A coal mine roadway roof drilling lithology identification system and method

Abstract

The application discloses a coal mine roadway roof drilling lithology identification system and method, relates to the technical field of mine support, and comprises a data acquisition system, a signal conversion system, a data processing system, an intelligent algorithm lithology identification system and a drilling system; the data acquisition system is used for collecting drilling pressure, drilling torque, drilling rotating speed, drilling displacement and drilling vibration signals in the drilling process of the drilling system; the signal conversion system is electrically connected with the data acquisition system and is used for receiving and converting sensor signals of the data acquisition system; the data processing system is used for receiving, storing and processing signals transmitted by the signal conversion system; the intelligent algorithm lithology identification system adopts a meta-heuristic algorithm to optimize an extreme learning machine, constructs an improved marine predator-extreme learning machine lithology identification model to identify the roof lithology, and the drilling system provides drilling power for the whole system; the application realizes lithology identification, roof dynamic support, blasting parameter optimization and other tests in the fields of mine exploitation, roadway support, punching construction and the like, improves the underground operation efficiency, realizes real-time detection and accurate dynamic support of the roadway roof hidden danger area, greatly guarantees construction safety, and has important practical engineering and scientific research significance.

Description

Technical Field

[0001] This invention relates to the field of mine support, specifically to a system and method for identifying the lithology of the roof during drilling in coal mine roadways. Background Technology

[0002] From the perspective of coal mining, large-section roadway cutting support technology is a key technology in coal mine roadway construction, which can effectively ensure the safety of construction personnel and the normal operation of coal mining work. Large-section roadway cutting can easily lead to collapse of weak joints in the roof, posing a safety hazard during roadway support. Therefore, roof management should be emphasized during the construction of large-section roadways. Furthermore, as coal mining gradually expands from shallow to deep, problems such as deep ground pressure manifestation and complex geological structures become more prominent, further exacerbating the difficulties of large-section roadway cutting support.

[0003] Currently, the inability to detect potential roof hazards in roadways in real time, and the inability to identify roof rock mechanics parameters in real time, leads to reliance on single-mode roof support and poses safety risks. Rock mechanics parameters are fundamental parameters for the classification and evaluation of surrounding rock in underground engineering, as well as for the analysis and design of support parameters. Drilling-while-operating lithology identification technology and real-time roof hazard detection technology can play a crucial role in support design and construction. Establishing the response relationship between drilling parameters and rock parameters can provide a basis for dynamic support and promote intelligent coal mining. Therefore, it is urgent to construct lithology identification and prediction models and develop real-time roof lithology identification technology. Summary of the Invention

[0004] The purpose of this invention is to provide a coal mine roadway roof lithology identification system and method to solve the problems existing in the prior art. By installing sensors on a pneumatic rock bolt drilling rig, an improved marine predator-extreme learning machine lithology identification model is constructed to determine the rock parameters of the coal mine roadway roof in real time. This enables lightweight lithology identification of the roadway roof rock bolt drilling rig, greatly improves the roadway support efficiency, increases the accuracy and pertinence of dynamic roadway roof support, and greatly ensures construction safety.

[0005] To achieve the above objectives, the present invention provides the following solution:

[0006] This invention provides a coal mine roadway roof lithology identification system during drilling, comprising a data acquisition system, a signal conversion system, a data processing system, an intelligent algorithm lithology identification system, and a drilling system;

[0007] The data acquisition system is used to collect the rotational speed, torque, displacement, thrust, and vibration signals of the drilling system during the drilling process of the tunnel roof. The signal conversion system is electrically connected to the data acquisition system and is used to receive and convert the sensor signals of the data acquisition system. The data processing system is electrically connected to the signal conversion system and is used to receive, store, and process the sensor signals converted by the signal conversion system. The intelligent algorithm lithology identification system uses a metaheuristic algorithm to optimize the extreme learning machine and constructs an improved marine predator-extreme learning machine lithology identification model to identify the roof lithology. The drilling system provides drilling power for the entire system.

[0008] Preferably, the data acquisition system includes a torque-speed sensor, a rope displacement sensor, a rotary pressure sensor, and a vibration sensor;

[0009] The torque-speed sensor is used to collect torque and speed signals during the drilling process. The torque-speed sensor is connected to the main shaft of the transmission system of the drilling system through a first rigid sleeve and to the drill rod through a second rigid sleeve.

[0010] The pull rope displacement sensor is installed on the upper part of the air leg component of the drilling system. The pull rope displacement sensor is connected to the air leg handle of the drilling system through a pull rope. During the drilling process, the pull rope rises and falls with the air leg to monitor the displacement in real time.

[0011] The rotary pressure sensor is located at the bottom of the drilling system and is connected to the top of the drilling system through a rigid threaded sleeve to monitor the propulsion force during the drilling process.

[0012] The vibration sensor is installed on the upper part of the motor drive assembly of the drilling system and is connected by magnetic attraction to monitor vibration signals during the drilling process.

[0013] Preferably, the signal conversion system includes a coal mine safety explosion-proof signal box, a first power transformer, a second power transformer, an air switch, a junction box, a first XSE5 digital display alarm controller, a second XSE5 digital display alarm controller, a signal conversion module, a first RS-485 serial port, and a second RS-485 serial port. The signal conversion system is connected to various sensors in the data acquisition system via signal lines to receive and process sensor signals.

[0014] The first and second power transformers are used to change the voltage of the signal conversion system to meet the various operating voltages of the sensor; the input terminal of the first power transformer is connected to an external power source, and the output terminal is connected to an air switch; the input terminal of the second power transformer is connected to the junction box, and the output terminal is connected to the vibration sensor and the signal conversion module respectively.

[0015] The air switch integrates control and multiple protection functions, and can protect against short circuits, severe overloads and undervoltages in the signal conversion system, ensuring the safe operation of the signal conversion system. The input terminal of the air switch is connected to the first power transformer, and the output terminal is connected to the junction box.

[0016] The junction box is a 3-in-12-out junction box that branches the main line to output more voltage signals. The input end of the junction box is connected to the air switch, and the output end branches out 4 groups of signal lines that are respectively connected to the second power transformer, the first XSE5 digital display alarm controller, the second XSE5 digital display alarm controller, and the data processing system.

[0017] The first and second XSE5 digital display alarm controllers use fully isolated 18-bit AD conversion for input, with a measurement accuracy of 0.55%FS, and have digital filtering and alarm functions, enabling the measurement of physical quantities in vibration environments. The input terminal of the first XSE5 digital display alarm controller is connected to the junction box, and its output terminal is connected to the displacement sensor and the first RS-485 serial port, respectively. The input terminal of the second XSE5 digital display alarm controller is connected to the junction box, and its output terminal is connected to the rotary pressure sensor and the first RS-485 serial port, respectively.

[0018] The signal conversion module can precisely amplify the output signal of the torque-speed sensor, perform voltage stabilization and impedance matching inside the circuit, connect the input end to the second power transformer, and connect the output end to the torque-speed sensor and the first RS-485 transmission serial port respectively.

[0019] The first RS-485 serial port input terminal is connected to the first XSE5 digital display alarm controller, the second XSE5 digital display alarm controller, and the signal conversion module, respectively, and the output terminal is connected to the data processing system;

[0020] The second RS-485 serial port input is connected to the vibration sensor, and the output is connected to the data processing system.

[0021] Preferably, the data processing system is an explosion-proof laptop computer used to receive the RS-485 signal converted by the signal conversion system and to store and process the converted data;

[0022] The explosion-proof laptop is connected to the junction box, the first RS-485 serial port and the second RS-485 serial port in the signal conversion system.

[0023] Preferably, the intelligent algorithm lithology identification system includes an algorithm optimization module, a data preprocessing module, and a lithology intelligent identification module;

[0024] The algorithm optimization module addresses the problems of low solution accuracy, slow convergence speed, and difficulty in escaping local optima in metaheuristic algorithms. It introduces opposition learning and differential evolution strategies to improve the metaheuristic algorithm and uses the improved metaheuristic algorithm to optimize the extreme learning machine.

[0025] The data preprocessing module includes outlier removal and normalization of the training and test sets. Intelligent identification requires the establishment of a lithology identification database of drilling parameters of a certain size. This data is collected through the aforementioned data acquisition system. Due to factors such as non-standard operation of the drilling system, a small amount of abnormal data may be generated. This data must be removed to prevent it from entering the database. At the same time, to avoid significant differences in similar drilling parameters of the same type of roof rock layer caused by different roof rock layers, different types of anchor drilling rigs, and different personnel operations, the input drilling parameters are normalized to avoid affecting the accuracy of the algorithm.

[0026] The dataset of the lithology intelligent identification module is divided into a training set and a test set. The training set is established based on existing sample data and the results of underground geological exploration and roof strata drilling. The input variables of the training set and the test set are the drilling pressure, drilling torque, drilling speed and drilling displacement after the collected data has been normalized. For ease of analysis, the output variables of the training set and the test set are set to three: coal (1), mudstone (2) and sandstone (3).

[0027] Preferably, the Extreme Learning Machine is an algorithm based on a single hidden layer feedforward neural network structure. It randomly selects the input layer weights and hidden layer biases, and obtains the output layer weights through generalized matrix theory, thereby calculating the predicted value of the data. The parameters that need to be manually set for the Extreme Learning Machine include: the number of input layer nodes, the number of hidden layer neurons, the number of output layer nodes, and the activation function.

[0028] Preferably, the metaheuristic algorithm is the Marine Predator Algorithm (MPA);

[0029] The oppositional learning strategy is as follows:

[0030] Δ={x_1,x_2〖,…,x〗_d};

[0031] Δ - ={x - _1,x - _2〖,…,x - 〗_d};

[0032] x - _j = lb_j + ub_j - x_j;

[0033] In the above formula, Δ represents the position of a certain d-dimensional individual. - The opposite side is defined by lb and ub, which are the upper and lower bounds of the individual's position.

[0034] The basic mutation strategy of the differential evolution algorithm is as follows:

[0035] X_i(t+1)=X_r1(t)+F·(X_r2(t)-X_r3(t));

[0036] In the above formula, r_1, r_2, and r_3 are distinct integers within the population range and are not equal to i; F is a constant factor, F∈[0,2].

[0037] Preferably, the initial weights and thresholds randomly generated by the extreme learning machine are saturated, and the improved opposition-difference ocean predator algorithm is used to optimize the initial weights and thresholds to improve prediction accuracy.

[0038] Preferably, the number of original datasets collected is 1800, the ratio of training set to test set is set to 8:2, the number of training set data is 1440, and the number of test set data is 360.

[0039] Preferably, the drilling system includes an MQT-130 / 3.2 pneumatic anchor drill and a drill rod. The MQT-130 / 3.2 series pneumatic anchor drill is widely applicable to roadways with rock hardness coefficient f≤10, and is particularly suitable for coal roadway anchor support operations. It can drill both roof anchor holes and anchor cable holes to achieve the purpose of rock breaking drilling.

[0040] The MQT-130 / 3.2 pneumatic anchor bolt drilling rig includes a motor transmission component, a body component, an air leg component, and a control arm component. The motor transmission component includes a motor assembly, a transmission assembly, and a water jacket assembly, with the motor assembly providing power to the entire system. The transmission assembly connects to the drill rod and rotates as a whole with the drill rod. The water jacket assembly connects to a water pipe to supply water during the drilling process. The air leg component controls the lifting and lowering of the drilling system. The control arm component allows for manual control to stabilize the anchor bolt drilling rig body.

[0041] The drill pipe is connected to the torque-speed sensor via a second rigid sleeve.

[0042] Preferably, the drill rod has a connection port in the middle, and multiple drill rods can be connected to each other to meet the requirements of different drilling depths.

[0043] Based on the above-mentioned coal mine roadway roof drilling lithology identification system, the present invention also provides a coal mine roadway roof drilling lithology identification method, comprising the following steps:

[0044] 1) First, connect the MQT-130 / 3.2 pneumatic anchor drill to the blower and water pump respectively to ensure the required air pressure, air volume and water pressure when the pneumatic anchor drill is working. The oil mist lubricator of the pneumatic anchor drill must not be short of oil to ensure that the pneumatic anchor drill is adequately lubricated.

[0045] 2) Place the MQT-130 / 3.2 pneumatic anchor drill vertically, with the pressure sensor in contact with the ground, install the drill rod, and raise the air leg to a position where the drill rod is close to contacting the top plate, then begin drilling;

[0046] 3) During the drilling process, the torque-speed sensor, cable displacement sensor, pressure sensor and vibration sensor in the data acquisition system synchronously and in real time acquire drilling parameters, and store the drilling parameters through the data processing system.

[0047] 4) Perform data preprocessing to unify the test set and training set, and establish an extreme learning machine classification model; determine the number of neurons in the hidden layer and the activation function; initialize the parameters of the improved marine predator algorithm, including the initial population size and the maximum number of iterations; calculate the fitness of samples in the population to obtain the optimal fitness and position; next, calculate the population fitness value after differential evolution perturbation; determine whether the iteration conditions are met, and if so, output the optimal weight and threshold; otherwise, repeat the above actions until the optimal solution is obtained; output the prediction results of the intelligent recognition model to achieve real-time identification of the roof lithology during drilling.

[0048] 5) Based on the lithology of the roof strata identified during drilling, locate the potential hazard areas in the roadway roof and provide accurate and targeted dynamic support for these areas.

[0049] The beneficial effects of this invention are as follows:

[0050] 1. The coal mine roadway roof lithology identification system of this invention, based on the MQT-130 / 3.2 pneumatic rock bolt drilling rig, includes a data acquisition system, a signal conversion system, a data processing system, an intelligent algorithm lithology identification system, and a drilling system. The data acquisition system collects drilling torque, drilling speed, drilling displacement, drilling pressure, and drilling vibration signals during the drilling process of the MQT-130 / 3.2 pneumatic rock bolt drilling rig; then, the signal conversion system converts the collected signals; the data processing system receives, stores, and processes the sensor signals converted by the signal conversion system; by acquiring drilling parameters in real time, the system establishes the response relationship between rock parameters and drilling parameters; and the intelligent algorithm lithology identification system performs lithology identification, real-time acquisition of the lithology of the roadway roof strata and the potential hazard areas in the roadway roof, increasing the accuracy and targeting of dynamic support for the roadway roof, promoting the intelligent development of underground equipment in coal mines, and greatly ensuring construction safety.

[0051] 2. In this invention, the opposing learning strategy and differential evolution strategy are used to improve the marine predator algorithm. The improved marine predator algorithm is used to optimize the initial weights and thresholds of the extreme learning machine. The improved algorithm has better performance and higher accuracy of the lithology prediction model.

[0052] 3. The drilling system set in this invention is not limited to the identification of roof lithology in coal mine roadways. After changing the drilling system to different models of anchor drilling rigs, it can be applied to the identification of surrounding rock lithology at different locations in coal mine roadways during drilling. Attached Figure Description

[0053] 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.

[0054] Figure 1 This is a schematic diagram of the structure of the coal mine roadway roof drilling lithology identification system of the present invention;

[0055] Figure 2 for Figure 1 Isometric side view of the data acquisition system and drilling system in China;

[0056] Figure 3 for Figure 1 A schematic diagram of the structure of the signal conversion system;

[0057] Figure 4 for Figure 1 Isometric side view of the explosion-proof signal box;

[0058] Figure 5 for Figure 1 Isometric side view of the drilling system;

[0059] Figure 6 This is a basic structural diagram of an Extreme Learning Machine;

[0060] Figure 7 Flowchart for improving the marine predator-extreme learning machine classification model;

[0061] In the diagram: 1-Data acquisition system and drilling system, 2-Explosion-proof signal box, 3-Data processing system;

[0062] 101-Torque-Speed ​​Sensor, 102-First Rigid Sleeve, 103-Second Rigid Sleeve, 104-Wire Rope Displacement Sensor, 105-Rotary Pressure Sensor, 106-Vibration Sensor;

[0063] 201-First power transformer, 202-Second power transformer, 203-Air switch, 204-Gateway, 205-First XSE5 digital display alarm controller, 206-Second XSE5 digital display alarm controller, 207-Signal conversion module, 208-First RS-485 serial port, 209-First RS-485 serial port;

[0064] 401-Body components, 402-Air leg components, 403-Operating arm components, 404-Motor transmission assembly, 405-Transmission spindle, 406-Air leg handle, 407-Top. Detailed Implementation

[0065] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0066] The purpose of this invention is to provide a system and method for identifying the lithology of the roof during drilling in coal mine roadways, in order to solve the problems existing in the prior art.

[0067] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0068] The coal mine roadway roof drilling-while-drilling lithology identification system in this embodiment, such as Figure 1 As shown, it includes a data acquisition system, a drilling system 1, an explosion-proof box (signal conversion system) 2, and a data processing system 3;

[0069] The data acquisition system is used to collect the rotational speed, torque, displacement, thrust, and vibration signals of the drilling system 1 during the drilling process of the roadway roof; the signal conversion system 2 is electrically connected to the data acquisition system and is used to receive and convert the sensor signals of the data acquisition system; the data processing system 3 is electrically connected to the signal conversion system 2 and is used to receive and store the sensor signals converted by the signal conversion system 2; the drilling system 1 provides drilling power for the entire system.

[0070] like Figure 2 and Figure 5 As shown, the data acquisition system includes a torque-speed sensor 101, a rope displacement sensor 104, a rotary pressure sensor 105, and a vibration sensor 106.

[0071] The torque-speed sensor 101 is used to collect torque and speed signals during the drilling process. The torque-speed sensor 101 is connected to the main shaft 405 of the transmission system of the drilling system through the first rigid sleeve 102 and to the drill rod through the second rigid sleeve 103.

[0072] The pull rope displacement sensor 104 is installed on the upper part of the air leg component 402 of the drilling system. The pull rope displacement sensor 104 is connected to the air leg handle 406 of the drilling system via a pull rope. During the drilling process, the pull rope rises and falls with the air leg 402 to monitor the displacement in real time.

[0073] A rotary pressure sensor 105 is located at the bottom of the drilling system and is connected to the top tip 407 of the drilling system through a rigid threaded sleeve to monitor the propulsion force during the drilling process.

[0074] Vibration sensor 106 is installed on the upper part of the motor drive assembly 404 of the drilling system and is connected by magnetic attraction to monitor vibration signals during the drilling process.

[0075] like Figure 3 As shown, the signal conversion system 2 includes a first power transformer 201, a second power transformer 202, an air switch 203, a junction box 204, a first XSE5 digital display alarm controller 205, a second XSE5 digital display alarm controller 206, a signal conversion module 207, a first RS-485 serial port 208, and a second RS-485 serial port 209, all connected to each other via signal lines. The signal conversion system 2 is connected to various sensors in the data acquisition system via signal lines to receive and process sensor signals.

[0076] The first power transformer 201 and the second power transformer 202 are used to change the voltage of the signal conversion system to meet the various working voltages of the sensor. The input terminal of the first power transformer 201 is connected to the external power supply, and the output terminal is connected to the air switch 203. The input terminal of the second power transformer 202 is connected to the junction box 204, and the output terminal is connected to the vibration sensor 106 and the signal conversion module 207 respectively.

[0077] The air switch 203 can protect against short circuits, severe overloads and undervoltages that occur in the signal conversion system 2, ensuring the safe operation of the signal conversion system 2. The input terminal of the air switch 203 is connected to the first power transformer 201, and the output terminal is connected to the junction box 204.

[0078] Junction box 204 branches the main line. The input end of junction box 204 is connected to air switch 203, and the output end branches out 4 groups of signal lines to the second power transformer 202, the first XSE5 digital display alarm controller 205, the second XSE5 digital display alarm controller 206, and the data processing system 3 respectively.

[0079] The first XSE5 digital display alarm controller 205 and the second XSE5 digital display alarm controller 206 have digital filtering and alarm functions, and can realize the measurement of physical quantities in vibration environments. The input terminal of the first XSE5 digital display alarm controller 205 is connected to the junction box 204, and the output terminal is connected to the displacement sensor 104 and the first RS-485 transmission serial port 208, respectively. The input terminal of the second XSE5 digital display alarm controller 206 is connected to the junction box 204, and the output terminal is connected to the rotary pressure sensor 105 and the first RS-485 transmission serial port 208, respectively.

[0080] The signal conversion module 207 can precisely amplify the output signal of the torque-speed sensor 101. Its input end is connected to the second power transformer 202, and its output end is connected to the torque-speed sensor 101 and the first RS-485 transmission serial port 208, respectively.

[0081] The input terminal of the first RS-485 serial port 208 is connected to the first XSE5 digital display alarm controller 205, the second XSE5 digital display alarm controller 206 and the signal conversion module 207 respectively, and the output terminal is connected to the data processing system 3.

[0082] The second RS-485 serial port 209 has its input terminal connected to the vibration sensor 106 and its output terminal connected to the data processing system 3.

[0083] like Figure 5 As shown, the drilling system includes a motor transmission component 404, a body component 401, a pneumatic leg component 402, and a control arm component 403; the motor transmission component 404 includes a motor assembly 404 and a transmission assembly 405; the motor assembly 404 provides power to the entire system; the pneumatic leg component 402 controls the lifting and lowering of the drilling system; the control arm component 403 is manually controlled to stabilize the anchor drilling machine body.

[0084] Based on the above-mentioned coal mine roadway roof drilling lithology identification system, the present invention also provides a coal mine roadway roof drilling lithology identification method, comprising the following steps:

[0085] 1) First, connect the drilling system 1 to the blower and water pump respectively to ensure the required air pressure, air volume and water pressure when the pneumatic anchor drilling machine is working. The oil mist lubricator of the pneumatic anchor drilling machine must not be short of oil to ensure that the pneumatic anchor drilling machine is adequately lubricated.

[0086] 2) Place the MQT-130 / 3.2 pneumatic anchor drill rig 1 vertically, with the pressure sensor 105 in contact with the ground, install the drill rod and raise the air leg 402 to a position where the drill rod is close to contacting the top plate, and start drilling;

[0087] 3) During the drilling process, the torque-speed sensor 101, the rope displacement sensor 104, the pressure sensor 105 and the vibration sensor 106 in the data acquisition system synchronously and in real time acquire drilling parameters, and store the drilling parameters through the data processing system 3.

[0088] 4) Perform data preprocessing to unify the test set and training set, and establish an extreme learning machine classification model; determine the number of neurons in the hidden layer and the activation function; initialize the parameters of the improved marine predator algorithm, including the initial population size and the maximum number of iterations; calculate the fitness of samples in the population to obtain the optimal fitness and position; next, calculate the population fitness value after differential evolution perturbation; determine whether the iteration conditions are met, and if so, output the optimal weight and threshold; otherwise, repeat the above actions until the optimal solution is obtained; output the prediction results of the intelligent recognition model to achieve real-time identification of the roof lithology during drilling.

[0089] 5) Based on the lithology of the roof strata identified during drilling, locate the potential hazard areas in the roadway roof and provide accurate and targeted dynamic support for these areas.

Claims

1. A coal mine roadway roof lithology identification system, characterized in that: It includes a data acquisition system, a signal conversion system, a data processing system, an intelligent algorithm lithology identification system, and a drilling system; The data acquisition system is used to collect torque, rotational speed, displacement, thrust, and vibration signals during the drilling process of the drilling system on the roadway roof. The signal conversion system is electrically connected to the data acquisition system and is used to receive and convert sensor signals from the data acquisition system. The data processing system is electrically connected to the signal conversion system and is used to receive, store, and process signals transmitted by the conversion system. The intelligent algorithm lithology identification system uses a metaheuristic algorithm to optimize the extreme learning machine and constructs an improved marine predator-extreme learning machine lithology identification model to identify the lithology of the roof. The drilling system provides drilling power for the entire system; The intelligent algorithm lithology identification system includes an algorithm optimization module, a data preprocessing module, and a lithology intelligent identification module; The algorithm optimization module addresses the problems of low solution accuracy, slow convergence speed, and difficulty in escaping local optima in metaheuristic algorithms. It introduces opposition learning and differential evolution strategies to improve the metaheuristic algorithm and uses the improved metaheuristic algorithm to optimize the extreme learning machine. The data preprocessing module includes outlier removal and normalization of the training and test sets. Intelligent identification requires the establishment of a lithology identification database of drilling parameters of a certain size. This data is collected through the aforementioned data acquisition system. Due to non-standard factors in the operation of the drilling system, a small amount of abnormal data may be generated. This data must be removed to prevent it from entering the database. At the same time, to avoid significant differences in similar drilling parameters of the same type of roof rock layer caused by different roof rock layers, different types of anchor drilling rigs, and different personnel operations, the input drilling parameters are normalized to avoid affecting the accuracy of the algorithm. The dataset of the lithology intelligent identification module is divided into a training set and a test set. The training set is established based on existing sample data and the results of underground geological exploration and roof strata drilling. The input variables of the training set and the test set are the drilling pressure, drilling torque, drilling speed and drilling displacement after the collected data are normalized. For ease of analysis, the output variables of the training set and the test set are set to three: coal (1), mudstone (2) and sandstone (3). The metaheuristic algorithm is the Ocean Predator Algorithm; The oppositional learning strategy is as follows: Δ={x_1,x_2〖,…,x〗_d}; Δ - ={x-_1,x-_2〖,…,x-〗_d}; x-_j = lb_j + ub_j - x_j; In the above formula, Δ represents the position of a certain d-dimensional individual. - The opposite side is defined by lb and ub, which are the upper and lower bounds of the individual's position. The basic mutation strategy of the differential evolution algorithm is as follows: X_i(t+1)=X_r1(t)+F·(X_r2(t)-X_r3(t)); In the above formula, r_1, r_2, and r_3 are distinct integers within the population range and are not equal to i; F is a constant factor, F∈[0,2].

2. The coal mine roadway roof drilling lithology identification system according to claim 1, characterized in that: The data acquisition system includes a torque-speed sensor, a rope displacement sensor, a pressure sensor, and a vibration sensor; The torque-speed sensor is used to collect torque and speed signals during the drilling process. The torque-speed sensor is connected to the transmission spindle of the drilling system through a first rigid sleeve and to the drill rod through a second rigid sleeve. The pull rope displacement sensor is installed on the upper part of the air leg component of the drilling system. The pull rope displacement sensor is connected to the air leg handle of the drilling system through a pull rope. During the drilling process, the pull rope rises and falls with the air leg to monitor the displacement in real time. The pressure sensor is located at the bottom of the drilling system and is connected to the top of the drilling system through a rigid threaded sleeve to monitor the propulsion force during the drilling process. The vibration sensor is installed on the upper part of the motor drive assembly of the drilling system and is connected by magnetic attraction to monitor vibration signals during the drilling process.

3. The coal mine roadway roof drilling lithology identification system according to claim 2, characterized in that: The signal conversion system includes a coal mine safety explosion-proof signal box, a first power transformer, a second power transformer, an air switch, a junction box, a first XSE5 digital display alarm controller, a second XSE5 digital display alarm controller, a signal conversion module, a first RS-485 serial port, and a second RS-485 serial port. The signal conversion system is connected to various sensors in the data acquisition system via signal lines to receive and process sensor signals. The first and second power transformers are used to change the voltage of the signal conversion system to meet the various operating voltages of the sensor; the input terminal of the first power transformer is connected to an external power source, and the output terminal is connected to an air switch; the input terminal of the second power transformer is connected to the junction box, and the output terminal is connected to the vibration sensor and the signal conversion module respectively. The air switch integrates control and multiple protection functions, and can protect against short circuits, severe overloads and undervoltages in the signal conversion system, ensuring the safe operation of the signal conversion system. The input terminal of the air switch is connected to the first power transformer, and the output terminal is connected to the junction box. The junction box is a 3-in-12-out junction box that branches the main line to output more voltage signals. The input end of the junction box is connected to the air switch, and the output end branches out 4 groups of signal lines that are respectively connected to the second power transformer, the first XSE5 digital display alarm controller, the second XSE5 digital display alarm controller, and the data processing system. The first and second XSE5 digital display alarm controllers use fully isolated 18-bit AD conversion for input, with a measurement accuracy of 0.55%FS, and have digital filtering and alarm functions, enabling the measurement of physical quantities in vibration environments. The input terminal of the first XSE5 digital display alarm controller is connected to the junction box, and its output terminal is connected to the pull rope displacement sensor and the first RS-485 serial port, respectively. The input terminal of the second XSE5 digital display alarm controller is connected to the junction box, and its output terminal is connected to the pressure sensor and the first RS-485 serial port, respectively. The signal conversion module is used to precisely amplify the output signal of the torque-speed sensor, perform voltage stabilization and impedance matching inside the circuit, and its input end is connected to the second power transformer. Its output end is connected to the torque-speed sensor and the first RS-485 transmission serial port, respectively. The first RS-485 serial port input terminal is connected to the first XSE5 digital display alarm controller, the second XSE5 digital display alarm controller, and the signal conversion module, respectively, and the output terminal is connected to the data processing system; The second RS-485 serial port input is connected to the vibration sensor, and the output is connected to the data processing system.

4. The coal mine roadway roof drilling lithology identification system according to claim 3, characterized in that: The data processing system is an explosion-proof laptop computer used to receive and store signals transmitted by the signal conversion system; The explosion-proof laptop is connected to the junction box, the first RS-485 serial port and the second RS-485 serial port in the signal conversion system.

5. The coal mine roadway roof drilling lithology identification system according to claim 1, characterized in that: The Extreme Learning Machine (ELM) is an algorithm based on a single hidden layer feedforward neural network structure. It randomly selects input layer weights and hidden layer biases, and obtains the output layer weights through generalized matrix theory, thereby calculating the predicted values ​​of the data. The parameters that need to be manually set for the ELM include: the number of input layer nodes, the number of hidden layer neurons, the number of output layer nodes, and the activation function.

6. The coal mine roadway roof drilling lithology identification system according to claim 1, characterized in that: The initial weights and thresholds randomly generated by the extreme learning machine are blind. The improved opposition-difference ocean predator algorithm is used to optimize the initial weights and thresholds to improve prediction accuracy.

7. The coal mine roadway roof drilling lithology identification system according to claim 1, characterized in that: The original dataset contains 1800 data points. The ratio of training set to test set is set to 8:2, with 1440 data points in the training set and 360 data points in the test set.

8. The coal mine roadway roof drilling lithology identification system according to claim 2, characterized in that: The drilling system includes an MQT-130 / 3.2 pneumatic rock bolt drill and drill rods. The MQT-130 / 3.2 series pneumatic rock bolt drill is used for rock bolt support operations in coal roadways with a rock hardness coefficient f≤10, and is used for drilling roof rock bolt holes and anchor cable holes. The MQT-130 / 3.2 pneumatic anchor bolt drilling rig includes a motor transmission component, a body component, an air leg component, and a control arm component. The motor transmission component includes a motor assembly, a transmission assembly, and a water jacket assembly, with the motor assembly providing power to the entire system. The transmission assembly connects to the drill rod and rotates as a whole with the drill rod. The water jacket assembly connects to a water pipe to supply water during the drilling process. The air leg component controls the lifting and lowering of the drilling system. The control arm component allows for manual control to stabilize the anchor bolt drilling rig body. The drill pipe is connected to the torque-speed sensor via a second rigid sleeve.

9. The coal mine roadway roof drilling lithology identification system according to claim 8, characterized in that: The drill pipe has a connection port in the middle, and multiple drill pipes can be connected to each other through the connection port to meet the requirements of different drilling depths.

10. A method for identifying the lithology of the roof during drilling in a coal mine roadway, applicable to any one of claims 8 and 9, characterized in that, Includes the following steps: 1) First, connect the MQT-130 / 3.2 pneumatic anchor drill to the blower and water pump respectively to ensure the required air pressure, air volume and water pressure when the pneumatic anchor drill is working. The oil mist lubricator of the pneumatic anchor drill must not be short of oil to ensure that the pneumatic anchor drill is adequately lubricated. 2) Place the MQT-130 / 3.2 pneumatic anchor drill vertically, with the pressure sensor in contact with the ground, install the drill rod, and raise the air leg to a position close to the contact point between the drill rod and the top plate to begin drilling; 3) During the drilling process, the torque-speed sensor, cable displacement sensor, pressure sensor and vibration sensor in the data acquisition system synchronously and in real time acquire drilling parameters, and store the drilling parameters through the data processing system. 4) Perform data preprocessing, normalize the training and test sets, and establish an extreme learning machine classification model; Determine the number of neurons in the hidden layer and the activation function; Initialize the parameters of the improved marine predator algorithm, including the initial population size and the maximum number of iterations; Calculate the fitness of samples in the population to obtain the optimal fitness and position; The next step is to calculate the fitness value of the population after differential evolution perturbation; Determine whether the iteration condition is met; if so, output the optimal weight and threshold. Otherwise, return to the steps of calculating the fitness of samples in the population, obtaining the optimal fitness and position, calculating the population fitness value after differential evolution perturbation, and determining whether the iteration conditions are met, until the iteration conditions are met and the optimal solution is obtained; Output the prediction results of the intelligent recognition model to achieve real-time identification of the roof lithology during drilling; 5) Based on the lithology of the roof strata identified during drilling, locate the potential hazard areas in the roadway roof and provide accurate and targeted dynamic support for these areas.

Images