Self-adaptive control method for frequency-amplitude alternating hydraulic cavitation of long directional borehole in coal mine underground

Abstract

The application discloses a coal mine underground directional long borehole frequency amplitude alternating hydraulic cavity forming self-adaptive control method and relates to the technical field of coal mine underground engineering, which comprises the following steps: presetting cavity forming equipment and a plurality of source sensors and adjusting the cavity forming equipment and the plurality of source sensors to a cavity forming target area; detecting and collecting the target cavity forming area, generating initial detection data, and storing and preprocessing the initial detection data to obtain preprocessed initial detection data; generating pump body output working conditions by using cavity forming target parameters and the preprocessed initial detection data, extracting a plurality of source real-time monitoring data; constructing a deep learning algorithm and a quantitative calculation model, analyzing the plurality of source real-time monitoring data, comparing analysis results with preset cavity forming target parameters; adjusting parameter adjustment suggestions by using a self-adaptive parameter adjustment mechanism; and inputting optimized pump control parameters into the plurality of source real-time monitoring data, the analysis comparison and parameter adjustment steps, and monitoring and adjusting to obtain a stable cavity. The application solves the problems of low efficiency and uncontrolled cavity size caused by fixed cavity forming parameters.

Description

Technical Field

[0001] This invention relates to the field of underground coal mine engineering technology, and more specifically, to an adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in underground coal mines. Background Technology

[0002] During underground coal mining, directional long borehole hydraulic cavity-making technology is one of the key technologies for achieving efficient gas extraction and reducing the risk of coal and gas outbursts. By using hydraulic cavity-making to form cavities of a certain size around the borehole, the exposed area of ​​the coal body can be expanded and the permeability of the coal seam can be improved, creating favorable conditions for gas extraction.

[0003] Currently, hydraulic cavity-making technology for directional long boreholes in coal mines mainly utilizes high-pressure water jets with fixed parameters for cavity creation. This presents several problems: First, it cannot dynamically adjust the water jet parameters based on changes in coal hardness. Cavity creation efficiency is low when the coal is hard, while excessive expansion of the cavity can lead to borehole collapse when the coal is soft. Second, there is a lack of real-time monitoring and feedback on the cavity morphology. Cavity size relies on manual judgment, resulting in poor accuracy and difficulty in meeting the cavity-making needs of different mining scenarios. Third, traditional cavity-making systems lack adaptive control capabilities, requiring frequent manual intervention to adjust equipment parameters. This increases labor intensity and may lead to decreased cavity quality or safety accidents due to untimely response.

[0004] There is currently no effective solution to the problems in the relevant technologies. Summary of the Invention

[0005] To address the problems in related technologies, this invention proposes an adaptive control method for hydraulic cavity creation with alternating frequency amplitude in directional long boreholes in coal mines, in order to overcome the aforementioned technical problems existing in existing related technologies.

[0006] Therefore, the specific technical solution adopted by the present invention is as follows:

[0007] In a first aspect, the present invention proposes an adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines, comprising:

[0008] S1. Place the pre-set cavity-making equipment and multi-source sensors in the target cavity-making area in the coal mine to complete the debugging of the cavity-making equipment and multi-source sensors.

[0009] S2. Use multi-source sensors to detect and collect data on the target cavitation area, generate initial detection data from the detection and collection results, and store and preprocess the initial detection data to obtain preprocessed initial detection data.

[0010] S3. Based on the preset target parameters for creating the cavity and the pre-processed initial detection data, set the output parameters for the variable frequency pulsating pump to generate pump output conditions that meet the target coal body conditions.

[0011] S4. Use multi-source sensors to extract multi-source real-time monitoring data during the pump output process;

[0012] S5. Based on the coal hardness calculation algorithm, pump pressure regulation mechanism, cavity formation efficiency prediction algorithm and cavity volume correction mechanism, a deep learning algorithm and a quantitative calculation model are constructed. The deep learning algorithm and quantitative calculation model are used to analyze multi-source real-time monitoring data. The analysis results are compared with the preset cavity formation target parameters to obtain parameter adjustment suggestions.

[0013] S6. The parameter adjustment suggestions are adjusted using an adaptive parameter adjustment mechanism to obtain optimized pump control parameters;

[0014] S7. Input the optimized pump control parameters back into steps S4-S6 to monitor and adjust the cavity-making parameters, and obtain a stable cavity-making chamber that meets the target size and efficiency requirements, thereby achieving the accurate completion of directional long-hole hydraulic cavity-making operations and the complete retention of operation process data.

[0015] Furthermore, the pre-set cavity-creating equipment and multi-source sensors are placed in the target cavity-creating area underground in the coal mine. This process, including the commissioning of the cavity-creating equipment and multi-source sensors, involves:

[0016] S11, the ultrasonic detector, the water jet pressure sensor and the water jet waveform acquisition device are multi-source sensors;

[0017] S12. Connect and adjust to the target area of ​​the hole-making device and multi-source sensor through the preset hole-making device;

[0018] S13. Use the terminal to test the preset hole-making equipment and multi-source sensors to complete the debugging of the hole-making equipment and multi-source sensors.

[0019] Furthermore, multi-source sensors are used to detect and collect data on the target cavitation area. The results of the detection and collection are used to generate initial detection data, which is then stored and preprocessed to obtain preprocessed initial detection data, including:

[0020] S21. Using an ultrasonic detector, perform multi-angle rotational scanning of the cavity-forming area to obtain reflected echo data, and generate a three-dimensional image of the initial shape of the coal body based on the sound wave propagation time and scanning angle.

[0021] S22. Use a water jet pressure sensor and a water jet waveform acquisition device to acquire and record the initial reference pressure and environmental waveform data respectively.

[0022] S23. Transmit the three-dimensional image, reference pressure, and environmental waveform data as initial detection data to the data processing terminal.

[0023] S24. Using a data processing terminal, the initial probe data is classified and stored. Then, outlier removal, noise smoothing, and standardization preprocessing are performed sequentially to obtain the preprocessed initial probe data.

[0024] Furthermore, based on the preset target parameters for creating the cavity and the pre-processed initial detection data, the output parameters of the variable frequency pulsating pump are set to generate pump output conditions that meet the target coal body conditions, including:

[0025] S31. Based on the preprocessed initial detection data and the preset cavity-forming target parameters, determine the initial hardness type of the target coal body;

[0026] S32. Based on the determined initial coal body hardness type, set the corresponding preset output parameter group;

[0027] S33. Set the initial output parameters of the variable frequency pulsating pump according to the corresponding preset output parameter group;

[0028] S34. Start the set variable frequency pulsating pump to form a water jet with alternating frequency and amplitude to impact the coal body and obtain the pump output condition of the target coal body.

[0029] Furthermore, the multi-source real-time monitoring data extracted during the pump output process using multi-source sensors includes:

[0030] S41. Trigger the ultrasonic detector according to the preset first cycle to perform scanning work on the cavity and generate real-time cavity morphology data.

[0031] S42. Drive the water jet pressure sensor according to the preset second sampling frequency to continuously monitor the water jet pressure and generate pressure fluctuation data.

[0032] S43. Based on the preset third cycle trigger water jet waveform acquisition device, acquire the water jet reflection waveform and extract the preset waveform feature parameters from it.

[0033] S44. Real-time cavity morphology data, pressure fluctuation data, and waveform characteristic parameters are treated as multi-source real-time monitoring data and transmitted to the data processing terminal in real time.

[0034] Furthermore, based on the coal hardness calculation algorithm, pump pressure regulation mechanism, cavity formation efficiency prediction algorithm, and cavity volume correction mechanism, a deep learning algorithm and quantitative calculation model are constructed. This deep learning algorithm and quantitative calculation model are then used to analyze multi-source real-time monitoring data. The analysis results are compared with preset cavity formation target parameters, and parameter adjustment suggestions are derived, including:

[0035] S51. Based on the waveform characteristic parameters in the multi-source real-time monitoring data, the real-time coal hardness parameters are calculated using the coal hardness calculation algorithm.

[0036] S52. Based on the cavity morphology data in the multi-source real-time monitoring data, the cavity volume correction mechanism is used to perform error compensation and obtain the corrected cavity volume parameters.

[0037] S53. Based on the pressure data and corrected cavity volume parameters in the multi-source real-time monitoring data, the cavity formation efficiency prediction algorithm is used to calculate and analyze the cavity formation efficiency status under the current parameter matching.

[0038] S54. Based on real-time coal hardness parameters, corrected cavity volume parameters, and cavity-forming efficiency status, and in conjunction with preset cavity-forming target parameters, the pump body pressure regulation mechanism is used to generate adjustment suggestions for the output parameters of the variable frequency pulsating pump.

[0039] Furthermore, based on the waveform characteristic parameters in the multi-source real-time monitoring data, and using the coal hardness calculation algorithm, the formula for calculating the real-time coal hardness parameter is as follows:

[0040] ;

[0041] In the formula, f This represents the Protodactyl hardness coefficient of the coal. F Indicates the frequency of the water jet reflected wave. A Indicates the amplitude of the water jet reflected wave. This represents the attenuation coefficient of the water jet reflected wave. and All of these represent correction factors. Indicates the reference hardness value. t Indicates the impact time.

[0042] Furthermore, an adaptive parameter adjustment mechanism is used to adjust the parameter adjustment suggestions, resulting in optimized pump control parameters including:

[0043] S61. Compare the parameter adjustment suggestions with the preset target parameters for creating a cavity. Based on the cavity size deviation rate and cavity creation efficiency deviation generated by the comparison, determine the current required parameter adjustment level.

[0044] S62. Based on the determined adjustment level, call the corresponding preset amplitude calculation rule, and then combine the pump body pressure adjustment mechanism and the cavitation efficiency prediction algorithm to calculate the target adjustment amount of each output parameter of the variable frequency pulsating pump.

[0045] S63. Based on the target adjustment amount, use linear interpolation to control the variable frequency pulsating pump so that its output parameters can smoothly transition from the current value to the target value.

[0046] S64. Verify the adjustment effect based on the target value and monitoring data. If the effect matches the preset optimization expectation, then lock the current parameter.

[0047] S65. Verify the adjustment effect based on the target value and monitoring data. If the effect does not meet expectations, return to step S62 to recalculate the adjustment amount and obtain optimized pump control parameters that meet the control requirements.

[0048] Furthermore, the optimized pump control parameters are input again into steps S4-S6 to monitor and adjust the cavity-forming parameters, thereby obtaining a stable cavity that meets the target size and efficiency requirements. This ensures the precise completion of directional long-hole hydraulic cavity-forming operations and the complete retention of operation process data, including:

[0049] S71. Continuously perform monitoring and parameter adjustment until the real-time size and efficiency data of the cavity reach the preset target range of the real-time size and efficiency data respectively.

[0050] S72. Once the standard is confirmed to be met, gradually reduce the output parameters of the variable frequency pulsating pump according to the grade order and duration, and maintain the preset duration under the final stage parameters, stop the pump body operation and shut it down to complete the hole-making operation.

[0051] S73. Generate a process report of this operation, including records of cavity morphology changes, parameter adjustments, and efficiency statistics, based on the completed hole-making operation process, and obtain the archived hole-making results.

[0052] Secondly, the present invention also provides an adaptive control system for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines, comprising:

[0053] The hole-making equipment and multi-source sensor debugging module is used to place the preset hole-making equipment and multi-source sensors in the target area of ​​hole-making in the coal mine, so as to complete the debugging of the hole-making equipment and multi-source sensors.

[0054] The data acquisition and preprocessing module is used to detect and collect data on the target cavitation area using multi-source sensors, generate initial detection data from the detection and collection results, and store and preprocess the initial detection data to obtain preprocessed initial detection data.

[0055] The working condition parameter setting module is used to set the output parameters of the variable frequency pulsating pump based on the preset cavity-forming target parameters and the pre-processed initial detection data, and generate pump output working conditions that meet the target coal body conditions.

[0056] The multi-source monitoring data acquisition module is used to extract multi-source real-time monitoring data during the pump output process using multi-source sensors.

[0057] The model building and analysis module is used to build a deep learning algorithm and a quantitative calculation model based on the coal hardness calculation algorithm, the pump pressure regulation mechanism, the cavity creation efficiency prediction algorithm, and the cavity volume correction mechanism. The deep learning algorithm and quantitative calculation model are used to analyze multi-source real-time monitoring data, and the analysis results are compared with the preset cavity creation target parameters to obtain parameter adjustment suggestions.

[0058] The parameter adaptive optimization module is used to adjust the parameter adjustment suggestions using an adaptive parameter adjustment mechanism to obtain optimized pump control parameters;

[0059] The pump control optimization and target achievement module is used to input the optimized pump control parameters back into steps S4-S6, monitor and adjust the cavity-making parameters, and obtain a stable cavity-making cavity that meets the target size and efficiency requirements, thereby achieving the accurate completion of directional long borehole hydraulic cavity-making operations and the complete retention of operation process data.

[0060] The technical effects achieved by this invention are as follows:

[0061] 1. This invention achieves adaptive adjustment of cavity-forming parameters. By using four quantitative models, the parameter adjustment perceived by the coal body is transformed from qualitative to quantitative, solving the problems of low efficiency and uncontrolled cavity size caused by fixed cavity-forming parameters in traditional methods.

[0062] 2. This invention improves the safety and reliability of cavity creation, ensuring precise matching between the system's safety threshold and the pump's pressure output range, thus preventing inconsistencies in pressure settings. Simultaneously, it utilizes a cavity volume correction model to eliminate dust interference, guaranteeing the accuracy and reliability of monitoring data. In the event of risks such as sudden pressure increases or cavity collapse, operations can be immediately suspended and an alarm issued, reducing the probability of safety accidents.

[0063] 3. This invention relies on reducing labor costs and labor intensity. The entire hole-making process is automatically completed through a closed loop of sensing, calculation, and adjustment, without the need for frequent human intervention.

[0064] 4. This invention relies on optimizing the quality and economy of cavity formation, using a cavity formation efficiency prediction model to optimize parameter matching relationships in advance, reducing the ineffective consumption of high-pressure water flow, and at the same time, the quantitative model ensures that the optimal parameters can be matched under different coal body conditions. Attached Figure Description

[0065] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0066] Figure 1 This is one of the flowcharts of the adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to an embodiment of the present invention.

[0067] Figure 2 This is a schematic diagram of the adaptive control system for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines, according to an embodiment of the present invention.

[0068] Figure 3 This is a schematic diagram of the connection of some modules of the adaptive cavity control system according to an embodiment of the present invention.

[0069] Figure 4 This is a schematic diagram of the assembly structure of the cavitation device, the multi-source sensor debugging module, and the water jet nozzle according to an embodiment of the present invention.

[0070] Figure 5 This is a cross-sectional view showing the positional relationship between the cavity and the directional long borehole according to an embodiment of the present invention.

[0071] Figure 6 This is the second flowchart of the adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to an embodiment of the present invention.

[0072] In the picture:

[0073] I. Hole-making equipment and multi-source sensor debugging module; II. Data acquisition and preprocessing module; III. Operating condition parameter setting module; IV. Multi-source monitoring data acquisition module; V. Model construction and analysis module; VI. Parameter adaptive optimization module; VII. Pump control optimization and target achievement module; 11. Variable frequency pulsating pump; 12. High-pressure drill rod; 13. Water jet nozzle; 21. Ultrasonic detector; 211. Ultrasonic probe; 212. Signal transmission line; 22. Water jet pressure sensor; 221. Pressure sensing core; 222. Data connector; 23. Water jet waveform acquisition device; 231. Acoustic sensor; 232. Fixed bracket; 5. Directional long borehole; 6. Hole-making cavity; 61. Water jet; 62. Target coal body; 7. Coal mine underground roadway wall. Detailed Implementation

[0074] To further illustrate the various embodiments, the present invention provides accompanying drawings, which are part of the disclosure of the present invention. These drawings are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these drawings, those skilled in the art should be able to understand other possible implementation methods and the advantages of the present invention.

[0075] According to an embodiment of the present invention, an adaptive control method for hydraulic cavity creation with alternating frequency amplitude of directional long boreholes in coal mines is proposed.

[0076] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 and Figure 6 As shown, the adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to an embodiment of the present invention includes:

[0077] Step S1: Place the preset cavity-making equipment and multi-source sensors in the target cavity-making area in the coal mine to complete the debugging of the cavity-making equipment and multi-source sensors.

[0078] Specifically, such as Figure 5 As shown, an operating platform is built at point 7 on the wall of the underground roadway in the coal mine. A variable frequency pulsating pump 11 and a data acquisition and preprocessing module II are fixed thereon. The variable frequency pulsating pump 11 and a high-pressure drill rod 12 are connected through a high-pressure water pipeline. The high-pressure drill rod 12 is sent into the target area of ​​the cavity along the axis of the directional long borehole 5. The cavity-making equipment and the multi-source sensor debugging module I are installed. The ultrasonic detector 21, the water jet pressure sensor 22, and the water jet waveform collector 23 are assembled with the water jet nozzle 13 through a fixed bracket 232. The contact between each sensor and the coal body and the detection direction are checked to ensure that the cavity-making cavity 6 area is covered. The communication line is connected, the system is powered on and debugged, the accuracy of the ultrasonic detector 21 is calibrated, and the water jet pressure sensor 22 has a pressure sensing core 221 and a data connector 222 inside. The accuracy is checked, and the data transmission of each module is tested to ensure that there is no communication interruption or data loss.

[0079] Specifically, this module, serving as the power source for cavity creation, includes a variable frequency pulse pump 11, a high-pressure drill rod 12, and a water jet nozzle 13. The variable frequency pulse pump 11 employs a variable displacement plunger pump structure, the high-pressure drill rod 12 is made of 20# seamless steel pipe, and the water jet nozzle 13 uses a multi-hole spray structure to ensure that the water jet can concentrate its impact on the target coal seam area. The high-pressure drill rod 12 is made of high-strength alloy material, and its length is adapted to the depth requirements of long directional drilling in coal mines. It is used to stably deliver high-pressure water flow to the water jet nozzle 13, converting the high-pressure water flow into a high-speed water jet to impact the coal seam, achieving directional cavity creation. The frequency-amplitude alternating hydraulic cavity creation module uses high-pressure water jets with periodically changing frequency and amplitude to impact the coal seam. Compared with constant water jets, this method can induce alternating stress within the coal seam. Based on the principle of fatigue damage accumulation, stress concentration will preferentially occur at the original micro-cracks in the coal seam, promoting crack propagation and thus achieving efficient crushing.

[0080] Step S2: Use multi-source sensors to detect and collect data on the target cavitation area, generate initial detection data from the detection and collection results, and store and preprocess the initial detection data to obtain preprocessed initial detection data.

[0081] Specifically, the data acquisition and preprocessing module II is activated. The ultrasonic detector 21 emits ultrasonic signals towards the target coal body 62, and rotates and scans at eight uniform angles corresponding to the 0-30° spray angle range, such as 0°, 4.5°, 9°...27°. Ten sets of continuous ultrasonic signals are emitted at each angle, and reflected echoes from coal bodies at different depths are received. The distance from the coal body surface to the probe is calculated by taking the average propagation time of each set of signals through the echo propagation time. The distance = speed of sound × propagation time / 2. A three-dimensional point cloud is generated by combining the scanning angle information. The initial three-dimensional image of the coal body morphology is generated by reconstructing the Poisson surface. Combined with correlation tests, the actual initial hardness of the target coal body is determined. The initial waveform is the environmental reference value and needs to be compared and corrected with the waveform after the coal body impact.

[0082] All initial data are transmitted to the data acquisition and preprocessing module II and stored according to a classification and partitioning strategy: 3D image data is stored in PLY format on a dedicated solid-state drive partition, and pressure and waveform data are stored in CSV format. During preprocessing, one set of abnormal waveform data is removed using the 3σ criterion, and the pressure data is smoothed using the moving average method. Five sets of continuous pressure data from the initial stage are taken, and the entire data is processed by sliding the window sequentially. The waveform data is filtered for interference using a power frequency notch filter, and then standardized and stored for later use.

[0083] Specifically, this module is responsible for real-time acquisition of coal and water jet state data during the cavity-forming process. It involves an ultrasonic detector 21, a water jet pressure sensor 22, and a water jet waveform collector 23. It continuously collects relevant data to ensure accurate acquisition of various state information of the coal and water jet during cavity-forming, providing a comprehensive and accurate data foundation for subsequent analysis and research. The ultrasonic detector 21 uses a high-frequency ultrasonic probe 211, covering the area around the water jet nozzle 13. Using a signal transmission line 212, it acquires the diameter, length, and contour data of the cavity 6 by transmitting and receiving ultrasonic signals. The water jet pressure sensor 22 is installed at the inlet of the water jet nozzle 13, monitoring the pressure changes when the water jet impacts the coal body in real time, reflecting the resistance of the coal body to the water jet. The water jet waveform collector 23 is installed on the side of the water jet nozzle 13, using an acoustic wave sensor 231 to capture the reflected waveform after the water jet impacts the coal body, recording parameters such as the reflected wave frequency, amplitude, and attenuation coefficient, providing basic data for coal hardness calculation.

[0084] The ultrasonic detector 21 is a high-frequency ultrasonic probe 211, which is mounted in front of the water jet nozzle 13 with the aid of a fixing bracket 232. The water jet pressure sensor 22 is a piezoelectric sensor, connected in series at the connection interface between the high-pressure drill rod 12 and the water jet nozzle 13. The water jet waveform acquisition device 23 is an acoustic sensor 231, which is fixed to the side of the water jet nozzle 13 with a bracket.

[0085] Step S3: Based on the preset target parameters for creating the cavity and the pre-processed initial detection data, set the output parameters for the variable frequency pulsating pump to generate pump output conditions that meet the target coal body conditions.

[0086] Specifically, the working condition parameter setting module III sends initial parameter instructions to the parameter adaptive optimization module VI based on the target coal body parameters; the parameter adaptive optimization module VI controls the variable frequency pulsating pump 11 to start, and the high-pressure water flow is delivered to the water jet nozzle 13 through the high-pressure drill rod 12 to form a water jet 61 with alternating frequency and amplitude to impact the target coal body 62 and start the cavity creation operation.

[0087] Step S4: Use multi-source sensors to extract multi-source real-time monitoring data during the pump output process;

[0088] Specifically, during the cavity creation process, the multi-source monitoring data acquisition module IV collects data at a preset frequency: the ultrasonic detector 21 detects once, measuring the cavity diameter and length during cavity creation. The water jet pressure sensor 22 records the pressure at the sampling frequency. The water jet waveform collector 23 collects the reflected waveform data of the cavity creation in real time via the communication line to the data acquisition and preprocessing module II, where Kalman filtering removes power frequency interference.

[0089] Step S5: Based on the coal hardness calculation algorithm, pump pressure regulation mechanism, cavity creation efficiency prediction algorithm and cavity volume correction mechanism, a deep learning algorithm and a quantitative calculation model are constructed. The deep learning algorithm and quantitative calculation model are used to analyze multi-source real-time monitoring data. The analysis results are compared with the preset cavity creation target parameters to obtain parameter adjustment suggestions.

[0090] Specifically, the data processing and analysis model. Module V, the model building and analysis module, analyzes data using deep learning algorithms. This module employs an industrial computer and communicates bidirectionally with the parameter adaptive optimization module VI and the cavity-forming equipment and multi-source sensor debugging module I. The built-in deep learning algorithms consist of convolutional neural network and recurrent neural network models, along with four major quantitative calculation models: a coal hardness calculation model, a pump pressure regulation model, a cavity-forming efficiency prediction model, and a cavity volume correction model.

[0091] Its main functions cover the following aspects: First, it stores historical monitoring data and cavity-building parameters to build a database for algorithm training. Second, it performs preprocessing operations on real-time monitoring data, including filtering, noise reduction, and data standardization, and uses a cavity volume correction model to eliminate dust interference, thereby improving data accuracy. Third, it uses deep learning algorithms to analyze monitoring data, explore the correlation between coal hardness, water jet parameters, and cavity-building effect, optimize cavity control strategies, and feed the optimized parameters back to the adaptive control module. Fourth, it generates a cavity-building process report, which includes cavity morphology change curves, parameter adjustment records, etc., providing a reference for subsequent cavity-building operations.

[0092] Step S6: Adjust the parameter adjustment suggestions using an adaptive parameter adjustment mechanism to obtain optimized pump control parameters;

[0093] Specifically, after receiving adjustment suggestions, the parameter adaptive optimization module VI adjusts the parameters according to the four-step process of the graded control strategy: parameter deviation judgment.

[0094] After adjustment and continuous monitoring, during hole creation, the diameter was measured by ultrasonic detector 21. D ,length L After revision V The deviation rate decreased. The frequency and amplitude of the reflected water jet were measured, and the hardness was calculated. The data acquisition and preprocessing module II generated a moderate adjustment suggestion. The parameter adaptive optimization module VI executed the moderate adjustment process and monitored subsequent data. The deviation rate continued to decrease, indicating that the adjustment was effective.

[0095] Specifically, this module, as the core control part of the system, uses a PLC controller and communicates with the working condition parameter setting module III and the cavitation device and multi-source sensor debugging module I via industrial Ethernet. The module is equipped with a cavity control logic algorithm and a pump pressure regulation model. It can receive monitoring data transmitted from the multi-source sensor debugging module I in real time, compare these data with the pre-set cavitation target parameters, such as cavity diameter and cavitation efficiency, and then generate parameter adjustment commands for the variable frequency pulse pump 11 based on the quantization model. This controls the output pressure, frequency and amplitude of the variable frequency pulse pump 11, and finally achieves dynamic adjustment of the cavitation parameters.

[0096] Meanwhile, this module possesses fault diagnosis capabilities. When monitored data exceeds the safety threshold, it can immediately issue an alarm signal and control the cavity-forming module to stop operating, thereby ensuring system safety. The parameter adaptive optimization module VI utilizes a PLC controller and communicates with each module via industrial Ethernet. The data acquisition and preprocessing module II uses an industrial computer, internally equipped with a deep learning algorithm combining convolutional neural networks (CNN) and recurrent neural networks (RNN), with 500 training samples, a prediction error of less than or equal to 3%, and four quantization models. Pre-set cavity-forming target parameters and a safety threshold set to trigger an alarm when pressure > 30 MPa or abnormal cavity morphology occurs.

[0097] Step S7: Input the optimized pump control parameters back into steps S4-S6 to monitor and adjust the cavity-forming parameters, thereby obtaining a stable cavity-forming chamber that meets the target size and efficiency requirements, thus achieving the accurate completion of directional long-hole hydraulic cavity-forming operations and the complete retention of operation process data.

[0098] Specifically, the ultrasonic detector 21 monitors the diameter and length of the cavity 6, and after correction, matches the target parameters. The parameter adaptive optimization module controls the variable frequency pulse pump 11 to gradually reduce the parameters and stop the pump operation, shutting down the coal body state sensing module and completing the cavity creation operation. The pump control optimization and target achievement module VII automatically generates a report on this cavity creation.

[0099] To verify the effectiveness of the implementation, after the cavity was created, the cavity 6 was observed in the field using a borehole inspection instrument. Subsequent gas extraction data showed that the cavity 6 improved the permeability of the coal seam, and the extraction efficiency met the design requirements, which verified the effectiveness and practicality of the system and method.

[0100] Specifically, it involves the cavitation equipment and the following modules: multi-source sensor debugging module I, data acquisition and preprocessing module II, operating condition parameter setting module III, multi-source monitoring data acquisition module IV, model construction and analysis module V, parameter adaptive optimization module VI, and pump control optimization and target achievement module VII. For example... Figure 4 As shown, the cavity-making equipment and multi-source sensor debugging module I includes a variable frequency pulsating pump 11, a high-pressure drill rod 12, a water jet nozzle 13, an ultrasonic detector 21, a water jet pressure sensor 22, and a water jet waveform collector 23. It is used to generate high-pressure water jets with adjustable frequency and amplitude. The ultrasonic detector 21 is used to detect the size and shape of the cavity-making chamber 6. The water jet pressure sensor 22 monitors the pressure change when the water jet impacts the coal body in real time. The water jet waveform collector 23 records the reflected waveform after the water jet impacts the coal body and extracts the frequency, amplitude, and attenuation coefficient of the reflected wave.

[0101] The parameter adaptive optimization module VI is electrically connected to the working condition parameter setting module III and the multi-source monitoring data acquisition module IV, respectively. It can receive the monitoring data from the multi-source monitoring data acquisition module IV and adjust the pressure, output frequency and amplitude of the variable frequency pulsating pump 11 according to the data.

[0102] The data acquisition and preprocessing module II communicates bidirectionally with the model building and analysis module V, and the cavity-forming device and multi-source sensor debugging module I. The model building and analysis module V incorporates a deep learning algorithm to analyze monitoring data, optimize cavity control parameters, and feed the optimization strategy back to the pump control optimization and target achievement module VII. The quantitative calculation models include a coal hardness calculation model, a pump pressure regulation model, a cavity-forming efficiency prediction model, and a cavity volume correction model.

[0103] In this optional embodiment, the pre-set cavity-creating device and multi-source sensors are placed in the target cavity-creating area underground in the coal mine to complete the debugging work of the cavity-creating device and multi-source sensors, including:

[0104] S11, ultrasonic detector 21, water jet pressure sensor 22 and water jet waveform acquisition device 23 are multi-source sensors;

[0105] S12. Connect and adjust to the target area of ​​the hole-making device and multi-source sensor through the preset hole-making device;

[0106] S13. Use the terminal to test the preset hole-making equipment and multi-source sensors to complete the debugging of the hole-making equipment and multi-source sensors.

[0107] Specifically, within the directional long borehole construction area of ​​the coal mine, the high-pressure drill rod 12 and water jet nozzle 13 of the frequency amplitude alternating hydraulic cavity-making module are sent to the target borehole location. The orientation of the water jet nozzle 13 is adjusted so that it faces the target cavity-making area, for example, perpendicular to the axis of the borehole. The ultrasonic detector 21, water jet pressure sensor 22, and water jet waveform collector 23 of the coal body state sensing module are installed at a pre-set position near the water jet nozzle. The communication lines and power lines of each module are connected, the system is started for power-on debugging, and it is checked whether each module can work normally. The accuracy of the ultrasonic detector is calibrated, the accuracy of the water jet pressure sensor is corrected, the cavity-making target parameters are set, and an alarm is triggered when the cavity shape is abnormal. The assembly and debugging of the system are completed.

[0108] In this optional embodiment, a multi-source sensor is used to detect and collect data on the target cavitation area. The results of the detection and collection are used to generate initial detection data, which is then stored and preprocessed to obtain preprocessed initial detection data, including:

[0109] S21. Using ultrasonic detector 21, perform multi-angle rotational scanning operation on the cavity-making area to obtain reflected echo data, and generate a three-dimensional image of the initial shape of the coal body based on the sound wave propagation time and scanning angle.

[0110] S22. Using water jet pressure sensor 22 and water jet waveform acquisition device 23, the initial reference pressure and environmental waveform data are acquired and recorded respectively.

[0111] S23. Transmit the three-dimensional image, reference pressure, and environmental waveform data as initial detection data to the data processing terminal.

[0112] S24. Using a data processing terminal, the initial probe data is classified and stored. Then, outlier removal, noise smoothing, and standardization preprocessing are performed sequentially to obtain the preprocessed initial probe data.

[0113] Specifically, after the coal body state sensing module is activated, the ultrasonic detector 21 begins to perform initial detection work on the initial shape and distribution of the coal body before the cavity is created. The water jet pressure sensor 22 and the water jet waveform acquisition device 23 enter standby mode and are responsible for recording the pressure and waveform reference values ​​under the initial environment. Then, the initial detection data is transmitted to the data processing module for storage and preprocessing operations.

[0114] A three-dimensional image of the initial coal body morphology is generated using multi-view scanning and echo signal 3D reconstruction. The ultrasonic probe selects from eight uniformly distributed angles (0°, 4.5°, 9°…27°) corresponding to the 0-30° injection angle range, and then performs a sequential rotational scan, emitting 10 sets of continuous ultrasonic signals at each angle to receive echoes reflected from coal bodies at different depths. The distance between the coal surface and the probe is then calculated using the echo propagation time: distance = speed of sound × propagation time / 2. Combining the scanning angle information, a voxelization reconstruction algorithm is used to transform the discrete distance data into a three-dimensional point cloud. A continuous three-dimensional image of the initial coal body morphology is then generated using Poisson surface reconstruction to determine the distribution of the coal body before cavity formation. The water jet pressure sensor 22 records the initial reference pressure, and the water jet waveform acquisition device 23 records the initial environmental waveform, which serves as a reference for subsequent comparisons. All initial detection data is transmitted to the data processing module for storage and preprocessing.

[0115] The storage method can adopt a classified and partitioned storage strategy. The three-dimensional image data of coal body morphology is stored in PLY format on solid-state drives. The solid-state drives can be partitioned separately, with a reserved redundant space of ≥50GB. Time-series data such as pressure and waveforms are stored in CSV format, named by date to timestamp. Detection parameters such as ultrasonic scanning angle and sampling frequency are stored together, and data can be quickly retrieved by time and parameter type.

[0116] Preprocessing methods can first remove outlier data using the 3σ criterion, such as data where the ultrasonic echo signal amplitude exceeds three times the normal range standard deviation. Then, the pressure data can be smoothed using a moving average method. Specific steps: Set the sliding window size to five consecutive data points, and select windows sequentially according to the time series. Calculate the average value of the five pressure data points within each window (p...). i-2 +p i-1 +p i +p i+1 +p i+2 ) / 5, where the formula yields the smoothed pressure value for the i-th window, p i The i-th original pressure data point within the window is used. When the window is less than 5 data points from the beginning or end of the data sequence, the average value is calculated based on the actual available data points. The original data is replaced with smoothed data to eliminate random fluctuation noise. Electromagnetic interference is filtered out using a power frequency notch filter on the waveform data. Initial error parameters are set using a cavity volume correction model.

[0117] In this optional embodiment, based on preset cavity-creating target parameters and pre-processed initial detection data, the output parameters of the variable frequency pulsating pump 11 are set to generate pump output conditions that meet the target coal body conditions, including:

[0118] S31. Based on the preprocessed initial detection data and the preset cavity-forming target parameters, determine the initial hardness type of the target coal body;

[0119] S32. Based on the determined initial coal body hardness type, set the corresponding preset output parameter group;

[0120] S33. Set the initial output parameters of the variable frequency pulsating pump 11 according to the corresponding preset output parameter group;

[0121] S34. Start the set variable frequency pulsating pump 11 to form a water jet with alternating frequency and amplitude to impact the coal body and obtain the pump output condition of the target coal body.

[0122] Specifically, after activating the frequency-amplitude alternating hydraulic cavity-forming module, the adaptive control module will initially set the output frequency, amplitude, and pressure of the variable frequency pulsating pump 11 according to the target cavity-forming parameters. The high-pressure water jet will be ejected from the water jet nozzle 13 via the high-pressure drill rod 12, impacting the coal body to initiate the cavity-forming operation. When the variable frequency pulsating pump 11 is started, high-pressure water is transported to the water jet nozzle 13 via the high-pressure drill rod 12, forming a frequency-amplitude alternating water jet to impact the coal body and begin the cavity-forming operation.

[0123] In this optional embodiment, the extraction of multi-source real-time monitoring data during the pump output process using multi-source sensors includes:

[0124] S41. Trigger the ultrasonic detector 21 according to the preset first cycle to perform scanning work on the cavity 6 and generate real-time cavity morphology data.

[0125] S42. Drive the water jet pressure sensor 22 according to the preset second sampling frequency to continuously monitor the water jet pressure and generate pressure fluctuation data.

[0126] S43. Based on the preset third cycle, the water jet waveform acquisition device 23 is triggered to acquire the water jet reflection waveform and extract the preset waveform feature parameters from it.

[0127] S44. Real-time cavity morphology data, pressure fluctuation data, and waveform characteristic parameters are treated as multi-source real-time monitoring data and transmitted to the data processing terminal in real time.

[0128] Specifically, during the cavity creation process, the coal body state sensing module operates in real time: each time the ultrasonic detector 21 detects the size and shape of the cavity 6, it completes one cavity creation cavity 6 shape detection and generates real-time cavity size data, such as diameter. D ,length L The water jet pressure sensor 22 continuously monitors the water jet impact pressure at its sampling frequency and records the pressure fluctuation curve. The water jet waveform acquisition device 23 acquires the water jet reflection waveform and extracts the frequency of the reflected wave each time. F ,amplitude A and attenuation coefficient All monitoring data is transmitted to the data processing module in real time.

[0129] In this optional embodiment, a deep learning algorithm and a quantitative calculation model are constructed based on a coal hardness calculation algorithm, a pump pressure regulation mechanism, a cavity-forming efficiency prediction algorithm, and a cavity volume correction mechanism. This deep learning algorithm and quantitative calculation model are then used to analyze multi-source real-time monitoring data. The analysis results are compared with preset cavity-forming target parameters, and parameter adjustment suggestions are derived, including:

[0130] S51. Based on the waveform characteristic parameters in the multi-source real-time monitoring data, the real-time coal hardness parameters are calculated using the coal hardness calculation algorithm.

[0131] S52. Based on the cavity morphology data in the multi-source real-time monitoring data, the cavity volume correction mechanism is used to perform error compensation and obtain the corrected cavity volume parameters.

[0132] S53. Based on the pressure data and corrected cavity volume parameters in the multi-source real-time monitoring data, the cavity formation efficiency prediction algorithm is used to calculate and analyze the cavity formation efficiency status under the current parameter matching.

[0133] S54. Based on the real-time coal hardness parameters, the corrected cavity volume parameters, and the cavity-forming efficiency status, and in conjunction with the preset cavity-forming target parameters, the pump body pressure regulation mechanism is used to generate adjustment suggestions for the output parameters of the variable frequency pulsating pump 11.

[0134] Specifically, the coal hardness calculation algorithm, pump pressure regulation mechanism, cavity formation efficiency prediction algorithm, and cavity volume correction mechanism are the coal hardness calculation model, pump pressure regulation model, cavity formation efficiency prediction model, and cavity volume correction model.

[0135] Environmental interference signals are removed using Kalman filtering and noise reduction algorithms. The data processing module analyzes the monitoring data using built-in deep learning algorithms and quantization calculation models.

[0136] Based on the frequency of water jet reflection waves F ,amplitude A and attenuation coefficient The Protodyakonov hardness coefficient of coal was calculated using a coal hardness calculation model. f To determine changes in coal hardness. f The value will be used as a key output parameter for subsequent calculations of the pump body pressure regulation model.

[0137] In this optional embodiment, based on the waveform characteristic parameters in the multi-source real-time monitoring data, the formula for calculating the real-time coal hardness parameter using the coal hardness calculation algorithm is as follows:

[0138] ;

[0139] In the formula, f This represents the Protodyakonov hardness coefficient of the coal. F Indicates the frequency of the water jet reflected wave. A Indicates the amplitude of the water jet reflected wave. This represents the attenuation coefficient of the water jet reflected wave. and All of these represent correction factors. Indicates the reference hardness value. t Indicates the impact time.

[0140] Specifically, the diameter of the cavity for ultrasonic detection D ,length L The actual cavity volume is calculated using a cavity volume correction model. V The formula is:

[0141] ;

[0142] In the formula, V This indicates the corrected cavity volume; D Indicates the diameter of the ultrasonic probe; LIndicates the length of the ultrasonic probe; , These represent the detection error rates for diameter and length, respectively. , This represents the volume correction factor. The corrected volume. V This will be used as an input parameter for the subsequent calculation of the cavity-forming efficiency prediction model and the pump body pressure regulation model.

[0143] The actual cavity-forming efficiency is calculated based on the change of the corrected cavity volume V over time. The efficiency under the current parameter matching was verified by using a hole-forming efficiency prediction model. The hole-forming efficiency prediction model is as follows:

[0144] ;

[0145] In the formula, This indicates the predicted hole-forming efficiency. V This indicates the corrected cavity volume. Indicates time interval, Indicates the pump body output pressure. F Indicates the pump body output frequency. , Indicates the reference pressure and frequency. This represents the efficiency correction factor. a , b An index representing the influence of pressure and frequency. If... If the value is less than 0.8η, it indicates that the parameter matching is unreasonable and the combination of frequency and amplitude needs to be optimized.

[0146] Parameter adjustment suggestions are generated. Based on the above analysis results, the target output pressure of the variable frequency pulsating pump 11 is calculated using the pump body pressure regulation model.

[0147] ;

[0148] In the formula, This indicates the output pressure of the variable frequency pulsating pump 11 after adjustment. Indicates the current pump body pressure. This represents the real-time coal hardness coefficient. This indicates the initial setting of the coal body hardness coefficient. V This indicates the corrected cavity volume. Indicates the volume of the target cavity. This represents the coefficient of influence of hardness. This indicates the volume deviation influence coefficient. The adjusted pressure value. This will be used as an optimized pump control parameter, input into the pump body actuator, to achieve dynamic adjustment of the cavity-forming parameters.

[0149] In this optional embodiment, an adaptive parameter adjustment mechanism is used to adjust the parameter adjustment suggestions to obtain optimized pump control parameters, including:

[0150] S61. Compare the parameter adjustment suggestions with the preset target parameters for creating a cavity. Based on the cavity size deviation rate and cavity creation efficiency deviation generated by the comparison, determine the current required parameter adjustment level.

[0151] S62. Based on the determined adjustment level, call the corresponding preset amplitude calculation rule, and then combine the pump body pressure adjustment mechanism and the cavitation efficiency prediction algorithm to calculate the target adjustment amount of each output parameter of the variable frequency pulsating pump 11.

[0152] S63. Based on the target adjustment amount, use linear interpolation to control the variable frequency pulsating pump so that its output parameters can smoothly transition from the current value to the target value.

[0153] S64. Verify the adjustment effect based on the target value and monitoring data. If the effect matches the preset optimization expectation, then lock the current parameter.

[0154] S65. Verify the adjustment effect based on the target value and monitoring data. If the effect does not meet expectations, return to step S62 to recalculate the adjustment amount and obtain the optimized pump control parameters that meet the control requirements.

[0155] Specifically, the adaptive control module receives adjustment suggestions from the data processing module and dynamically adjusts the output parameters of the variable frequency pulsating pump 11. When the coal hardness increases, the pump pressure and amplitude are increased according to the pump pressure adjustment model, while the output frequency is reduced. If the size of the cavity 6 is close to the target value, the pump pressure and amplitude are reduced, and the output frequency is increased. If the cavity efficiency is lower than the preset threshold, the frequency and amplitude matching relationship is optimized using the cavity efficiency prediction model to ensure that the cavity creation process can proceed stably. The target cavity creation parameters are compared, and parameter adjustment suggestions are generated in conjunction with the pump pressure adjustment model. The deep learning algorithm uses a model combining convolutional neural networks and recurrent neural networks. Its input features include water jet reflection waveform features, cavity size data, and water jet pressure. The output is the pressure, frequency, and amplitude adjustment of the variable frequency pulsating pump 11. The model training samples contain 500 sets of historical data with different coal hardness and different cavity creation parameters. Parameter deviation judgment: Extract the cavity size deviation rate |δ| and the actual cavity creation efficiency. , with the target hole-forming efficiency The adjustment is divided into three levels: fine adjustment, medium adjustment, and deep adjustment.

[0156] Adjustment calculation: Fine adjustment is calculated based on the current parameter amplitude, medium adjustment is calculated based on the preset amplitude, and deep adjustment is calculated based on the set amplitude; pressure adjustment is combined with the pump body pressure adjustment model, and frequency and amplitude adjustment are based on the coal body hardness change and cavity-forming efficiency prediction model.

[0157] Smooth parameter output: Linear interpolation is used to achieve smooth parameter transition. For example, the transition time is divided into fine-tuning, medium-tuning and deep-tuning to avoid sudden parameter changes.

[0158] After all parameter adjustments have been calculated, linear interpolation is used to achieve a smooth transition of parameters in order to prevent sudden changes in the water jet impact intensity caused by parameter mutations, which could lead to cavity collapse or a sudden decrease in cavity-building efficiency.

[0159] Adjustment effect feedback: Monitor the next 1 to 2 sets of data. If the adjustment is effective, maintain the parameter. If it is ineffective, adjust the adjustment amount by 50% of the original level.

[0160] After the parameters are output, the system continuously monitors the next 1-2 sets of data using an ultrasonic detector 21 and a pressure sensor. If the deviation rate decreases and the efficiency increases after adjustment, the current parameters are maintained. If the effect does not meet expectations, the system returns to the adjustment calculation step to recalculate the adjustment amount. During the adjustment process, the system monitors the impact of parameter changes on the acupoint creation effect in real time and uses the acupoint creation efficiency prediction model to predict the efficiency after adjustment, thus avoiding abrupt parameter changes that could lead to uncontrolled acupoint creation.

[0161] In this optional embodiment, the optimized pump control parameters are input again into steps S4-S6 to monitor and adjust the cavity-forming parameters, thereby obtaining a stable cavity-forming chamber that meets the target size and efficiency requirements. This achieves precise completion of directional long-hole hydraulic cavity-forming operations and complete retention of operation process data, including:

[0162] S71. Continuously perform monitoring and parameter adjustment until the real-time size and efficiency data of the cavity 6 reach the preset target range of the real-time size and efficiency data respectively.

[0163] S72. Once the standard is confirmed to be met, the output parameters of the variable frequency pulsating pump 11 are gradually reduced according to the grade order and duration, and the preset duration is maintained at the final stage parameter. The pump body is then stopped and shut down to complete the hole-making operation.

[0164] S73. Generate a process report of this operation, including records of cavity morphology changes, parameter adjustments, and efficiency statistics, based on the completed hole-making operation process, and obtain the archived hole-making results.

[0165] Specifically, steps S4 to S6 are repeated to continuously monitor and adjust the cavity-forming parameters. When the ultrasonic detector 21 detects that the size of the cavity-forming chamber 6 has reached the target value and the cavity-forming efficiency is stable within the target range, the adaptive control module controls the variable frequency pulse pump 11 to gradually reduce the output parameters and observes whether the shape of the chamber is stable. If it is stable, the pressure, frequency, and amplitude are then reduced. Finally, the variable frequency pulse pump 11 is stopped, the coal body state sensing module is turned off, and one cavity-forming operation is completed. The data processing module automatically generates a report of this cavity-forming process, which includes data such as the chamber shape change curve, parameter adjustment records, and cavity-forming efficiency statistics, for use in optimizing subsequent operations. The adaptive control module controls the frequency-amplitude alternating hydraulic cavity-forming module to gradually reduce the output parameters, stop the cavity-forming operation, and complete one adaptive cavity control process.

[0166] like Figure 2 and Figure 3 As shown, according to another embodiment of the present invention, an adaptive control system for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines is also provided, comprising:

[0167] The hole-making equipment and multi-source sensor debugging module I is used to place the preset hole-making equipment and multi-source sensors in the target area of ​​hole-making in the coal mine, so as to complete the debugging of the hole-making equipment and multi-source sensors.

[0168] Data acquisition and preprocessing module II is used to detect and collect data on the target cavitation area using multi-source sensors, generate initial detection data from the detection and collection results, and store and preprocess the initial detection data to obtain preprocessed initial detection data.

[0169] The working condition parameter setting module Ⅲ is used to set the output parameters of the variable frequency pulsating pump 11 based on the preset cavity-forming target parameters and the pre-processed initial detection data, and generate pump output working conditions that meet the target coal body conditions.

[0170] Multi-source monitoring data acquisition module IV is used to extract multi-source real-time monitoring data during the pump output process using multi-source sensors;

[0171] The model building and analysis module V constructs a deep learning algorithm and a quantitative calculation model based on the coal hardness calculation algorithm, the pump pressure regulation mechanism, the cavity creation efficiency prediction algorithm, and the cavity volume correction mechanism. The deep learning algorithm and quantitative calculation model are used to analyze multi-source real-time monitoring data, and the analysis results are compared with the preset cavity creation target parameters to obtain parameter adjustment suggestions.

[0172] The parameter adaptive optimization module VI is used to adjust the parameter adjustment suggestions using an adaptive parameter adjustment mechanism to obtain optimized pump control parameters;

[0173] The pump control optimization and target achievement module VII is used to input the optimized pump control parameters back into the multi-source monitoring data acquisition module IV, the model construction and analysis module V, and the parameter adaptive optimization module VI to monitor and adjust the cavity-making parameters, thereby obtaining a stable cavity-making cavity 6 that meets the target size and efficiency requirements. This enables the precise completion of directional long-hole hydraulic cavity-making operations and the complete retention of operation process data.

[0174] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines, characterized in that... :include: S1. Place the pre-set cavity-making equipment and multi-source sensors in the target cavity-making area in the coal mine to complete the debugging of the cavity-making equipment and multi-source sensors. S2. Use multi-source sensors to detect and collect data on the target cavitation area, generate initial detection data from the detection and collection results, and store and preprocess the initial detection data to obtain preprocessed initial detection data. S3. Based on the preset target parameters for creating the cavity and the pre-processed initial detection data, set the output parameters for the variable frequency pulsating pump to generate pump output conditions that meet the target coal body conditions. S4. Use multi-source sensors to extract multi-source real-time monitoring data during the pump output process; S5. Based on the coal hardness calculation algorithm, pump pressure regulation mechanism, cavity formation efficiency prediction algorithm and cavity volume correction mechanism, a deep learning algorithm and a quantitative calculation model are constructed. The deep learning algorithm and quantitative calculation model are used to analyze multi-source real-time monitoring data. The analysis results are compared with the preset cavity formation target parameters to obtain parameter adjustment suggestions. Specifically, it includes: Based on waveform characteristic parameters from multi-source real-time monitoring data, a coal hardness calculation algorithm is used to calculate real-time coal hardness parameters. Based on cavity morphology data from multi-source real-time monitoring data, a cavity volume correction mechanism is used for error compensation to obtain corrected cavity volume parameters. Based on pressure data from multi-source real-time monitoring data and corrected cavity volume parameters, a cavity-forming efficiency prediction algorithm is used to calculate and analyze the cavity-forming efficiency status under the current parameter matching. Based on real-time coal hardness parameters, corrected cavity volume parameters, and cavity-forming efficiency status, and in conjunction with preset cavity-forming target parameters, a pump pressure regulation mechanism is used to generate adjustment suggestions for the output parameters of the variable frequency pulsating pump. S6. The parameter adjustment suggestions are adjusted using an adaptive parameter adjustment mechanism to obtain optimized pump control parameters; S7. Input the optimized pump control parameters back into steps S4-S6 to monitor and adjust the cavity-making parameters, and obtain a stable cavity-making chamber that meets the target size and efficiency requirements, thereby achieving the accurate completion of directional long-hole hydraulic cavity-making operations and the complete retention of operation process data.

2. The adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to claim 1, characterized in that, The process of placing the pre-set cavity-creating equipment and multi-source sensors in the target cavity-creating area underground in the coal mine to complete the debugging of the cavity-creating equipment and multi-source sensors includes: S11, the ultrasonic detector, the water jet pressure sensor and the water jet waveform acquisition device are multi-source sensors; S12. Connect and adjust to the target area of ​​the hole-making device and multi-source sensor through the preset hole-making device; S13. Use the terminal to test the preset hole-making equipment and multi-source sensors to complete the debugging of the hole-making equipment and multi-source sensors.

3. The adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to claim 1, characterized in that, The process involves using multi-source sensors to detect and collect data on the target cavitation area, generating initial detection data from the collected data, and then storing and preprocessing the initial detection data to obtain the preprocessed initial detection data, which includes: S21. Using an ultrasonic detector, perform multi-angle rotational scanning of the cavity-forming area to obtain reflected echo data, and generate a three-dimensional image of the initial shape of the coal body based on the sound wave propagation time and scanning angle. S22. Use a water jet pressure sensor and a water jet waveform acquisition device to acquire and record the initial reference pressure and environmental waveform data respectively. S23. Transmit the three-dimensional image, reference pressure, and environmental waveform data as initial detection data to the data processing terminal. S24. Using a data processing terminal, the initial probe data is classified and stored. Then, outlier removal, noise smoothing, and standardization preprocessing are performed sequentially to obtain the preprocessed initial probe data.

4. The adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to claim 1, characterized in that, The process of setting output parameters for the variable frequency pulsating pump based on preset cavity-forming target parameters and pre-processed initial detection data, and generating pump output conditions that meet the target coal body conditions, includes: S31. Based on the preprocessed initial detection data and the preset cavity-forming target parameters, determine the initial hardness type of the target coal body; S32. Based on the determined initial coal body hardness type, set the corresponding preset output parameter group; S33. Set the initial output parameters of the variable frequency pulsating pump according to the corresponding preset output parameter group; S34. Start the set variable frequency pulsating pump to form a water jet with alternating frequency and amplitude to impact the coal body and obtain the pump output condition of the target coal body.

5. The adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to claim 1, characterized in that, The method of using multi-source sensors to extract multi-source real-time monitoring data during the pump output operation process: S41. Trigger the ultrasonic detector according to the preset first cycle to perform scanning work on the cavity and generate real-time cavity morphology data. S42. Drive the water jet pressure sensor according to the preset second sampling frequency to continuously monitor the water jet pressure and generate pressure fluctuation data. S43. Based on the preset third cycle trigger water jet waveform acquisition device, acquire the water jet reflection waveform and extract the preset waveform feature parameters from it. S44. Real-time cavity morphology data, pressure fluctuation data, and waveform characteristic parameters are treated as multi-source real-time monitoring data and transmitted to the data processing terminal in real time.

6. The adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to claim 1, characterized in that, The formula for calculating the real-time coal hardness parameter based on the waveform characteristic parameters in the multi-source real-time monitoring data and using the coal hardness calculation algorithm is as follows: ； In the formula, This represents the Protodyakonov hardness coefficient of the coal. Indicates the frequency of the water jet reflected wave. Indicates the amplitude of the water jet reflected wave. This represents the attenuation coefficient of the water jet reflected wave. and All of these represent correction factors. Indicates the reference hardness value. t Indicates the impact time.

7. The adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to claim 1, characterized in that, The method of adjusting the parameter adjustment suggestions using an adaptive parameter adjustment mechanism to obtain optimized pump control parameters includes: S61. Compare the parameter adjustment suggestions with the preset target parameters for creating a cavity. Based on the cavity size deviation rate and cavity creation efficiency deviation generated by the comparison, determine the current required parameter adjustment level. S62. Based on the determined adjustment level, call the corresponding preset amplitude calculation rule, and then combine the pump body pressure adjustment mechanism and the cavitation efficiency prediction algorithm to calculate the target adjustment amount of each output parameter of the variable frequency pulsating pump. S63. Based on the target adjustment amount, use linear interpolation to control the variable frequency pulsating pump so that its output parameters can smoothly transition from the current value to the target value. S64. Verify the adjustment effect based on the target value and monitoring data. If the effect matches the preset optimization expectation, then lock the current parameter. S65. Verify the adjustment effect based on the target value and monitoring data. If the effect does not meet expectations, return to step S62 to recalculate the adjustment amount and obtain optimized pump control parameters that meet the control requirements.

8. The adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines according to claim 1, characterized in that, The step of re-inputting the optimized pump control parameters into steps S4-S6, monitoring and adjusting the cavity-forming parameters, and obtaining a stable cavity-forming chamber that meets the target size and efficiency requirements, thereby achieving the precise completion of directional long-hole hydraulic cavity-forming operations and the complete retention of operation process data includes: S71. Continuously perform monitoring and parameter adjustment until the real-time size and efficiency data of the cavity reach the preset target range of the real-time size and efficiency data respectively. S72. Once the standard is confirmed to be met, gradually reduce the output parameters of the variable frequency pulsating pump according to the grade order and duration, and maintain the preset duration under the final stage parameters, stop the pump body operation and shut it down to complete the hole-making operation. S73. Generate a process report of this operation, including records of cavity morphology changes, parameter adjustments, and efficiency statistics, based on the completed hole-making operation process, and obtain the archived hole-making results.

9. An adaptive control system for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines, used to implement the adaptive control method for frequency-amplitude alternating hydraulic cavity creation in directional long boreholes in coal mines as described in any one of claims 1-8, characterized in that... include: The hole-making equipment and multi-source sensor debugging module is used to place the preset hole-making equipment and multi-source sensors in the target area of ​​hole-making in the coal mine, so as to complete the debugging of the hole-making equipment and multi-source sensors. The data acquisition and preprocessing module is used to detect and collect data on the target cavitation area using multi-source sensors, generate initial detection data from the detection and collection results, and store and preprocess the initial detection data to obtain preprocessed initial detection data. The working condition parameter setting module is used to set the output parameters of the variable frequency pulsating pump based on the preset cavity-forming target parameters and the pre-processed initial detection data, and generate pump output working conditions that meet the target coal body conditions. The multi-source monitoring data acquisition module is used to extract multi-source real-time monitoring data during the pump output process using multi-source sensors. The model building and analysis module is used to build a deep learning algorithm and a quantitative calculation model based on the coal hardness calculation algorithm, the pump pressure regulation mechanism, the cavity creation efficiency prediction algorithm, and the cavity volume correction mechanism. The deep learning algorithm and quantitative calculation model are used to analyze multi-source real-time monitoring data, and the analysis results are compared with the preset cavity creation target parameters to obtain parameter adjustment suggestions. The parameter adaptive optimization module is used to adjust the parameter adjustment suggestions using an adaptive parameter adjustment mechanism to obtain optimized pump control parameters; The pump control optimization and target achievement module is used to input the optimized pump control parameters back into steps S4-S6, monitor and adjust the cavity-making parameters, and obtain a stable cavity-making cavity that meets the target size and efficiency requirements, thereby achieving the accurate completion of directional long borehole hydraulic cavity-making operations and the complete retention of operation process data.

Images