Device for detecting composite rock stratum structure while drilling, and intelligent identification method

Abstract

A device for detecting a composite rock stratum structure while drilling, and an intelligent identification method. The device comprises an apparatus base (8), wherein a vertically arranged supporting rod is fixedly connected to the middle portion of the apparatus base (8), a drilling hydraulic oil cylinder (4) is fixedly connected on one side of the bottom of the supporting rod, and a piston end of the drilling hydraulic oil cylinder (4) is fixedly connected to an adaptive while-drilling data acquisition module (5) and is in sliding connection with the supporting rod; a sample table (1) is arranged above the adaptive while-drilling data acquisition module (5) and is fixedly connected to the supporting rod, and a top limiting system (3) is fixedly connected to the top end of the supporting rod; and a rock gripper (2) is arranged between the sample table (1) and the top limiting system (3), and the adaptive while-drilling data acquisition module (5), a data acquisition device (9) and a data analysis terminal (10) are all electrically connected to each other. A real mechanical environment of a coal mine site is reproduced, multiple parameters are acquired by means of the adaptive while-drilling data acquisition module (5), and the property parameters of a composite rock stratum are represented by using a machine learning method, so as to achieve the intelligent identification of rock lithology, such that support can be provided for intelligent control of surrounding rock of a coal mine roadway.

Description

Composite rock structure detection while drilling device and intelligent identification method Technical Field

[0001] The present invention belongs to the technical field of surrounding rock sensing during coal mine tunneling, and in particular relates to a composite rock layer structure detection while drilling device and an intelligent identification method. Background Art

[0002] Coal seam tunnels are subject to significant geology fluctuations. Simplified support systems in long tunnels often result in excessive or insufficient support strength, leading to wasted support costs and reduced tunnel maintenance and control effectiveness. Furthermore, the development of intelligent mining faces has made the issue of mining-excavation transitions increasingly prominent. Therefore, transparent visualization of the rock mass structure in coal tunnel roofs has become a key issue for rapid, intelligent tunneling in coal tunnels.

[0003] Borehole imaging and microseismic monitoring, the current mainstream technologies for detecting coal tunnel roof structures, are limited by the size of the equipment and construction efficiency, making them incapable of fully detecting roof structures and failing to meet the specific requirements of rapid coal tunneling. While-drilling (LWD), a simple and convenient detection method, was initially used in the oil and large-scale geological exploration sectors. In recent years, driven by the development of intelligent tunneling, it has been gradually introduced to the coal mining sector. This technology primarily collects parameters such as torque, speed, and thrust during drilling, thereby establishing a mapping relationship between different drilling parameters and the rock mass.

[0004] CN114000866A proposes a device and method for testing rock mass structural characteristics while drilling. This device collects drilling process parameters through a rock mass fixing device and a drilling part. However, the device cannot set the confining pressure and cannot realize multi-source collection of drilling parameters. Therefore, it lacks a true simulation of the actual engineering conditions. CN112145151A proposes a dynamic detection and assessment method for impact hazard based on drilling response parameters. It is mainly used to reasonably analyze different parameters during drilling to realize the stress environment of the coal seam where drilling is carried out, and is used for impact hazard assessment. This is similar to the experimental method and device for measuring coal mass stress while drilling proposed by CN114017029B. Both cannot detect and identify rock mass properties. CN115059449A proposes a self-identifying multi-parameter near-drill bit while drilling device and method for measuring while drilling in coal mines. It is mainly used for large-scale drilling equipment such as large-diameter pressure relief drills in coal mines. It is not suitable for equipment such as anchor drills used in coal roadway roof support. CN115077607A proposes a system and method for intelligent identification of surrounding rock properties while drilling in coal mine tunnels. This system uses a designed experimental setup to intelligently identify surrounding rock properties. However, this system has three major shortcomings: 1. The drilling posture is horizontal, which is inconsistent with the actual operating posture of a roof bolting rig, affecting the accuracy of drilling parameter identification; 2. It cannot apply confining pressure, significantly different from the actual working conditions of the surrounding rock underground; and 3. The drill bit used in the drill rig is inconsistent with conventional double-wing polycrystalline diamond (PDC) drill bits. Therefore, this method is not suitable for identifying the lithology of surrounding rock in coal tunnels.

[0005] The mainstream downhole detection devices in the current coal mining field have the following main problems: 1. The current functions are mainly focused on surrounding rock stress and gas detection, and there are few downhole detection equipment for rock properties; 2. The drill bit and drilling rig working conditions of the experimental device are simulated for a large-diameter pressure relief drill, which is inconsistent with the drilling size of the actual anchor drill; 3. The drilling posture of the existing downhole detection experimental device is mainly horizontal, which is inconsistent with the drilling posture of the coal roadway anchor drill; 4. The existing experimental device usually fixes the experimental rock block with linear constraints and cannot apply confining pressure, which is quite different from the underground rock stress field environment.

[0006] Therefore, it is necessary to design a composite rock structure while-drilling detection device and an intelligent identification method to solve the above problems.

[0007] Summary of the Invention

[0008] The purpose of the present invention is to provide a composite rock structure detection while drilling device and an intelligent identification method to solve the above problems, achieve the simulation and restoration of the real stress field environment, and realize the purpose of identifying and distinguishing lithology through the response characteristics of different parameters during drilling of different rocks.

[0009] To achieve the above-mentioned purpose, the present invention provides the following scheme: a composite rock formation structure downhole detection device, comprising an equipment base, a vertically arranged support rod fixedly connected to the middle of the top of the equipment base, a drilling hydraulic cylinder fixedly connected to one side of the bottom of the support rod, an adaptive downhole data acquisition module fixedly connected to the piston end of the drilling hydraulic cylinder, the adaptive downhole data acquisition module is slidably connected to the support rod, a sample table is correspondingly arranged above the adaptive downhole data acquisition module, the sample table is fixedly connected to the support rod, a top limit system is fixedly connected to the top of the support rod, a rock clamp is arranged between the sample table and the top limit system, the adaptive downhole data acquisition module is electrically connected to the data acquisition device, and the data acquisition device is electrically connected to the data analysis terminal.

[0010] Preferably, the rock clamp includes a bottom plate, the middle of the bottom plate is provided with a square through hole, the four corners of the bottom end of the bottom plate are respectively provided with universal rollers, the universal rollers are in rolling contact with the top of the sample table, the four corners of the top end of the bottom plate are respectively fixedly connected to a supporting hydraulic cylinder, the piston end of the supporting hydraulic cylinder passes through the bottom plate and is fixedly connected to the supporting column of the rock sample clamping device, the bottom end of the supporting column of the rock sample clamping device abuts against the top end of the sample table, the four supporting columns of the rock sample clamping device are located between the four universal rollers, the top end of the bottom plate is fixedly connected to two fixed limiting steel plates parallel to each other and vertically arranged, the two fixed limiting steel plates are located between the four supporting hydraulic cylinders, and two hydraulic cylinder steel plates parallel to each other and vertically arranged are fixedly connected between the two fixed limiting steel plates, the hydraulic cylinder steel plate and the top of the fixed limiting steel plate are jointly fixedly connected to the bottom end of the top plate, and the top end of the top plate abuts against the bottom of the top limit system.

[0011] Preferably, the adaptive downhole data acquisition module includes a fixed plate, the bottom end of the fixed plate is fixedly connected to the piston end of the drilling hydraulic cylinder, one side of the fixed plate is fixedly connected to a drilling guide rail groove, the drilling guide rail groove is slidably connected to the support rod, the bottom of the other side of the fixed plate is fixedly connected to a speed acquisition unit, the output end of the speed acquisition unit is fixedly connected to one end of the torque acquisition unit, the other end of the torque acquisition unit is fixedly connected to a drill rod groove, an axial force acquisition unit is arranged between the drill rod groove and the torque acquisition unit, and the axial force acquisition unit, the torque acquisition unit, and the speed acquisition unit are electrically connected to the data acquisition device.

[0012] Preferably, the sample table includes a sample table plane, the sample table plane is fixedly connected to the support rod, a drilling hole is opened in the middle of the sample table plane, the universal roller is in rolling contact with the sample table plane, the bottom end of the rock sample clamping device support column is in contact with the top of the sample table plane, rock sample placement entrances are respectively provided on both sides of the top of the sample table plane, and a sample table guardrail is fixedly connected to one side wall of the sample table plane.

[0013] Preferably, the bottom side wall of the support rod is fixedly connected to a guide rail, and the drilling guide rail groove is slidably matched with the guide rail.

[0014] The experimental method of the composite rock formation structure detection while drilling device includes the following steps:

[0015] S1. Prepare rock samples of different strengths;

[0016] S2. Testing the basic parameters of the rock sample in S1;

[0017] S3, placing the rock sample in S2 in the rock holder and fixing it;

[0018] S4, controlling the adaptive while-drilling data acquisition module to drill the rock sample and monitor parameter changes during the drilling process;

[0019] S5, after drilling is completed, the adaptive while-drilling data acquisition module is removed from the rock sample;

[0020] S6. Repeat S2-S5 to conduct experiments on rock samples of different strengths and composites;

[0021] S7, repeating S2-S5, setting different confining pressures, and conducting experiments on the rock samples with different strengths;

[0022] S8. Repeat S2-S5 to conduct experiments on the rock samples with the same strength but different damage degrees;

[0023] S9. Establish and analyze a database based on experimental data, and use machine learning to train and learn with the database as the target to achieve intelligent recognition of roof structure.

[0024] Preferably, in S1, the rock samples of different strengths are prepared according to similarity criteria or based on in-situ rocks.

[0025] Preferably, the basic parameters in S2 include uniaxial compressive strength, elastic modulus, internal friction angle, and Poisson's ratio.

[0026] Preferably, the axial force acquisition unit, the speed acquisition unit, and the torque acquisition unit in the adaptive while-drilling data acquisition module are positively correlated with the rock strength. The preliminary mapping relationship between rock strength and drilling parameters is shown in the formula, and the mapping relationship is integrated into the machine learning algorithm to achieve intelligent identification of rock types. In formula (1), E represents rock strength, i represents the number of data, k is the empirical coefficient, N is the speed, M is the torque, F is the drilling force, a is the empirical coefficient of the drilling force, and b is a constant.

[0027] Compared with the prior art, the present invention has the following advantages and technical effects:

[0028] (1) The present invention can simulate the drill bit size and drilling posture of a real anchor drill in a coal tunnel, and simultaneously realize the application of confining pressure on rock samples, thus realistically simulating the stress state of the surrounding rock;

[0029] (2) By designing different gradients of torque, rotation speed, drilling speed, thrust and confining pressure, the response laws of rock masses with different strengths, different damage degrees and different numbers of cracks during the drilling process can be realized;

[0030] (3) Based on the drilling data, the mapping relationship between rock masses with different strengths, damage degrees, and crack numbers and different drilling parameters can be obtained by inversion, which can ultimately be used for dynamic identification of rock mass properties.

[0031] (4) Based on the adaptive while-drilling data acquisition module, the rapid improvement of the coal mine underground anchor drilling rig can quickly realize the field application of while-drilling detection of composite rock formation structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive work.

[0033] FIG1 is a schematic structural diagram of an experimental device of the present invention;

[0034] FIG2 is a schematic structural diagram of a rock holder according to the present invention;

[0035] FIG3 is a schematic structural diagram of an adaptive while-drilling data acquisition module according to the present invention;

[0036] FIG4 is a schematic diagram of the structure of the sample table of the present invention.

[0037] Among them, 1. Sample table; 11. Entrance for placing rock samples; 12. Drilling hole; 13. Sample table guardrail; 14. Sample table plane; 2. Rock clamp; 21. Rock sample; 22. Universal roller; 23. Support column for rock sample clamping device; 24. Hydraulic cylinder steel plate; 25. Fixed limit steel plate; 26. Support hydraulic cylinder; 3. Top limit system; 4. Drilling hydraulic cylinder; 5. Adaptive downhole data acquisition module; 51. Drill rod slot; 52. Axial force acquisition unit; 53. Torque acquisition unit; 54. Speed ​​acquisition unit; 55. Drilling guide rail slot; 6. Displacement sensor; 7. Guide rail; 8. Equipment base. DETAILED DESCRIPTION

[0038] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0039] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0040] 1 to 4 , the present invention provides a drilling-while-drilling detection device for a composite rock formation structure, comprising an equipment base 8, wherein a vertically arranged support rod is fixedly connected to the middle of the top of the equipment base 8, a drilling hydraulic cylinder 4 is fixedly connected to one side of the bottom of the support rod, an adaptive drilling-while-drilling data acquisition module 5 is fixedly connected to the piston end of the drilling hydraulic cylinder 4, the adaptive drilling-while-drilling data acquisition module 5 is slidably connected to the support rod, a sample table 1 is correspondingly arranged above the adaptive drilling-while-drilling data acquisition module 5, the sample table 1 is fixedly connected to the support rod, a top limit system 3 is fixedly connected to the top of the support rod, a rock clamp 2 is arranged between the sample table 1 and the top limit system 3, the adaptive drilling-while-drilling data acquisition module 5 is electrically connected to a data acquisition device 9, and the data acquisition device 9 is electrically connected to a data analysis terminal 10.

[0041] Furthermore, a displacement sensor 6 is fixedly connected to the bottom end of the support rod for monitoring the displacement value.

[0042] The data acquisition device 9 can realize wireless and efficient transmission of drilling data through Bluetooth and Wireless Fidelity technology, which solves the problem of cumbersome data transmission underground in coal mines. The data analysis terminal 10 can realize intelligent analysis and rock formation identification of drilling data.

[0043] A further optimization scheme is provided, the rock clamp 2 includes a base plate, a square through hole is opened in the middle of the base plate, and universal rollers 22 are respectively provided at the four corners of the bottom end of the base plate, and the universal rollers 22 are in rolling contact with the top of the sample table 1, and the four corners of the top end of the base plate are respectively fixedly connected to support hydraulic cylinders 26, and the piston end of the support hydraulic cylinder 26 passes through the base plate and is fixedly connected to the rock sample clamping device support column 23, and the bottom end of the rock sample clamping device support column 23 abuts the top end of the sample table 1, and the four rock sample clamping device support columns 23 are located between the four universal rollers 22, and the top end of the base plate is fixedly connected to two fixed limit steel plates 25 arranged parallel to each other and vertically, and the two fixed limit steel plates 25 are located between the four support hydraulic cylinders 26, and two hydraulic cylinder steel plates 24 arranged parallel to each other and vertically are fixedly connected between the two fixed limit steel plates 25, and the top ends of the hydraulic cylinder steel plates 24 and the fixed limit steel plates 25 are jointly fixedly connected to the bottom end of the top plate, and the top end of the top plate abuts the bottom end of the top limit system 3.

[0044] The hydraulic cylinder steel plate 24 and the fixed limit steel plate 25 are used to realize the application of confining pressure to the rock sample 21; the universal roller 22 is relied upon to realize the free movement of the rock clamp 2 on the sample table 1; and the rock clamp 2 is fixed between the top limit system 3 and the sample table 1 by supporting the hydraulic cylinder 26.

[0045] The rock holder 2 can hold rocks with a size range of 50-200 mm in length, 50-200 mm in width, and 100-600 mm in height.

[0046] Further optimization scheme, the adaptive drilling data acquisition module 5 includes a fixed plate, the bottom end of the fixed plate is fixedly connected to the piston end of the drilling hydraulic cylinder 4, one side of the fixed plate is fixedly connected to the drilling guide groove 55, the drilling guide groove 55 is slidably connected to the support rod, the bottom of the other side of the fixed plate is fixedly connected to the speed acquisition unit 54, the output end of the speed acquisition unit 54 is fixedly connected to one end of the torque acquisition unit 53, the other end of the torque acquisition unit 53 is fixedly connected to the drill rod groove 51, an axial force acquisition unit 52 is arranged between the drill rod groove 51 and the torque acquisition unit 53, the axial force acquisition unit 52, the torque acquisition unit 53, the speed acquisition unit 54 and the data acquisition device 9 are electrically connected.

[0047] The adaptive while-drilling data acquisition module 5 moves as a whole on the support rod, generating displacement. The rotation speed of the drilling rod body in the drill rod slot 51 is provided by the rotation speed acquisition unit 54, and finally the data changes are monitored by the axial force acquisition unit 52 and the torque acquisition unit 53.

[0048] The drilling power part of the adaptive drilling data acquisition module 5 can realize the drilling of mining B19, B22 and any conventional drill rod and related drill bit with a diameter of more than 22 mm, and is widely used.

[0049] A further optimized solution is provided in which the sample table 1 includes a sample table plane 14, which is fixedly connected to the support rod. A drilling hole 12 is provided in the middle of the sample table plane 14, the universal roller 22 is in rolling contact with the sample table plane 14, the bottom end of the rock sample clamping device support column 23 is in contact with the top of the sample table plane 14, and rock sample entrances 11 are respectively provided on both sides of the top of the sample table plane 14, and a sample table guardrail 13 is fixedly connected to one side wall of the sample table plane 14.

[0050] The rock holder 2 is moved from the middle of the sample table 1 to above the rock sample inlets 11 on both sides via the universal rollers 22. The rock sample 21 is placed into the rock holder 2 from the rock sample inlets 11, and then moved to above the drilling hole 12 via the universal rollers 22 to prepare for the next experiment. The sample table guardrail 13 provides protection for the safe movement of the rock holder 2 on the sample table plane 14.

[0051] To further optimize the solution, the bottom side wall of the support rod is fixedly connected to the guide rail 7, and the guide rail groove 55 is drilled into the guide rail groove 55 to slide with the guide rail 7.

[0052] The experimental method of the composite rock formation structure detection while drilling device includes the following steps:

[0053] S1. Prepare rock samples 21 of different strengths. The rock samples 21 can be cast in the laboratory using different lime-sand ratios, or can be drilled from typical composite roofs on-site in a coal mine.

[0054] S2, testing the basic parameters of the rock sample 21 in S1;

[0055] S3, placing the rock sample 21 in S2 in the rock holder 2 and fixing it;

[0056] Place the rock sample 21 into the sample slot from the rock sample entrances 11 on both sides of the sample table 1, apply confining pressure to the rock sample 21 in the X and Y directions through the rock clamp 2, move the rock clamp 2 to the middle of the sample table 1, operate the control device to start the rock sample clamping device support column 23, and allow the rock clamp 2 to leave the sample table 1 under the action of the universal roller 22.

[0057] S4, controlling the adaptive while-drilling data acquisition module 5 to drill the rock sample 21 and monitoring changes in parameters such as torque and axial force during the drilling process;

[0058] The drill rod and drill bit are installed on the drill rod slot 51. The electro-hydraulic servo control system is used to activate the speed acquisition unit 54 and the propulsion displacement and control the real-time parameters, so that the drill bit begins drilling. The parameters during the drilling process are collected through torque, axial force, displacement and speed sensors. When the drilling rig advances, the drilling hydraulic cylinder 4 moves in the Z direction to realize the overall movement of the adaptive drilling data acquisition module 5, giving the drill bit an initial propulsion force. During the drilling process, the speed acquisition unit 54 sets the rated speed of the drilling spindle, and the speed sensor and torque acquisition unit 53 monitor the real-time changes of the drilling process parameters.

[0059] S5, drilling is completed, and the adaptive while-drilling data acquisition module 5 is removed from the rock sample 21;

[0060] When drilling is completed, the displacement of the adaptive while-drilling data acquisition module 5 is stopped. By controlling the displacement, the adaptive while-drilling data acquisition module 5 is retreated, driving the drill bit of the drill pipe to withdraw from the rock sample 21 .

[0061] S6, repeat S2-S5 and conduct experiments on rock samples 21 of different strengths and composites;

[0062] Configure rock samples 21 of different strengths and composites, use the experimental device to repeat S2-S5, record the working parameters of torque, axial force, displacement and speed during the drilling process of rock samples 21 of different strengths and composites, study the response characteristics of the main drilling parameters during the drilling process of rock masses of different strengths and composites, and establish a mapping relationship between the two to facilitate the realization of while-drilling detection of rock structures.

[0063] S7, repeat S2-S5, set different confining pressures, and conduct experiments on rock samples 21 with different strengths;

[0064] By changing the confining pressure with different gradients and using rock samples 21 with different strengths, repeating S2-S5, the drilling response characteristics of rock samples 21 with different strengths under the influence of confining pressure were simulated and studied. The relationship between the surrounding rock stress environment and rock strength identification was studied, which provided a reference for the identification of surrounding rock properties at different depths.

[0065] S8, repeat S2-S5 and conduct experiments on rock samples 21 with the same strength and different damage degrees;

[0066] Use rock samples 21 with the same strength but different damage degrees, or prefabricate different structures and numbers of cracks in the rock sample 21. Repeat S2-S5 using the experimental device, record the response patterns of different drilling parameters in rock samples 21 with different damage degrees, and simulate the while-drilling detection and identification of rock damage degree / cracks.

[0067] S9. Establish and analyze a database based on experimental data, and use machine learning to train and learn with the database as the target to achieve intelligent recognition of roof structure.

[0068] Based on the massive amount of drilling experimental data obtained, a corresponding database is established, and structured processing and analysis are carried out based on data statistics. Artificial intelligence methods such as machine learning are used to carry out artificial intelligence training and learning with the database as the target, and ultimately realize the intelligent identification of roof structure.

[0069] The present invention uses a deep learning model-based end-to-end regression algorithm. The algorithm is trained using multiple data sources such as continuous value data and geological images to achieve high-precision prediction of roof structure.

[0070] The specific implementation method is as follows:

[0071] Data preparation: Continuous value data is collected during the drilling process, and combined with data sources such as geological images to generate training and test datasets.

[0072] Model Design: We use a combination of convolutional neural networks (CNNs) and long short-term memory (LSTM) networks to build an end-to-end regression model. The input layer accepts a variety of data sources, including continuous-valued data and geological images. After a series of convolution, pooling, and LSTM processing, the final output is the prediction result.

[0073] Model training: The model is trained using a training dataset and validated using a test dataset. During training, techniques such as data augmentation and regularization can be used to improve the model's robustness and generalization capabilities.

[0074] Model optimization: Further optimize the performance of the model, including model structure adjustment, hyperparameter optimization, ensemble learning and other methods to improve the prediction accuracy and stability of the model.

[0075] Model application: Deploy the trained model to actual drilling exploration to predict roof structure. Adjust and optimize based on the prediction results, and continuously iterate and optimize the algorithm and model.

[0076] The specific structure of the combination of convolutional neural network and long short-term memory network is:

[0077] Input layer: accepts various data sources such as continuous value data and geological images.

[0078] Convolutional layer 1: Use a 3x3 convolution kernel to extract underlying features and output a feature map of 64 channels.

[0079] Max pooling layer 1: Performs a max pooling operation on the output of convolutional layer 1 to reduce the size of the feature map.

[0080] Convolutional layer 2: Use a 3x3 convolution kernel to further extract intermediate layer features and output a feature map of 128 channels.

[0081] Max pooling layer 2: Performs a max pooling operation on the output of convolutional layer 2.

[0082] LSTM layer 1: A 128-unit LSTM layer is used to process the output of pooling layer 2 and retain the time series information of the intermediate layer features.

[0083] Dropout layer: To prevent overfitting, the output of LSTM layer 1 is randomly deactivated.

[0084] LSTM layer 2: Continue to use the 128-unit LSTM layer to process the output of the Dropout layer and further fuse the time series information.

[0085] Fully connected layer 1: After flattening the output of LSTM layer 2, it passes through fully connected layer 1 for dimension transformation and feature mapping, outputting a 256-dimensional feature vector.

[0086] Fully connected layer 2: The fully connected layer is used again to map the feature vector to the final prediction result space and output continuous values ​​to achieve the while-drilling detection results of the roof structure.

[0087] To further optimize the scheme, rock samples 21 of different strengths are prepared in S1 according to similarity criteria or based on in-situ rocks.

[0088] The similarity criterion is to simulate the uniaxial compressive strength of different types of rocks by pouring concrete-like rock materials, so as to achieve similarity in uniaxial compressive strength between the cast specimens and the real rocks. Usually, it is necessary to cast 100*100*100mm specimens according to different proportions of water, cement and river sand to carry out uniaxial compressive tests and measure the compressive strength. If the error value with the uniaxial compressive strength of the real rock is within 10%, it is considered to meet the similarity requirements.

[0089] To further optimize the scheme, the basic parameters in S2 include uniaxial compressive strength, elastic modulus, internal friction angle, and Poisson's ratio.

[0090] Laboratory basic mechanical tests were used to determine basic parameters such as uniaxial compressive strength, elastic modulus, internal friction angle and Poisson's ratio of rock samples 21 with different strengths.

[0091] To further optimize the solution, the axial force acquisition unit 52, the speed acquisition unit 54, and the torque acquisition unit 53 in the adaptive while-drilling data acquisition module 5 are positively correlated with the rock strength. The preliminary mapping relationship between rock strength and drilling parameters is shown in Formula 1. The mapping relationship is integrated into the machine learning algorithm to achieve intelligent identification of rock types. In Formula 1, E represents rock strength, i represents the number of data, k is the empirical coefficient, N is the speed, M is the torque, F is the drilling force, a is the empirical coefficient of the drilling force, and b is a constant.

[0092] In the description of the present invention, it should be understood that the terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating the orientation or position relationship, are based on the orientation or position relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

Claims

1. A composite rock formation structure detection device while drilling, characterized in that: The invention comprises a data analysis terminal (10), a data acquisition device (9), and an equipment base (8). A vertically arranged support rod is fixedly connected to the middle of the top of the equipment base (8), a drilling hydraulic cylinder (4) is fixedly connected to one side of the bottom of the support rod, an adaptive drilling data acquisition module (5) is fixedly connected to the piston end of the drilling hydraulic cylinder (4), the adaptive drilling data acquisition module (5) is slidably connected to the support rod, a sample table (1) is correspondingly arranged above the adaptive drilling data acquisition module (5), the sample table (1) is fixedly connected to the support rod, a top limit system (3) is fixedly connected to the top of the support rod, a rock clamp (2) is arranged between the sample table (1) and the top limit system (3), the adaptive drilling data acquisition module (5) is electrically connected to the data acquisition device (9), and the data acquisition device (9) is electrically connected to the data analysis terminal (10).

2. The composite rock formation structure detection while drilling device according to claim 1, characterized in that: The rock clamp (2) comprises a bottom plate, a square through hole is provided in the middle of the bottom plate, universal rollers (22) are respectively arranged at the four corners of the bottom end of the bottom plate, the universal rollers (22) are in rolling contact with the top end of the sample table (1), the four corners of the top end of the bottom plate are respectively fixedly connected with supporting hydraulic cylinders (26), the piston end of the supporting hydraulic cylinder (26) passes through the bottom plate and is fixedly connected with a rock sample clamping device support column (23), the bottom end of the rock sample clamping device support column (23) is in contact with the top end of the sample table (1), and the four rock sample clamping device support columns (23) are respectively fixedly connected with the top end of the sample table (1). The support column (23) is located between the four universal rollers (22); two mutually parallel and vertically arranged fixed limit steel plates (25) are fixedly connected to the top of the bottom plate; the two fixed limit steel plates (25) are located between the four supporting hydraulic cylinders (26); two mutually parallel and vertically arranged hydraulic cylinder steel plates (24) are fixedly connected between the two fixed limit steel plates (25); the top ends of the hydraulic cylinder steel plates (24) and the fixed limit steel plates (25) are fixedly connected to the bottom end of the top plate; the top end of the top plate abuts against the bottom end of the top limit system (3).

3. The composite rock formation structure detection while drilling device according to claim 2, characterized in that: The adaptive drilling data acquisition module (5) comprises a fixed plate, the bottom end of the fixed plate is fixedly connected to the piston end of the drilling hydraulic cylinder (4), one side of the fixed plate is fixedly connected to a drilling guide groove (55), the drilling guide groove (55) is slidably connected to the support rod, the bottom of the other side of the fixed plate is fixedly connected to a speed acquisition unit (54), the output end of the speed acquisition unit (54) is fixedly connected to one end of a torque acquisition unit (53), the other end of the torque acquisition unit (53) is fixedly connected to a drill rod groove (51), an axial force acquisition unit (52) is arranged between the drill rod groove (51) and the torque acquisition unit (53), and the axial force acquisition unit (52), the torque acquisition unit (53), and the speed acquisition unit (54) are electrically connected to the data acquisition device (9).

4. The composite rock formation structure detection while drilling device according to claim 3, characterized in that: The sample table (1) comprises a sample table plane (14), the sample table plane (14) is fixedly connected to the support rod, a drilling hole (12) is provided in the middle of the sample table plane (14), the universal roller (22) is in rolling contact with the sample table plane (14), the bottom end of the rock sample clamping device support column (23) is in contact with the top of the sample table plane (14), rock sample entrances (11) are respectively provided on both sides of the top of the sample table plane (14), and a sample table guardrail (13) is fixedly connected to one side wall of the sample table plane (14).

5. The composite rock formation structure detection while drilling device according to claim 3, characterized in that: The bottom side wall of the support rod is fixedly connected with a guide rail (7), and the drilling guide rail groove (55) is slidably matched with the guide rail (7).

6. An intelligent identification method for a composite rock formation structure detection while drilling device according to any one of claims 1 to 5, characterized in that: The following steps are involved: S1. preparing rock samples (21) of different strengths; S2, testing basic parameters of the rock sample (21) in S1; S3, placing the rock sample (21) in S2 in the rock holder (2) and fixing it Certainly; S4, controlling the adaptive while-drilling data acquisition module (5) to drill the rock sample (21), and monitoring parameter changes during the drilling process; S5, when drilling is completed, the adaptive while-drilling data acquisition module (5) is removed from the rock sample (21); S6, repeating S2-S5, and conducting experiments on the rock samples (21) of different strengths and composites; S7, repeating S2-S5, setting different confining pressures, and conducting experiments on the rock samples (21) of different strengths; S8, repeating S2-S5, and conducting experiments on the rock samples (21) with the same strength and different damage degrees; S9. Establish and analyze a database based on experimental data, use machine learning to train and learn with the database as the target, and realize intelligent identification of roof structure.

7. The intelligent identification method of the composite rock structure while drilling detection device according to claim 6 is characterized in that: In the S1, the rock samples (21) of different strengths are prepared according to similarity criteria or based on in-situ rocks.

8. The intelligent identification method of the composite rock formation structure while drilling detection device according to claim 6 is characterized in that: The basic parameters in S2 include uniaxial compressive strength, elastic modulus, internal friction angle, and Poisson's ratio.

9. The intelligent identification method of the composite rock structure while drilling detection device according to claim 6, characterized in that: The axial force acquisition unit (52), the rotation speed acquisition unit (54) and the torque acquisition unit (53) in the adaptive drilling data acquisition module (5) are positively correlated with the rock strength. The preliminary mapping relationship between the rock strength and the drilling parameters is shown in formula (1). The mapping relationship is integrated into the machine learning algorithm to achieve intelligent identification of rock types. In formula (1), E represents rock strength, i represents the number of data, k is the empirical coefficient, N is the rotation speed, M is the torque, F is the drilling force, a is the empirical coefficient of the drilling force, and b is a constant.

Images