

[0075] Step 5: The calculation and analysis terminal calculates the value of each detection point based on the 20kHz waveform data received by each receiving sensor 13. ,in, = (Transmission distance - Reception distance) / First wave arrival time, calculate for each detection point ,in, =20×log10(reference amplitude / measured amplitude) / propagation distance; Simultaneously, based on the spectral centroid change value of the 50kHz waveform data received by each receiving sensor 13, the following is obtained: ;

[0076] Step 6: The terminal calculates the longitudinal wave velocity descent rate based on the 20kHz waveform data received by each receiving sensor 13 according to formula (1). The rate of increase of the attenuation coefficient is calculated according to formula (2). Simultaneously, based on the 50kHz waveform data received by each receiving sensor 13, the spectral centroid offset is calculated according to formula (3). The fracture density is calculated according to formula (4). ;

[0077] (1);

[0078] (2);

[0079] (3);

[0080] (4);

[0081] In the formula, The calibration coefficient is determined experimentally and can be set to 0.005 mm. -1 ; , and It was obtained through a detection experiment based on a dense and crack-free ore pile benchmark model;

[0082] Step 7: When any parameter at a detection point exceeds the standard, a 0.8m radius area is delineated as the key monitoring area centered on that detection point. Based on the degree to which the parameter deviates from the benchmark value, the risk level of the ore pile is determined, and a graded early warning action is taken.

[0083] When 5%≤ <10% and 20% ≤ When <50%, or when When a minor crack is detected, a Level 1 risk is identified and a Level 1 warning is issued. When a Level 1 warning is issued, an alarm is triggered to emit an audible and visual alarm signal to notify relevant personnel to further check the actual situation of the monitored area.

[0084] When 10%≤ <15% and 50% ≤ When <80%, or when When there is moderate cracking, a level 2 risk is identified and a level 2 warning is issued. When a level 2 warning is issued, relevant personnel will spray quick-setting material on the monitored area to solidify the surface of the mine pile in the risk area and give it the set bearing capacity.

[0085] When 15% < And 80% < At that time, or when When a severe crack occurs, a level 3 risk is identified and a level 3 warning is issued. When a level 3 warning is issued, relevant personnel are guided to evacuate along a safe path, and emergency support measures are initiated.

[0086] Step 8: After eliminating all safety risks through proper emergency response measures, the mechanical dog 17 is repositioned in the mining space 11 to retrieve the various receiving sensors 13 on the surface of the ore pile.

[0087] Step 9: Proceed to the next rock drilling and blasting operation.

[0088] As a preferred option, in step one, the height of the formed mining space 11 is 2~3m.

[0089] As a preferred embodiment, in step two, the receiving sensor 13 is a piezoelectric receiving sensor.

[0090] As a preferred embodiment, in step two, a receiving sensor 13 is arranged at each of the four corners, the inner center, and the midpoint of each side of each rhomboid region.

[0091] As a preferred embodiment, in step three, during the flight of the UAV platform 14 in the recovery space 11, the RTK real-time dynamic positioning module 15-2 is used to perceive the position information of the UAV platform 14 in real time and send it to the controller. The controller then sends it to the computing and analysis terminal through the wireless communication module. At the same time, the visual SLAM module 15-4 is used to collect image data around the UAV platform 14 in real time and send it to the controller. The controller then sends it to the computing and analysis terminal through the wireless communication module. Meanwhile, the controller uses the lidar module 15-3 to obtain the obstacle situation around the UAV platform 14 in real time and dynamically adjusts the flight path according to the surrounding obstacle situation.

[0092] This invention employs multi-frequency acoustic wave collaborative scanning technology, which solves the problem of traditional single-frequency acoustic waves being unable to simultaneously achieve both depth and resolution through a complementary mechanism of low-frequency penetration of the ore body and high-frequency capture of surface fractures. By integrating UAV-mechanical dog collaborative operation, multi-source data fusion, and intelligent response, it overcomes the limitations of low efficiency and poor accuracy of manual exploration, realizing real-time dynamic monitoring and risk classification management of ore pile cavities. It features high precision, high safety, and strong engineering practicality.

[0093] This method can efficiently and accurately monitor the cavities inside ore piles, effectively solving the shortcomings of existing technologies such as low detection efficiency, poor real-time performance, and high risk of missed detection. It is especially suitable for the shallow-hole ore-holding method in metal and non-metal mines, enabling dynamic identification and graded early warning of cavities inside ore piles, effectively ensuring the safety of mining personnel.

Claims

1. A UAV-mechanical dog cooperative multi-frequency acoustic wave mine heap cavity monitoring system, comprising a mine chamber (16), a stoping space (11) located in the mine chamber (16), intercolumns (5) distributed on both side edges of the mine chamber (16), a top column (2) located at the top of the mine chamber (16), a bottom column (7) located at the bottom of the mine chamber (16), a raise (3) opened in the intercolumn (5), a crossway (4) communicating the raise (3) and the mine chamber (16), a along-vein transportation main roadway (9) located at the bottom of the raise (3) and extending transversely along the mine chamber (16), and a ore chute (8) opened in the bottom plate of the mine chamber (16) and communicating with the along-vein transportation main roadway (9), characterized in that, It also includes receiving sensors (13), unmanned aerial vehicle platforms (14), mechanical dogs (17), multi-sensor integrated units (15) and computing analysis terminals. The plurality of receiving sensors (13) are evenly arranged on the ore heap surface in the stoping space (11). The unmanned aerial vehicle platform (14) is arranged in the stoping space (11). The mechanical dog (17) is arranged in the stoping space (11). The multi-sensor integrated unit (15) includes a mounting platform (15-5), a sound wave emission module (15-1), an RTK real-time dynamic positioning module (15-2), a laser radar module (15-3), a visual SLAM module (15-4), a wireless communication module and a controller, the controller is connected with the sound wave emission module (15-1), the RTK real-time dynamic positioning module (15-2), the laser radar module (15-3), the visual SLAM module (15-4) and the wireless communication module respectively; the mounting platform (15-5) is fixedly installed at the bottom of the unmanned aerial vehicle platform (14), the sound wave emission module (15-1), the RTK real-time dynamic positioning module (15-2), the laser radar module (15-3) and the visual SLAM module (15-4) are installed on the mounting platform (15-5) at intervals; the wireless communication module and the controller are installed on the unmanned aerial vehicle platform (14); the controller is connected with the sound wave emission module (15-1), the RTK real-time dynamic positioning module (15-2), the laser radar module (15-3), the visual SLAM module (15-4), the unmanned aerial vehicle platform (14) and the wireless communication module respectively; The computing analysis terminal is connected with the mechanical dog (17), the multi-sensor integrated unit (15) and the plurality of receiving sensors (13) respectively.

2. The unmanned aerial vehicle - robot dog cooperative multi-frequency acoustic wave mine heap cavity monitoring system according to claim 1, characterized in that, The computing analysis terminal is an industrial computer.

3. The unmanned aerial vehicle - robot dog cooperative multi-frequency acoustic wave mine heap cavity monitoring system according to claim 2, characterized in that, The controller is a PLC controller.

4. A method for monitoring mine heap cavities by drone-mechanical dog cooperation multi-frequency sound waves, using the drone-mechanical dog cooperation multi-frequency sound wave mine heap cavity monitoring system of claim 1, characterized in that, The method comprises the following steps: Step one: when the metal and non-metal mine uses the shallow hole shrinkage method for mining operation, the ore is discharged through the ore discharge funnel (8), 1 / 3 of the ore mined each time is discharged, 2 / 3 of the remaining ore (6) is left as a temporary workbench for the next rock drilling and blasting work, and a stoping space (11) is formed above the temporary workbench; Step two: after completing the ore discharge operation each time, the mechanical dog (17) is arranged in the stoping space (11), and the mechanical dog (17) is used to arrange a plurality of receiving sensors (13) in a diamond grid manner on the ore heap surface at the bottom of the stoping space (11). Step three: control the unmanned aerial vehicle platform (14) to traverse each receiving sensor (13) arrangement point in turn, and hover above each arrangement point for an initial detection period, in each initial detection period, a 20kHz standard sound wave pulse is emitted downward by the sound wave emission module (15-1), and a 20kHz sound wave signal is received in real time by the receiving sensor (13) below and sent to the computing analysis terminal through wireless communication, the computing analysis terminal records the time of the first wave arrival and the amplitude attenuation rate of each receiving sensor (13), and generates an initial sound wave propagation parameter matrix combined with the accumulation height of the remaining ore (6); Step four: control the unmanned aerial vehicle platform (14) to traverse each receiving sensor (13) arrangement point in turn again, and hover above each arrangement point for a formal detection period, in each formal detection period, a 20kHz main frequency detection signal and a 50kHz auxiliary frequency detection signal are emitted downward in turn by the sound wave emission module (15-1), at the same time, a 20kHz sound wave signal and a 50kHz sound wave signal are received synchronously by the receiving sensor (13) below, and sent to the computing analysis terminal through wireless communication, the computing analysis terminal records the waveform data of each receiving sensor (13); Step five: The analysis terminal calculates the 20 kHz waveform data received by each receiving sensor (13) to calculate the wherein, = (transmitting distance - receiving distance) / first wave arrival time, the 50 kHz waveform data received by each receiving sensor (13) to calculate the wherein, = 20 x log10 (reference amplitude / measured amplitude) / propagation distance; and the spectrum barycenter change value of the 50 kHz waveform data received by each receiving sensor (13) to obtain ; Step six: the analysis terminal calculates the 20 kHz waveform data received by each receiving sensor (13), calculates the longitudinal wave speed drop rate according to formula (1) , calculates the attenuation coefficient rise rate according to formula (2) ; at the same time, based on the 50 kHz waveform data received by each receiving sensor (13), calculates the spectral barycenter offset according to formula (3) , and calculates the fracture density according to formula (4) ; （1）； （2）； （3）； （4）； wherein is a calibration factor; , and are obtained based on a dense and fissure-free stockpile reference model implementation; Step seven: when any one of the detection point parameters exceeds the standard, a 0.8m radius area centered on the detection point is designated as a key monitoring area, the ore pile risk level is determined according to the degree of parameter deviation from the reference value, and a graded early warning action is taken; When 5%≤ < 10% and 20%≤ < 50%, or when a first level risk is determined to occur, and a first level warning is given. When 10%≤ < 15% and 50%≤ < 80%, or when a secondary risk is determined to occur, and a secondary warning is given. When 15% < And 80% < At that time, or when When a level three risk is identified, a level three warning is issued. Step eight: after all safety risks are eliminated by correct emergency disposal means, the mechanical dog (17) is arranged again in the stoping space (11) to recover the receiving sensors (13) on the surface of the ore pile one by one; Step nine: carry out the next rock drilling and blasting operation.

5. The method of claim 4, wherein the method further comprises: In step one, the height of the stoping space (11) formed is 2-3m.

6. The unmanned aerial vehicle - mechanical dog cooperative multi-frequency sound wave mine heap cavity monitoring method according to claim 5, characterized in that, In step two, the receiving sensor (13) is a piezoelectric receiving sensor.

7. The method of claim 5, wherein the method further comprises: In step two, one receiving sensor (13) is arranged at each corner of the diamond-shaped area, the internal center and the midpoint of each side.

8. The method for monitoring cavities in a mine pile using a drone-mechanical dog collaborative multi-frequency acoustic wave method according to claim 5, characterized in that, In step three, during the flight of the unmanned aerial vehicle platform (14) in the stoping space (11), the RTK real-time dynamic positioning module (15-2) is used to sense the position information of the unmanned aerial vehicle platform (14) in real time and send it to the controller, the controller sends it to the computing analysis terminal through the wireless communication module, at the same time, the visual SLAM module (15-4) is used to collect image data around the unmanned aerial vehicle platform (14) in real time and send it to the controller, the controller sends it to the computing analysis terminal through the wireless communication module, at the same time, the controller uses the laser radar module (15-3) to obtain the obstacle situation around the unmanned aerial vehicle platform (14) in real time, and dynamically adjusts the flight path according to the obstacle situation around.

Images