Boulder stratum drilling pile construction and drilling parameter intelligent control method

Abstract

The application provides a cobble stratum drilling pile construction and drilling parameter intelligent control method, relates to the technical field of artificial intelligence and drilling pile construction, and comprises the following steps: setting an acoustic wave emission array at the end of a drill bit to collect cobble distribution information, identifying a weak connection belt as a passable area, calculating an optimal obstacle avoidance track and setting a posture control point, adjusting drill bit displacement, mud jetting and drill rod deformation parameters, and realizing intelligent threading of a drilling tool along the obstacle avoidance track. The application can effectively avoid large cobble stones, reduce drilling resistance, improve drilling efficiency and reduce equipment wear.

Description

TECHNICAL FIELD

[0001] The present application relates to artificial intelligence and drilling pile construction technology, and particularly relates to a cobble stratum drilling pile construction and drilling parameter intelligent control method. BACKGROUND

[0002] In the construction of a drilling pile in a cobble stratum, a traditional manual experience control method lacks real-time sensing ability for a drilling environment, cannot accurately obtain stratum structure characteristics and obstacle distribution information in front of a drill bit, and leads to great blindness in drilling trajectory planning and parameter optimization.

[0003] A current intelligent control system mainly relies on single displacement or pressure sensing data for drilling parameter adjustment, cannot establish a mapping relationship between a drilling tool motion state and stratum characteristics, and cannot realize self-adaptive response to a complex stratum environment. Meanwhile, due to lack of fusion analysis and real-time decision-making ability for multi-source information in a drilling process, it is difficult to guarantee continuity of a drilling trajectory and construction efficiency.

[0004] Therefore, an urgent need exists for a cobble stratum drilling pile construction and drilling parameter intelligent control method, which realizes accurate sensing of a drilling environment and self-adaptive regulation of drilling parameters through artificial intelligence algorithms such as sound wave detection, feature extraction, and trajectory planning, and improves intelligent level of drilling pile construction. SUMMARY

[0005] Embodiments of the present application provide a cobble stratum drilling pile construction and drilling parameter intelligent control method, which can solve the problems in the prior art.

[0006] In a first aspect, the present application provides a cobble stratum drilling pile construction and drilling parameter intelligent control method, which comprises:

[0007] An annular sound wave emission array is arranged at an end portion of the drill bit, reflection signals of the detection sound waves among the cobble stones are collected, the reflection signals are subjected to frequency spectrum analysis and waveform superposition processing, attenuation characteristics and reflection characteristics of the sound waves in different media are extracted, and a cobble distribution density graph in a three-dimensional space range in front of the drill bit is drawn;

[0008] According to the cobble distribution density graph, a weak connection zone between adjacent cobble stones is identified as a passable area, distances and direction angles from a gravity center of the drilling tool to centers of each weak connection zone are calculated in the passable area, a weak connection zone sequence with the shortest distance and the smallest direction angle change is selected as an obstacle avoidance trajectory, and an attitude control point is arranged on the obstacle avoidance trajectory;

[0009] For each attitude control point, extension displacement of each quadrant portion of the drill bit and bending deformation of the drill rod are calculated, and liquid outlet directions and jetting pressures of each quadrant mud nozzle are adjusted to form a directional softening channel, so that the drilling tool sequentially passes through each control point;

[0010] The spatial position and attitude angle data of the drilling tool are collected in real time, the weak connection belt strength and strike characteristics of the current drilling area are extracted, the position deviation of the drilling tool and each attitude control point is calculated, the drilling tool stretch displacement, the drilling rod bending deformation and the mud jet pressure parameters are adjusted in real time according to the position deviation and the distribution law of the weak connection belt, so that the drilling tool keeps moving along the obstacle avoidance trajectory.

[0011] In an alternative embodiment,

[0012] The reflection signals of the detection sound waves between the cobblestones are collected, the reflection signals are subjected to frequency spectrum analysis and waveform superposition processing, the attenuation characteristics and reflection characteristics of the sound waves in different media are extracted, and the cobblestone distribution density graph in the three-dimensional space range in front of the drilling head is drawn, including:

[0013] The reflection signals are subjected to wavelet decomposition to obtain energy components of different frequency bands, the amplitude attenuation rate and phase delay value of each frequency band signal are extracted, and the acoustic impedance variation amount of the medium on the propagation path is calculated according to the difference in the attenuation degree of the soil and the cobblestone to different frequency sound waves;

[0014] Based on the sudden change point of the acoustic impedance variation amount, the medium interface spatial coordinates on the sound wave propagation path are determined, the spatial coordinates are taken as the target position, the received reflection signals are subjected to time delay superposition by using the spatial beam synthesis method, the time delay compensation amount of each reflection signal is dynamically adjusted according to the phase delay value, and the echo signal at the target position is in phase and enhanced;

[0015] The three-dimensional space in front of the drilling head is divided into a plurality of concentric circular grid voxels, for each voxel position, the cobblestone existence probability value corresponding to the voxel position is calculated according to the echo signal intensity, the phase delay value and the acoustic impedance variation amount;

[0016] The cobblestone existence probability value is subjected to continuous processing by using a three-dimensional distance weight interpolation algorithm, and a cobblestone distribution density graph in the three-dimensional space range in front of the drilling head is generated.

[0017] In an alternative embodiment,

[0018] According to the cobblestone distribution density graph, the weak connection belt between adjacent cobblestones is identified as a passable area, including:

[0019] The cobblestone distribution density graph is subjected to multi-scale density threshold segmentation, the optimal segmentation threshold is determined by using an adaptive iteration method, and a binary density image of the cobblestone area is obtained;

[0020] A cobblestone boundary contour is extracted from the binary density image, a morphological closing operation with a structure radius adjustable is performed to eliminate the cavities inside the boundary, a morphological opening operation with edge preservation is performed to smooth the boundary, and a regularized cobblestone region is obtained, an anisotropic distance field is constructed for the regularized cobblestone region, a distance weight is adaptively adjusted based on the density gradient information, and a directional distance gradient map is calculated;

[0021] A continuous region with the maximum gradient value is extracted as an initial division belt in the distance gradient map, density sampling is performed along the center line of the initial division belt to obtain a density sampling sequence, region growing is performed based on the density sampling sequence, a continuous region with a density value lower than a dynamic threshold is expanded and marked as a weak connection belt, and morphological optimization is performed on the weak connection belt to obtain a final passable region.

[0022] In an optional embodiment,

[0023] In the passable region, the distance and direction angle of the drill center of gravity to the center of each weak connection belt are calculated, a weak connection belt sequence with the shortest distance and the smallest direction angle change is selected as an obstacle avoidance trajectory, and a posture control point is set on the obstacle avoidance trajectory, including:

[0024] The weak connection belts in the passable region are classified according to the width to obtain weak connection belts of different width levels, and the center lines of the weak connection belts of each level are extracted to establish a multi-level weak connection belt network containing the center lines and the transverse connection relationship;

[0025] In the multi-level weak connection belt network, a dynamic coordinate system is established with the drill center of gravity as the origin, the distance and direction angle of the drill center of gravity to the center line of each weak connection belt are calculated, and a pass evaluation function is constructed based on the distance, direction angle and transverse connection cost, wherein the transverse connection cost decreases with the increase of the width of the weak connection belt;

[0026] Bidirectional search is performed in the multi-level weak connection belt network using the pass evaluation function: the search space is expanded forward from the current position to the target position, and the search space is expanded backward from the target position to the current position; the optimal expansion direction is determined according to the evaluation function values of the two search directions, and a weak connection belt sequence with the minimum pass evaluation function value is selected as the obstacle avoidance trajectory;

[0027] Posture control points are set on the obstacle avoidance trajectory, the distance between adjacent posture control points is proportional to the width of the corresponding weak connection belt, and transition control points are added by minimizing the local curvature change rate in the area where the width of the weak connection belt changes, to ensure the continuous and smooth transition of the trajectory.

[0028] In an optional embodiment,

[0029] For each attitude control point, the extension displacement of each quadrant of the drill bit and the bending deformation of the drill pipe are calculated, and the liquid outlet direction and jet pressure of each quadrant mud nozzle are adjusted to form a directional softening channel, so that the drilling tool sequentially passes through each control point, including:

[0030] A local coordinate system is established at the attitude control point, the drill bit is divided into four quadrants, the extension displacement of each quadrant of the drill bit is calculated, the bending deformation of the drill pipe is calculated based on the elastic deformation theory of the drill pipe, and the ratio of the extension displacement to the bending deformation is calculated to obtain the extension correction coefficient of each quadrant.

[0031] The rock hardness difference between each quadrant of the drill bit and the well wall is collected, the rock breaking power ratio of each quadrant is calculated, and the product of the rock breaking power ratio and the extension correction coefficient is calculated to obtain the initial softening weight of each quadrant.

[0032] According to the initial softening weight, the contact pressure between each quadrant of the drill bit and the well wall, and the bending deformation of the drill pipe, the initial adjustment of the mud nozzle of each quadrant is performed to obtain the initial jet state parameters of each quadrant.

[0033] The spatial deviation of the drilling tool position and the attitude control point is collected in real time, the initial softening weight is corrected to obtain the real-time softening weight based on the initial jet state parameters, the jet pressure and liquid outlet direction of each quadrant mud nozzle are dynamically adjusted based on the real-time softening weight, and a directional softening channel is formed to make the drilling tool sequentially pass through each attitude control point.

[0034] In an alternative embodiment,

[0035] The spatial position and attitude angle data of the drilling tool are collected in real time, the weak connection belt strength and strike characteristics of the current drilling area are extracted, and the position deviation of the drilling tool and each attitude control point is calculated, including:

[0036] A motion sensor group is arranged on the drilling tool body in the axial direction to collect original motion data; a strain sensor array is arranged around the installation point of the motion sensor group to establish a local deformation vector field of the drilling tool; the radial displacement and axial displacement of the sensor installation point are calculated by using the deformation vector field, and the radial displacement and axial displacement are eliminated from the original motion data to obtain the compensated real-time spatial position and attitude angle of the drilling tool.

[0037] A plurality of groups of acoustic wave transducers are arranged in the circumferential direction at the end of the drill bit, each group of transducers including a transmitting unit and a receiving unit; adjacent transducer groups are controlled to alternately emit detection signals and collect reflection signals; the reflection signals are segmented and phase superimposed to enhance the boundary characteristics, and a corresponding relationship between the circumferential scanning angle of the drilling tool and the reflection signal amplitude is constructed; the connection belt boundary and the density change are identified based on the corresponding relationship, and the spatial distribution range and strike characteristics of the weak connection belt are determined in combination with the compensated real-time spatial position of the drilling tool.

[0038] The corrected real-time spatial position and attitude angle of the drilling tool are projected to the coordinate systems of the attitude control points respectively through spatial coordinate transformation, force influence coefficients at the control points are established according to the strength and strike characteristics of the weak connection belt, the projected coordinates are corrected, vector differences between the corrected coordinates and the corresponding attitude control points are calculated, and position deviations between the drilling tool and the attitude control points are obtained.

[0039] In an alternative embodiment,

[0040] According to the position deviation and the distribution law of the weak connection belt, the drilling bit extension displacement, the drilling rod bending deformation and the mud jet pressure parameters are adjusted in real time, so that the drilling tool keeps moving along the obstacle avoidance trajectory, which comprises:

[0041] According to the size of the position deviation and the distribution law of the weak connection belt, the drilling process is divided into a stable drilling section with a deviation less than a preset deviation threshold and an obstacle avoidance adjustment section with a deviation greater than the preset deviation threshold.

[0042] When the drilling tool is located in the stable drilling section, the extension displacement of each quadrant of the drilling bit is calculated based on the position deviation, the bending deformation of the adjacent drilling rod sections is determined according to the distribution law of the weak connection belt, and the adjustment instruction is output according to the set matching relationship between the extension displacement and the bending deformation.

[0043] When the drilling tool is located in the fast obstacle avoidance section, the optimal obstacle avoidance direction is calculated according to the distribution law of the weak connection belt, the circumferential arrangement position of the mud nozzle is determined based on the optimal obstacle avoidance direction, the mud jet pressure parameters of each nozzle are adjusted according to the circumferential arrangement position, the obstacle avoidance propulsion force is formed, and the adjustment instruction is output according to the obstacle avoidance propulsion force.

[0044] The movement trajectory of the drilling tool after executing the adjustment instruction is collected, the new position deviation is calculated, the drilling force data is collected, and the movement stability is evaluated, and the new position deviation and the movement stability are taken as evaluation parameters.

[0045] According to the evaluation parameters, when the new position deviation is reduced and the movement stability meets the requirements, the current adjustment parameters are kept; otherwise, the new position deviation and the movement stability are returned as inputs, the drilling bit extension displacement, the drilling rod bending deformation and the mud jet pressure parameters are recalculated until the drilling tool keeps moving along the obstacle avoidance trajectory.

[0046] In a second aspect of the embodiments of the present application, an electronic device is provided, comprising:

[0047] a processor;

[0048] a memory for storing processor-executable instructions;

[0049] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0050] A third aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0051] In this embodiment, by setting a ring-shaped acoustic emission array at the drill bit tip, the distribution of cobblestone strata is accurately detected and visualized, significantly improving the ability to identify the geological structure during drilling and effectively reducing the risk of drill bit damage and stuck drill bit caused by encountering large cobblestones. Based on the cobblestone distribution density map, weak connection zones are intelligently identified and optimal obstacle avoidance trajectories are planned. By adjusting the extension and retraction displacement of each quadrant of the drill bit and the bending deformation of the drill rod, precise navigation of the drilling tool in complex strata is achieved, significantly improving drilling efficiency, reducing construction time, and lowering project costs. An intelligent control method that dynamically adjusts drilling parameters in real time automatically optimizes the mud injection direction and pressure according to the geological characteristics of different areas, forming a directional softening channel. This allows the drilling tool to advance smoothly along a predetermined trajectory, improving construction quality and safety, and is particularly suitable for bored pile construction in complex strata containing a large number of irregular cobblestones. Attached Figure Description

[0052] Figure 1 This is a flowchart illustrating the intelligent control method for drilling parameters in cobblestone strata borehole pile construction according to an embodiment of the present invention.

[0053] Figure 2 The above is a thermal image of the optimized pebble boundary before and after morphological treatment in an embodiment of the present invention.

[0054] Figure 3 This is a flowchart illustrating the trajectory planning framework of an embodiment of the present invention;

[0055] Figure 4 This is a thermogram showing the adjustment of mud nozzle injection pressure and liquid outlet direction in an embodiment of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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 some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0057] The technical solutions of the present application are described in detail below with specific examples. The following specific examples can be combined with each other, and the same or similar concepts or processes may not be described in detail in some examples.

[0058] Figure 1 The flowchart of the intelligent control method for the construction and drilling parameters of the cobble stratum drilling pile is shown in FIG. Figure 1 The method comprises the following steps:

[0059] An annular acoustic wave emission array is arranged at the end of the drill bit, the reflection signals of the detection acoustic waves between the cobbles are collected, the reflection signals are subjected to frequency spectrum analysis and waveform superposition processing, the attenuation characteristics and reflection characteristics of the acoustic waves in different media are extracted, and the cobble distribution density graph in the three-dimensional space range in front of the drill bit is drawn;

[0060] According to the cobble distribution density graph, the weak connection belts between the adjacent cobbles are identified as passable areas, the distances and direction angles from the gravity center of the drilling tool to the centers of the weak connection belts in the passable areas are calculated, the weak connection belt sequence with the shortest distance and the smallest direction angle change is selected as the obstacle avoidance trajectory, and the attitude control points are arranged on the obstacle avoidance trajectory;

[0061] For each attitude control point, the extension displacement of each quadrant part of the drill bit and the bending deformation amount of the drill pipe are calculated, the liquid outlet direction and the jet pressure of each quadrant mud nozzle are adjusted, the directional softening channel is formed, and the drilling tool sequentially passes through each control point;

[0062] The spatial position and attitude angle data of the drilling tool are collected in real time, the strength and strike characteristics of the weak connection belts in the current drilling area are extracted, the position deviation of the drilling tool from each attitude control point is calculated, the position deviation and the distribution law of the weak connection belts are adjusted in real time, the extension displacement of the drill bit, the bending deformation amount of the drill pipe and the mud jet pressure parameters are adjusted in real time, and the drilling tool is kept moving along the obstacle avoidance trajectory.

[0063] In an optional implementation, collecting the reflection signals of the detection acoustic waves between the cobbles, subjecting the reflection signals to frequency spectrum analysis and waveform superposition processing, extracting the attenuation characteristics and reflection characteristics of the acoustic waves in different media, and drawing the cobble distribution density graph in the three-dimensional space range in front of the drill bit comprise the following steps:

[0064] The reflection signals are subjected to wavelet decomposition to obtain energy components of different frequency bands, the amplitude attenuation rate and the phase delay value of each frequency band signal are extracted, the acoustic impedance variation amount of the medium on the propagation path is calculated according to the attenuation degree difference of the soil and the cobbles to different frequency acoustic waves;

[0065] A medium interface spatial coordinate on the sound wave propagation path is determined based on a sudden change point of the sound impedance variation amount, the spatial coordinate is taken as a target position, a spatial beam synthesis method is used to delay and superimpose the received reflection signals, a time delay compensation amount of each reflection signal is dynamically adjusted according to the phase delay value, and echo signals at the target position are in phase and enhanced;

[0066] A three-dimensional space in front of the drill bit is divided into a plurality of layers of concentric circular grid voxels, and for each voxel position, a cobblestone existence probability value corresponding to the voxel position is calculated according to echo signal strength, a phase delay value and a sound impedance variation amount;

[0067] A three-dimensional distance weight interpolation algorithm is used to continuously process the cobblestone existence probability value, and a cobblestone distribution density map in a three-dimensional space range in front of the drill bit is generated.

[0068] The embodiment provides a system implementation method for detecting cobblestone distribution by sound waves, which realizes accurate drawing of cobblestone distribution density in a three-dimensional space range in front of the drill bit through sound wave acquisition, signal processing and space modeling technology.

[0069] In actual operation, the sound wave detection system is configured at the front end of the drill bit and includes one transmitting transducer and at least four receiving transducers. The transmitting transducer emits a detection sound wave signal with a frequency range of 1 kHz to 10 kHz and a duration of 10 ms. The system sampling frequency is set to 100 kHz, 500 ms of reflection signals are collected each time for detection, and the data accuracy is 16 bits.

[0070] After the signal acquisition is completed, wavelet decomposition processing is performed to obtain energy components of different frequency bands. In a specific implementation, six-layer wavelet decomposition is used to decompose the original signal into six frequency bands: 0-0.78 kHz, 0.78-1.56 kHz, 1.56-3.13 kHz, 3.13-6.25 kHz, 6.25-12.5 kHz and 12.5-25 kHz. Practice shows that for cobblestones with a diameter of 5-30 cm, the 3.13-6.25 kHz frequency band signal has the best recognition effect. The energy envelope of each frequency band signal is calculated, and the amplitude decay rate is obtained by analyzing the slope change of the energy curve. For example, in the cobblestone-containing area, the amplitude decay rate of the 3.13-6.25 kHz frequency band signal is usually 0.6-0.8 dB / m, while in the ordinary soil body, it is 1.2-1.5 dB / m.

[0071] The extraction of phase delay values is achieved by Hilbert transform. After Hilbert transform of the signal, the instantaneous phase of the signal is calculated. By comparing the phase changes of signals in different frequency bands during propagation, the phase delay curve is established. For example, in the actual measurement case, when the sound wave encounters cobblestones with a diameter of about 15 cm, the phase delay values of the 3.13-6.25 kHz frequency band signals increase by 27° to 35° compared with the predicted values of the standard soil model.

[0072] Based on the obtained amplitude attenuation rate and phase delay value, the change amount of acoustic impedance of the medium on the sound wave propagation path is calculated. In practical application, the typical value of the acoustic impedance of soil is about 1.5×10 2 s), and the typical value of the acoustic impedance of cobblestone is about 7.8×10 6 kg / (m 2 s). In the calculation, the sliding window method is used, the window width is 10 sampling points, and the step is 5 sampling points. When the calculated acoustic impedance change amount exceeds 1.2×10 6 kg / (m 2 s) in three consecutive windows, it is determined that there is a medium interface.

[0073] After the acoustic impedance change mutation point is determined, the spatial coordinates of these interfaces are determined by using a three-dimensional coordinate calculation model. In practical application, according to the time difference from emission to reception of the sound wave and the propagation speed of the sound wave in the soil (about 1500 m / s), combined with the spatial layout information of the receiving transducer (four receiving transducers are arranged in a square shape with a side length of 10 cm), the spatial coordinates of the reflection points are calculated by the principle of triangulation. For example, in a certain test case, it is detected that there is a cobblestone with a diameter of about 20 cm at a distance of 78 cm in front of the drill bit. After the target position is determined, the spatial beam synthesis processing is performed. A grid with a spatial resolution of 5 cm×5 cm×5 cm is selected. For each grid point, the time delay compensation amount is calculated according to the distance difference from each receiving transducer. In the processing, the signals obtained by each receiving transducer are delayed and superimposed, and the time delay compensation amount is dynamically adjusted according to the previously calculated phase delay value. For example, when the phase delay value of a certain grid point is 30°, the corresponding time delay compensation adjustment amount is 0.026 ms. This processing enhances the in-phase echo signal of the target position and significantly improves the signal-to-noise ratio, which can be improved by about 6-9 dB in actual measurement.

[0074] To draw the cobblestone distribution density map of the three-dimensional space in front of the drill bit, the space is divided into a plurality of concentric circular grid voxels, with a radius range of 0.5-3 meters, an angular resolution of 5°, and a radial resolution of 10 cm. For each voxel position, a cobblestone existence probability value is calculated, and the calculation is based on the echo signal intensity (normalized to 0-1), the phase delay value (weighting coefficient 0.3), and the acoustic impedance change amount (weighting coefficient 0.7). Tests show that when the comprehensive probability value exceeds 0.75, the cobblestone is extremely likely to exist at this position.

[0075] A three-dimensional distance-weighted interpolation algorithm is used to continuously process the cobblestone existence probability value to generate a visual density distribution map. In specific implementation, an interpolation radius of 15 cm and a distance weight factor of 2 are selected to smooth the probability value. The generated density map uses a red-yellow-green-blue four-color gradient to represent the probability level, where the red area (probability value > 0.85) represents the highest cobblestone density area, and the blue area (probability value < 0.3) represents the cobblestone-free area.

[0076] Based on the above technical solutions, high-precision detection and visualization of cobblestone distribution in complex geological structures in front of the drill bit can be achieved, significantly improving the resolution and reliability of geological exploration. By extracting the attenuation characteristics and phase information of sound waves in different frequency bands through wavelet decomposition, different media such as soil and cobblestones can be accurately distinguished, and their spatial distribution interfaces can be identified. By introducing spatial beam synthesis and dynamic time delay compensation technology, the signal-to-noise ratio of the echo signal at the target position can be enhanced, interference and multipath effects can be suppressed, and image clarity and accuracy can be improved. By using voxel probability modeling and three-dimensional interpolation algorithms, the distribution density of cobblestones in the area in front of the drill bit can be continuously expressed, providing reliable basis for drilling path optimization, risk assessment, and automatic control.

[0077] In an alternative embodiment, identifying the weak connection zone between adjacent cobblestones as a passable area based on the cobblestone distribution density map includes:

[0078] Performing multi-scale density threshold segmentation on the cobblestone distribution density map, using an adaptive iterative method to determine the optimal segmentation threshold, to obtain a binary density image of the cobblestone region;

[0079] Extracting the cobblestone boundary contour from the binary density image, using a morphological operator with adjustable structure radius to perform a closing operation to eliminate internal cavities, and using a morphological opening operation with edge preservation to smooth the boundary, to obtain a regularized cobblestone region, constructing an anisotropic distance field for the regularized cobblestone region, adaptively adjusting the distance weight based on the density gradient information, and calculating a directional distance gradient map;

[0080] Extract a continuous region with the maximum gradient value in the distance gradient map as an initial boundary zone, perform density sampling along the center line of the initial boundary zone, obtain a density sampling sequence, perform region growing based on the density sampling sequence, expand a continuous region with a density value lower than a dynamic threshold value to a weak connection zone, and perform morphological optimization on the weak connection zone to obtain a final passable region.

[0081] The embodiment provides a method for identifying a weak connection zone between adjacent cobblestones as a passable region according to a cobblestone distribution density map. First, a cobblestone distribution density map is obtained, and a gray-scale image is formed through preprocessing, wherein a gray-scale value represents a cobblestone distribution density, and a higher gray-scale value indicates a greater density.

[0082] The cobblestone distribution density map is subjected to multi-scale density threshold segmentation. Specifically, an adaptive iteration method is used to determine an optimal segmentation threshold. The method first selects an initial threshold T0 (for example, taking the average gray value 128 of the density map), divides the image into a foreground region R1 and a background region R2, calculates the average gray values μ1 and μ2 of R1 and R2 respectively, then uses (μ1+μ2) / 2 as a new threshold T1, and repeats the iteration until the threshold change value is less than a preset threshold (such as 0.5). In actual application, the initial threshold can be set to 128 (8-bit image), and after 6 iterations, the final threshold T=143 is obtained. The region with a density value greater than 143 is marked as a cobblestone region, and the region with a density value less than or equal to 143 is marked as a non-cobblestone region, forming a binary density image. Next, the cobblestone boundary contour is extracted from the binary density image. An 8-neighbor boundary tracking algorithm is used to detect boundary points to generate a boundary contour sequence. For a typical cobblestone region (with an area of about 500 square pixels), the number of boundary contour points is about 80-120.

[0083] The extracted boundary contour is optimized. First, a morphological closing operation with a structure radius that can be adjusted is used to eliminate internal cavities of the boundary. The structure element is selected to be a disc, and the radius is adaptively adjusted according to the average size of the cobblestone. In practice, for a cobblestone with an average diameter of 30 pixels, a structure element radius of 5 pixels is appropriate. After closing operation, a morphological opening operation with edge preservation is used to smooth the boundary, and the same shape but a structure element with a radius of 3 pixels is used to retain the main morphological features of the cobblestone while removing small protrusions. After these two steps of operation, a regularized cobblestone region is obtained.

[0084] An anisotropic distance field is constructed based on the regularized cobblestone region. The traditional Euclidean distance field cannot accurately reflect the actual passability in the case of uneven density distribution, so the density information is introduced to modify the distance calculation. First, the Euclidean distance d(x, y) from each non-cobblestone region pixel point to the nearest cobblestone boundary is calculated, and the density value p(x, y) of the point is obtained. Then, the distance weight is adjusted adaptively according to the density gradient information, and the distance value D(x, y) is modified as D(x, y) = d(x, y) x (1 + a x p(x, y) / pmax), where a is the weight coefficient (0.8 in the experiment), and pmax is the maximum value of the density map (255). This makes the distance value in the high-density area be increased, indicating that the passability is increased. After calculating the modified distance of all pixel points, the gradient of the distance field is calculated to obtain the directional distance gradient map.

[0085] In the distance gradient map, the continuous region with the maximum gradient value is extracted as the initial division band. Set the gradient threshold Tg (75% of the maximum gradient value in the experiment, about 0.3), and mark the region with gradient value greater than Tg as the initial division band. The initial division band usually has a band-like distribution, with a width of about 8-15 pixels. Density sampling is performed along the center line of the initial division band to obtain a density sampling sequence. The center line of the initial division band is extracted using a thinning algorithm, and the density values are sampled at an interval of 5 pixels along the center line to form a density sampling sequence {p1, p2,..., pn}. In a typical scene, a division band with a length of about 200 pixels will generate about 40 density sampling points. Based on the density sampling sequence, region growing is performed to label the weak connection band. The mean value p and the standard deviation s of the density sampling sequence are calculated, and the dynamic threshold Td = p - b x s is set, where b is the adjustment coefficient (1.5 in the experiment). Starting from the sampling point with the lowest density value, expand to both sides, and mark the region with continuous density value lower than Td as the weak connection band candidate area. In the example, p is about 110, s is about 20, and Td is about 80.

[0086] The weak connection band candidate area is morphologically optimized. First, conditional dilation is performed, and the structure element is a 3x3 square. The density value of the expanded region is limited to not more than 1.2 x Td to prevent excessive expansion to the high-density region. Then, morphological smoothing is performed using a 5x5 Gaussian kernel to filter out jagged edges. Finally, path width optimization is performed to ensure that the connection band width is not less than the preset minimum passable width (such as 15 pixels), and the narrow part is appropriately expanded. After this series of optimization, the final passable region is obtained.

[0087] In the prior art, the recognition of the cobblestone area in front of the drill bit mainly relies on fixed threshold segmentation and simple morphological processing, which is difficult to cope with complex geological environments with dramatic density changes and blurred boundaries, resulting in low recognition accuracy and poor robustness of the passable path. To solve the above problems, the present application can dynamically determine the optimal segmentation threshold according to the actual density distribution characteristics by using multi-scale density threshold segmentation and adaptive iteration method, effectively adapting to the density differences in different regions; the morphological closing operation with adjustable structure radius and the opening operation with edge preservation are introduced to enhance the flexibility and smoothness of boundary processing, and to improve the problems of boundary discontinuity and misrecognition; by constructing an anisotropic distance field and adjusting the weight based on the density gradient information, a directional distance gradient map is obtained, and directional recognition of the internal structure of the cobblestone dense area is realized; on this basis, the gradient extremum continuous region is extracted and density sampling and region growing are performed, and the weak connection belt with loose structure and low density is further recognized, and the passable area is effectively marked. The overall improvement takes improving the recognition accuracy and adaptability as the starting point, and finally realizes the accurate extraction and dynamic optimization of the cobblestone area passable path in complex dense environment, and enhances the intelligent level of the drilling navigation and obstacle avoidance system.

[0088] Figure 2 For the cobblestone boundary optimization heat map before and after the morphological processing of the embodiment of the present application, as shown in Figure 2 , the figure shows the cobblestone boundary contour optimization effect before and after the morphological processing. The left side of the figure is the cobblestone boundary before processing, and it can be clearly seen that there are three obvious internal cavities, the boundary is irregular in jagged shape, and the density value is unevenly distributed. The right side is the cobblestone boundary after the morphological optimization of the present technical solution, the internal cavities of the boundary are eliminated by using the morphological closing operation with adjustable structure radius, and the boundary is smoothed by the morphological opening operation with edge preservation. From the comparison of the density values, the region density value before processing is D=0.45 to D=0.72, which is unevenly distributed; the density value after processing is improved and more uniform, ranging from D=0.48 to D=0.75. Quantitative analysis shows that the average boundary roughness before processing is 0.65, and the average boundary roughness after processing is reduced to 0.23, and the boundary smoothness is significantly improved. Especially the low-density region with D=0.51 is improved to D=0.54 after processing, which improves the density uniformity. The optimization process maintains the original boundary shape characteristics while improving the regularity of the region, laying a foundation for the subsequent distance field construction and boundary extraction.

[0089] In an optional embodiment, in the passable area, the distance and direction angle of the drill center of gravity to the center of each weak connection belt are calculated, the weak connection belt sequence with the shortest distance and the smallest direction angle change is selected as the obstacle avoidance trajectory, and the attitude control point is set on the obstacle avoidance trajectory, including:

[0090] The weak connection belts in the passable area are classified according to the width, and weak connection belts of different width levels are obtained; the center lines of the weak connection belts of each level are extracted, and a multi-level weak connection belt network containing the center lines and the transverse connection relationship is established;

[0091] In the multi-level weak connection belt network, a dynamic coordinate system is established with the center of gravity of the drilling tool as the origin, the distance and the direction angle of the center of gravity of the drilling tool to the center line of each weak connection belt are calculated; and a passable evaluation function is constructed based on the distance, the direction angle and the transverse connection cost, wherein the transverse connection cost decreases with the increase of the width of the weak connection belt;

[0092] The passable evaluation function is used to perform bidirectional search in the multi-level weak connection belt network: the search space is expanded forward from the current position to the target position, and the search space is expanded backward from the target position to the current position; the optimal expansion direction is determined according to the evaluation function values of the two search directions, and the weak connection belt sequence with the minimum passable evaluation function value is selected as the obstacle avoidance trajectory;

[0093] The posture control points are set on the obstacle avoidance trajectory, the distance between adjacent posture control points is proportional to the width of the corresponding weak connection belt; in the area where the width of the weak connection belt changes, transition control points are added by minimizing the local curvature change rate to ensure the continuous and smooth transition of the trajectory.

[0094] Exemplarily, after the classification, the center line of each level of weak connection belt area is extracted. The center line represents the optimal passable path of the connection belt in space, which can be realized by skeleton extraction algorithm, morphological distance transformation or center axis search method. The extraction of the center line needs to ensure that it does not deviate from the path boundary inside the connection belt, and at the same time, the continuity and representativeness of the path direction are preserved as much as possible. After the center line extraction is completed, a multi-level weak connection belt network containing the center line information and the transverse connection relationship is established in combination with the intersection and connection relationship between the connection belts of different levels. The network not only contains the level and position attributes of each weak connection belt, but also contains the type and cost attributes of the connection relationship. The transverse connection relationship is used to describe the conversion channel between connection belts in different directions, and the connection cost decreases with the increase of the width level of the weak connection belt, that is, the transverse connection between wider connection belts is smoother, and the connection cost between narrower connection belts is higher.

[0095] After the multi-level network is constructed, a dynamic coordinate system is established with the current center of gravity of the drilling tool as the reference origin, and the center lines of the weak connection belts are mapped in space. The spatial distance between the center of gravity of the drilling tool and each center line, and the included angle relationship between the current advancing direction of the drilling tool and the extension direction of the center line are calculated. Among them, the distance is used to measure the cost of the drilling tool approaching a specific connection belt, and the direction angle is used to evaluate the operation strength required for attitude adjustment. When performing the above calculation, the actual operation constraints such as the size of the drilling tool, the turning radius and the forward inertia should be considered.

[0096] A passing evaluation function is constructed based on distance, direction angle and lateral connection cost. The function is used to measure the comprehensive cost of the drill string passing through a certain connection zone in the current dynamic coordinate system. To ensure the rationality of the function construction, the following constraints should be set: the shorter the passing distance, the smaller the deviation of the direction angle, and the wider the connection zone, the smaller the evaluation value, i.e. the better. In the construction process, specific formulas are avoided, and a logical weight model is constructed based on the above three factors. The passing path is evaluated step by step in the order of distance and direction angle. For example, for a passing path that starts from the current drill position, passes through two narrow connection zones and then enters a wide connection zone, although the overall distance is not long, the width of the first two connection zones is narrow, the posture needs to be adjusted frequently, and there is a turning blind area, so the evaluation value will be significantly higher than that of a path that is slightly longer but consists of wide connection zones.

[0097] To improve the efficiency and accuracy of obstacle avoidance path planning, bidirectional search is performed in the multi-level weak connection zone network based on the above passing evaluation function. Bidirectional search refers to expanding the search space from the current position of the drill to the target area while performing reverse search from the target area to the current position. The expansion process is progressive, i.e. the direction with a smaller passing evaluation function value is given priority. During the expansion process, a priority queue or buffer structure is set to record the cumulative path cost and path direction state, dynamically update the passing evaluation value of the known path, and real-time sort all connection zones to be expanded, ensuring that the optimal direction is selected at each expansion step. When the two search directions intersect in space, a feasible path is considered to be found. Further analysis of the passing evaluation value of the intersection point corresponding to the path is performed, and the path with the smallest passing evaluation function value is selected from all possible intersection paths as the optimal obstacle avoidance trajectory.

[0098] After the obstacle avoidance trajectory is determined, the posture control points are set on it. The posture control point is an important reference point for fine adjustment and stable movement of the drill posture, and its position should fully consider the width and connection type of the weak connection zone. In the passing path, the distance between adjacent posture control points is proportional to the width of the connection zone where the path is located, i.e. the control point distance is larger in wide connection zones and should be appropriately reduced in narrow connection zones to enhance the frequency and accuracy of drill posture adjustment. For example, when the drill passes through a connection zone with a lateral width of three meters, the control point distance can be set to two meters, while in a connection zone with a width of only one meter, the control point distance should be adjusted to zero point five meters or even smaller. In addition, in areas where the width of the connection zone changes significantly, such as a sudden transition from a wide connection zone to a narrow connection zone, a transition control point should be set to ensure the continuity and stability of the trajectory. The setting of the transition control point should aim to minimize the rate of change of the curvature of the drill movement, i.e. by adding additional control points in the transition area to adjust the degree of path curvature and avoid sharp turns or shaking.

[0099] For example, in a certain underground drilling scenario, the drilling tool needs to move from entrance A to target B, and needs to pass through a tunnel area with complex structure and compact space. First, a three-dimensional laser scanning device is used to obtain spatial data of the tunnel area, and the spatial segmentation and width calculation of the passing path are performed to identify dozens of weak connection belts, which are divided into three categories according to their widths. Then, the center line extraction algorithm is used to extract the center line of each weak connection belt, and a network structure is constructed based on the connection relationship to form a multi-level network containing information such as position, direction, and connection cost. Then, taking the current position of the drilling tool as the origin, the spatial distance and direction angle of each center line are calculated, and a passing evaluation function is established. After constructing a bidirectional search structure, the system simultaneously expands the path search from point A and point B, and finally establishes an intersection path between ten meters from the entrance and fifteen meters from the target. The path with the minimum passing evaluation value is composed of seven weak connection belts of different levels, with a total length of thirty meters and a minimum width of one meter. Based on the path, attitude control points are set, six control points are set in the narrowest area, and three transition control points are set in the transition area, to ensure that the drilling tool trajectory remains continuous and smooth during the passing process, and no yaw phenomenon occurs.

[0100] In the prior art, the path planning of the drilling tool in the complex geological structure usually relies on static maps or one-way search algorithms, which cannot dynamically adapt to the changes in the spatial structure of the cobblestone area, and lack precise control of the drilling attitude changes, which can easily lead to path deviation or drill bit jamming. The present application constructs a multi-level weak connection belt network in the passable area, classifies the weak connection belts by width and extracts the center line, clearly defines the hierarchical relationship of the path structure, and introduces a dynamic network structure containing lateral connections to provide more comprehensive passing information support for subsequent path selection. By constructing a passing evaluation function containing distance, direction angle and lateral connection cost, and combining the bidirectional search mechanism of forward and reverse, the global optimality and search efficiency of the path planning are improved. To solve the problem of insufficient attitude control at the path turning point, attitude control points are further set on the obstacle avoidance trajectory, and the control point spacing is dynamically adjusted according to the width of the weak connection belt, and the minimum curvature change rate constraint is introduced in the width change area to ensure the continuity and smoothness of the path. The overall improvement takes the adaptability, passing efficiency and attitude control accuracy of the path planning as the starting point, and finally realizes the stable and smooth obstacle avoidance passing ability of the drilling tool in complex geological conditions.

[0101] Figure 3 The trajectory planning framework flowchart of the embodiment of the present application is as follows, Figure 3As shown, first, the passable area is divided into n levels according to the width of the weak connection belt, and each level is denoted as Wi(i=1, 2,..., n); then the center line of each level of the weak connection belt is extracted, and the lateral connection relationship is established to form a network topology; then a dynamic coordinate system is established with the center of gravity of the drilling tool as the origin, and the distance and direction angle are calculated; and a pass evaluation function is constructed based on these parameters. In the search stage, a bidirectional search strategy is adopted, that is, the forward search (expansion times 83, expansion range 28.4m) from the starting point to the ending point is performed at the same time as the reverse search (expansion times 76, expansion range 25.7m) from the ending point to the starting point, to ensure the search efficiency. Based on the bidirectional search results, the obstacle avoidance trajectory with the minimum evaluation function value is determined, and the pass cost of the trajectory is 0.37 and the trajectory length is 52.7m. Finally, the attitude control points are set on the obstacle avoidance trajectory, the distance between the control points is proportional to the width of the weak connection belt, and the transition control points are added in the weak connection belt width change area through the minimum curvature change rate criterion to ensure smooth transition of the trajectory. The whole process fully embodies the core features of the technical solution: multi-level weak connection belt network construction, bidirectional search strategy, width adaptive control point setting, and curvature smoothing optimization.

[0102] In an optional embodiment, for each attitude control point, the extension displacement of each quadrant of the drill bit and the bending deformation of the drill pipe are calculated, and the liquid outlet direction and jet pressure of each quadrant mud nozzle are adjusted to form a directional softening channel, so that the drilling tool sequentially passes through each control point, including:

[0103] A local coordinate system is established at the attitude control point, the drill bit is divided into four quadrants, the extension displacement of each quadrant of the drill bit is calculated, the bending deformation of the drill pipe is calculated based on the elastic deformation theory of the drill pipe, and the ratio of the extension displacement to the bending deformation is calculated to obtain the extension correction coefficient of each quadrant;

[0104] The rock hardness difference between each quadrant of the drill bit and the well wall is collected, the rock breaking power ratio of each quadrant is calculated, and the product of the rock breaking power ratio and the extension correction coefficient is calculated to obtain the initial softening weight of each quadrant;

[0105] According to the initial softening weight, the contact pressure between each quadrant of the drill bit and the well wall, and the bending deformation of the drill pipe, the initial adjustment of the mud nozzle of each quadrant is performed to obtain the initial jet state parameters of each quadrant;

[0106] The real-time collection of the spatial deviation of the drilling tool position and the attitude control point is combined with the initial jet state parameters to modify the initial softening weight to obtain the real-time softening weight, and the jet pressure and liquid outlet direction of each quadrant mud nozzle are dynamically adjusted based on the real-time softening weight to form a directional softening channel, so that the drilling tool sequentially passes through each attitude control point.

[0107] For example, a local coordinate system is first established at the attitude control point, dividing the drill bit into four quadrants: up, down, left, and right. A right-handed coordinate system is then established with the drill bit's axis as the origin, where the x-axis points to the right, the y-axis points upwards, and the z-axis moves forward along the drill bit's axis. Pressure and position sensors are used to collect the drill bit's position and attitude data in this coordinate system in real time. For instance, a drill bit center position of (0.02m, -0.03m, 0m) indicates that the drill bit center is offset 2cm to the right and 3cm downwards relative to the control point.

[0108] When calculating the extension and retraction displacement of each quadrant of the drill bit, displacement sensors are used to measure the deviation of each quadrant from the ideal trajectory. Taking the right quadrant as an example, when a deviation of 2.5cm is detected between the right side of the drill bit and the ideal trajectory, the right quadrant needs to contract by approximately 2.5cm to correct the deviation. Similarly, the extension and retraction displacement values ​​are calculated for the upper, lower, and left quadrants respectively, such as an extension of 1.8cm in the upper quadrant, a contraction of 2.2cm in the lower quadrant, and an extension of 1.5cm in the left quadrant. The bending deformation of the drill rod is calculated based on the elastic deformation theory of the drill rod. Strain data of each section of the drill rod are collected using strain gauges, and combined with the material properties of the drill rod (such as steel with an elastic modulus of 210GPa), the bending deformation of the drill rod in each direction is calculated. For example, at a section of the drill rod 10m away from the drill bit, the bending deformation in the x-direction is 3.2cm, and the bending deformation in the y-direction is 2.8cm.

[0109] The ratio of expansion displacement to bending deformation is used to calculate the expansion correction factor for each quadrant. Taking the right quadrant as an example, the expansion displacement is 2.5 cm, and the corresponding bending deformation is 3.2 cm, so the expansion correction factor is 2.5 / 3.2 = 0.78. Similarly, the correction factors for other quadrants are calculated: 0.64 for the upper quadrant, 0.79 for the lower quadrant, and 0.47 for the left quadrant. The rock hardness difference between the drill bit and the wellbore is collected in each quadrant. The rock hardness of the wellbore is measured using a microhardness tester or sonic logging instrument, and the difference between the hardness and the drill bit is calculated. For example, if the rock hardness in contact with the wellbore in the right quadrant is 80 MPa and the drill bit hardness is 900 MPa, the hardness difference is 820 MPa; for the upper quadrant, it is 790 MPa; for the lower quadrant, it is 850 MPa; and for the left quadrant, it is 810 MPa. The rock-breaking power ratio for each quadrant is calculated. Based on the hardness difference and parameters such as drill bit speed and pressure, the rock-breaking power for each quadrant is calculated. For example, the rock-breaking power in the right quadrant is 12kW, in the upper quadrant it is 10.5kW, in the lower quadrant it is 13kW, and in the left quadrant it is 11kW. The total power of the drill bit is 46.5kW. Therefore, the rock-breaking power ratio in the right quadrant is 12 / 46.5 = 0.258, in the upper quadrant it is 0.226, in the lower quadrant it is 0.280, and in the left quadrant it is 0.236.

[0110] The initial softening weight of each quadrant is calculated by multiplying the rock breaking power ratio and the expansion correction coefficient. Taking the right quadrant as an example, the rock breaking power ratio is 0.258, and the expansion correction coefficient is 0.78, so the initial softening weight is 0.258 x 0.78 = 0.201. Similarly, the initial softening weights of other quadrants are calculated, such as 0.145 for the upper quadrant, 0.221 for the lower quadrant, and 0.111 for the left quadrant.

[0111] According to the initial softening weight, the contact pressure between the drill bit quadrants and the well wall, and the bending deformation of the drill pipe, the mud nozzles of each quadrant are initially adjusted. The contact pressure is collected by a pressure sensor, for example, the contact pressure of the right quadrant is 18 MPa, the upper quadrant is 15 MPa, the lower quadrant is 20 MPa, and the left quadrant is 16 MPa. Adjust the liquid direction and jet pressure of the mud nozzle to obtain the initial jet state parameters of each quadrant. Taking the right quadrant as an example, the initial softening weight is 0.201, the contact pressure is 18 MPa, and the mud pump output pressure is 30 MPa, so the initial jet pressure of the right quadrant nozzle is set to 30 x 0.201 = 6.03 MPa, and the nozzle angle is adjusted to 15° with the normal direction of the well wall. Similarly, set the initial jet state parameters of other quadrants.

[0112] Real-time acquisition of spatial deviation of drill position and attitude control point. The actual position and attitude of the drill are measured in real time by gyroscopes, accelerometers and other sensors, compared with the preset control point, and the spatial deviation is calculated. For example, the actual position of the drill is (0.03m, -0.02m, 0.01m), and the position deviation from the control point is 3.74 cm. Modify the initial softening weight based on the initial jet state parameters to obtain the real-time softening weight. The correction formula considers factors such as spatial deviation, actual drilling speed and rock feedback. For example, the initial softening weight of the right quadrant is 0.201, and the spatial deviation correction factor is 1.15, so the real-time softening weight is adjusted to 0.201 x 1.15 = 0.231. Similarly, modify the softening weight of other quadrants.

[0113] Based on the real-time softening weight, dynamically adjust the jet pressure and liquid direction of the mud nozzle of each quadrant. For example, the real-time softening weight of the right quadrant is 0.231, and the mud pump output pressure is 30 MPa, so the jet pressure of the right quadrant nozzle is dynamically adjusted to 30 x 0.231 = 6.93 MPa, and the nozzle angle is adjusted to 18° with the normal direction of the well wall, forming a stronger right directional softening channel.

[0114] The directional softening channel formed by the above steps makes the force of each quadrant of the drill bit balanced, and the drill can sequentially pass through each attitude control point along the preset trajectory, realizing precise directional drilling. The system performs a complete calculation and adjustment cycle every second to ensure real-time control and precise navigation during drilling.

[0115] In the prior art, when the drilling tool passes through a complex path or a turning area, it usually relies on mechanical control or fixed parameter adjustment of the nozzle to soften the rock formation, lacks accurate perception and response to the attitude of the drill bit and the contact state of the well wall, and is prone to cause local resistance increase or drill bit deviation, affecting the drilling efficiency and path accuracy. To solve this problem, the present application introduces a local coordinate system centered on the attitude control point, refines the coordinated calculation of the extension displacement of each quadrant of the drill bit and the bending deformation amount of the drill pipe, obtains an extension correction coefficient that truly reflects the deformation state of the drilling tool, and combines the hardness difference between the drill bit and the well wall to construct an initial softening weight, thereby realizing the differentiated expression of the rock breaking difficulty of different quadrants. On this basis, the initial jetting state of the mud nozzle is adjusted in combination with the contact pressure and deformation degree of each quadrant, thereby improving the accuracy and response speed of softening control. By continuously collecting the attitude deviation of the drilling tool, the initial softening weight is dynamically corrected, and the liquid outlet direction and pressure of the nozzle are adjusted in real time, so that a directional softening channel is formed in front of the drill bit, which is constantly updated according to the path change. The overall improvement takes improving the attitude control accuracy and directional drilling capability as the starting point, and finally realizes smooth passing of the drilling tool through the control point in the complex path, reduces the resistance and deviation risk, and improves the path tracking and passing efficiency.

[0116] Figure 4 The heat map for adjusting the jetting pressure and liquid outlet direction of the mud nozzle of the embodiment of the present application is shown in FIG. 3. Figure 4 As shown in FIG. 3, the heat map directly shows the relationship between the jetting pressure and the liquid outlet direction adjustment of the mud nozzle. The horizontal axis in the figure represents the liquid outlet direction adjustment angle, ranging from -15° to 25°; the vertical axis represents the serial number of 7 different attitude control points. The heat map uses a gray scale gradient to represent the jetting pressure, and the darker the color, the higher the pressure. As can be clearly seen from the figure, at control point 3 and control point 4, when the liquid outlet direction adjustment angle is 0°, the jetting pressure reaches the highest value, which is 19.8 MPa and 19.5 MPa respectively, and these two regions are also marked as the optimal adjustment region. In contrast, the jetting pressure of control point 6 at angles of -15° and 25° is 14.2 MPa and 10.3 MPa respectively, which is significantly lower than the center region. The pressure range of the first quadrant is 16.5-19.8 MPa, and the pressure range of the second quadrant is 10.3-14.5 MPa, indicating that the pressure required by different quadrants is significantly different. From the overall distribution, the best jetting angle mainly concentrates in the 0°~5° interval, which provides an important reference for accurately controlling the drilling direction. The technical solution forms a directional softening channel by fine adjustment of the jetting pressure and liquid outlet direction of the mud nozzle in each quadrant, which is significantly superior to the uniform pressure distribution method used in traditional jetting drilling technology (such as conventional pressure drilling technology). The traditional method usually uses fixed pressure and angle, which is difficult to cope with complex formations and accurate trajectory requirements, while the optimal parameter combination determined by the heat map analysis in the present solution can realize accurate passing of each attitude control point, greatly improving the efficiency and accuracy of complex trajectory drilling.

[0117] In an alternative embodiment, the spatial position and attitude angle data of the drilling tool are collected in real time, the strength and strike of the weak connecting zone in the current drilling area are extracted, and the positional deviation of the drilling tool from each attitude control point is calculated, comprising:

[0118] A motion sensor group is arranged on the drilling tool body in the axial direction to collect raw motion data; a strain sensor array is arranged around the mounting point of the motion sensor group to establish a local deformation vector field of the drilling tool; the radial displacement and axial displacement of the sensor mounting point are calculated by using the deformation vector field decomposition, the radial displacement and axial displacement are eliminated from the raw motion data, and the compensated real-time spatial position and attitude angle of the drilling tool are obtained.

[0119] A plurality of groups of acoustic wave transducers are arranged in the circumferential direction at the end of the drill bit, each group of transducers including a transmitting unit and a receiving unit; adjacent transducer groups are controlled to alternately emit detection signals and collect reflected signals; the reflected signals are subjected to segmented phase superposition to enhance the boundary features, and the corresponding relationship between the circumferential scanning angle of the drilling tool and the amplitude of the reflected signals is constructed; the connecting zone boundary and the density change are identified based on the corresponding relationship, and the spatial distribution range and strike of the weak connecting zone are determined in combination with the compensated real-time spatial position of the drilling tool.

[0120] The corrected real-time spatial position and attitude angle of the drilling tool are projected into the coordinate system of each attitude control point through spatial coordinate transformation, the force influence coefficient of the drilling tool at each control point is established according to the strength and strike of the weak connecting zone, the projection coordinates are corrected, the vector difference between each corrected coordinate and the corresponding attitude control point is calculated, and the positional deviation of the drilling tool from each attitude control point is obtained.

[0121] The embodiment provides a method for real-time acquisition of drilling tool space position and attitude angle data and feature extraction of weak connection zone. A group of motion sensors are arranged on the drilling tool body in the axial direction to collect original motion data. Specifically, a group of three-axis acceleration sensors and three-axis gyroscopes are installed on the drilling tool body every 0.5 meters to form a sensor network. The measurement ranges of these sensors are ±16g and ±2000° / s respectively, and the sampling frequency is 200Hz, which can capture the acceleration and angular velocity changes of the drilling tool in three-dimensional space in real time. The original data are transmitted to a data acquisition unit through a CAN bus and are subjected to preliminary filtering to remove high-frequency noise. In order to compensate for the influence of local deformation of the drilling tool on position measurement, a strain sensor array is arranged around the installation point of each motion sensor group. Specifically, eight strain gauges are arranged uniformly around the installation point of the sensor in the circumferential direction, with a measurement range of ±3000με and a sensitivity of 2mV / με. The collected strain data are used to establish a local deformation vector field of the drilling tool. Through interpolation fitting of multi-point strain values, the deformation distribution within a range of 200mm around the installation point can be obtained. The least square method is applied to decompose the deformation field to calculate the displacement components of the sensor installation point in the radial and axial directions, with typical values of 0.2-1.5mm and 0.1-0.8mm respectively. These displacement values are removed from the original motion data to obtain the compensated real-time space position and attitude angle data of the drilling tool, with a position accuracy of ±3cm and an angle accuracy of ±0.5°.

[0122] When detecting the weak connection zone at the end of the drill bit, 12 groups of acoustic transducers are arranged uniformly in the circumferential direction. Each group of transducers includes a transmitting unit and a receiving unit. The transmitting unit generates a narrow pulse ultrasonic signal with a frequency of 200kHz, a peak power of 50W and a pulse width of 10μs. Adjacent transducer groups work in an alternating manner, and the transmission timing interval is 5ms to avoid signal interference. When the detection distance is 300mm, the typical reflection signal intensity is-40dB to-60dB, and the signal-to-noise ratio is about 15-25dB.

[0123] The collected reflection signals are processed to enhance the boundary features. The specific steps are as follows: first, the reflection signals are divided into 30 segments according to the propagation time, and each segment corresponds to a 10μs time window; then, the envelope extraction and phase alignment are performed on each segment of signal; then, the phase superposition of adjacent 5 segments of signal is performed to enhance the response of continuous boundary; finally, a mapping relationship table of 360° scanning angle and reflection signal amplitude is constructed, with a resolution of 1°. In practical application, the signal amplitude at the boundary of the weak connection zone is usually 6-12dB higher than that of the surrounding rock mass, showing obvious peak characteristics.

[0124] Based on the mapping relationship, the connection belt boundary and the density change are identified. When the reflected signal amplitude is continuously detected to be higher than the threshold value (usually 8 dB higher than the average background noise) in a certain angular direction, and the continuous angular range exceeds 15°, it is determined that there is a weak connection belt. The strength of the reflected signal is negatively correlated with the density of the connection belt. For every 3 dB increase in the amplitude of the reflected signal, the density of the corresponding connection belt decreases by about 20%. By marking the start angle and the end angle of the high-amplitude area in the scanning result, the distribution range of the weak connection belt in the circumferential direction of the drilling tool is determined. Combined with the compensated real-time spatial position data of the drilling tool, a three-dimensional spatial coordinate transformation is used to map the local scanning result to the global coordinate system, and the spatial distribution range and the strike characteristics of the weak connection belt are determined. For example, during a certain drilling process, it is detected that there is a weak connection belt located 15-30 cm in front of the drill bit, and significant reflected signals are detected within the range of 120° to 210° in the circumferential direction of the drilling tool, with an average signal strength of -42 dB. Combined with the drilling tool position data analysis, the connection belt has a strike of north by east 40°, an inclination of about 65°, a width of about 25 cm, and a density estimated to be 60% of the surrounding rock mass.

[0125] The position deviation of the drilling tool from each attitude control point is calculated. First, the corrected real-time spatial position and attitude angle of the drilling tool are projected to the coordinate system of each attitude control point through spatial coordinate transformation. The attitude control points are usually set at key positions of the drilling tool trajectory, such as turning points, target layer positions, etc. According to the strength and strike characteristics of the weak connection belt described above, the force influence coefficient of the drilling tool at each control point is established. Specifically, when the drilling tool passes through a weak connection belt with a strength of 60% of the standard rock mass, the force distribution of the drill bit is uneven, resulting in a drilling tool trajectory deviation coefficient of about 0.05-0.15 m / kN, which depends on the intersection angle between the connection belt and the drilling tool axis.

[0126] When correcting each projected coordinate, the force deviation of the drilling tool is considered. For example, when a weak connection belt with a strength of 65% of the surrounding rock mass is detected at an intersection angle of 30° in front of the drilling tool, the corresponding trajectory correction amount is 8-12 cm. By comparing the corrected coordinates with the attitude control point coordinates, the vector difference is calculated to obtain the position deviation of the drilling tool from each attitude control point. In practical applications, the system can provide a three-dimensional deviation vector of the drilling tool relative to the control points in real time, with a typical accuracy of ±5 cm and an update frequency of 10 Hz, providing a basis for trajectory adjustment during drilling.

[0127] In the embodiment, high-precision real-time perception and deviation calculation of the position and attitude of the drilling tool in a complex formation environment can be achieved. Compared with the existing positioning method relying on a single sensor, the scheme integrates motion sensor and strain sensor data to construct a local deformation vector field, effectively eliminating the interference of the deformation of the drilling tool on the attitude measurement, and improving the accuracy of the spatial position and attitude from the source. At the same time, by laying out the acoustic transducers and performing phase superposition enhancement, the recognition ability of the weak connection band boundary and the density change is improved, so that the drilling tool can accurately perceive the structural characteristics of the current drilling area. Combined with the spatial coordinate projection and the stress influence modeling of the connection band, the dynamic correction of the spatial deviation between the drilling tool and the attitude control point is realized. The problems of attitude error caused by the complexity of the formation and the deformation of the drilling tool are solved, and finally the path deviation is found and accurately corrected in the continuous drilling process, improving the trajectory tracking accuracy and drilling stability.

[0128] In an optional implementation, according to the position deviation and the distribution law of the weak connection band, the drilling bit telescopic displacement amount, the drilling rod bending deformation amount and the mud jet pressure parameter are adjusted in real time, so that the drilling tool keeps moving along the obstacle avoidance trajectory, which includes:

[0129] According to the size of the position deviation and the distribution law of the weak connection band, the drilling process is divided into a stable drilling section with a deviation less than a preset deviation threshold and an obstacle avoidance adjustment section with a deviation greater than the preset deviation threshold;

[0130] When the drilling tool is located in the stable drilling section, the telescopic displacement amount of each quadrant of the drilling bit is calculated based on the position deviation, and the bending deformation amount of the adjacent drilling rod sections is determined according to the distribution law of the weak connection band, and the adjustment instruction is output according to the set matching relationship of the telescopic displacement amount and the bending deformation amount;

[0131] When the drilling tool is located in the rapid obstacle avoidance section, the optimal obstacle avoidance direction is calculated according to the distribution law of the weak connection band, the circumferential arrangement position of the mud nozzle is determined based on the optimal obstacle avoidance direction, the mud jet pressure parameter of each nozzle is adjusted according to the circumferential arrangement position, the obstacle avoidance propulsion force is formed, and the adjustment instruction is output according to the obstacle avoidance propulsion force;

[0132] The motion trajectory of the drilling tool after executing the adjustment instruction is collected, the new position deviation is calculated, the stress data of the drilling tool is collected, and the motion stability is evaluated, and the new position deviation and the motion stability are taken as evaluation parameters;

[0133] According to the evaluation parameters, when the new position deviation is reduced and the motion stability meets the requirements, the current adjustment parameters are kept; otherwise, the new position deviation and the motion stability are returned as inputs, the telescopic displacement amount of the drilling bit, the bending deformation amount of the drilling rod and the mud jet pressure parameter are recalculated, and the drilling tool keeps moving along the obstacle avoidance trajectory.

[0134] Exemplarily, before the drilling operation starts, the system pre-sets the position deviation threshold value as 0.8 meters, and divides the drilling process into a stable drilling section with a deviation less than 0.8 meters and an obstacle avoidance adjustment section with a deviation greater than or equal to 0.8 meters. At the same time, the system determines the distribution characteristics of the weak connection zone through geological exploration data, including parameters such as azimuth angle, inclination angle and layer thickness. For example, in the drilling process of a certain coal seam, the main weak connection zone is located at an azimuth angle of 45°, an inclination angle of 15°, and a layer thickness of about 1.2 meters.

[0135] When operating in the stable drilling section, the real-time position data of the drilling tool is first collected, such as the current position coordinates being (125.6, 78.3, -56.2) meters and the target position being (125.8, 78.5, -56.4) meters, and the position deviation is calculated as 0.4 meters. Since the deviation is less than the pre-set threshold value of 0.8 meters, the system determines that the current is in the stable drilling section. Based on the position deviation value, the extension and retraction displacement amounts of each quadrant of the drill bit are calculated. The drill bit is divided into four quadrants, corresponding to the positive up, positive right, positive down and positive left directions respectively. For the above-mentioned deviation of 0.4 meters, the system calculates that the extension and retraction displacement amount of the first quadrant (positive up direction) of the drill bit is +3.5 millimeters, the second quadrant (positive right direction) is +2.8 millimeters, the third quadrant (positive down direction) is -1.6 millimeters, and the fourth quadrant (positive left direction) is -1.2 millimeters. This asymmetric extension and retraction configuration can generate a resultant force towards the target direction.

[0136] Considering the detected distribution law of the weak connection zone (azimuth angle 45°, inclination angle 15°), the bending deformation amount of adjacent drill rod joints is determined. For each node of the drill rod, the system sets different bending deformation amounts. For example, the bending deformation amount between the first node and the second node is 2.2 degrees, between the second node and the third node is 1.8 degrees, and between the third node and the fourth node is 1.5 degrees, decreasing in turn, forming a smooth bending profile, ensuring that the drilling tool can advance along the optimal path that avoids the weak connection zone.

[0137] According to the pre-set ratio of extension and retraction displacement amount to bending deformation amount of 1:5 (i.e. 1 millimeter of extension and retraction displacement corresponds to 5 degrees of bending deformation), adjustment instructions are generated and issued to the drilling tool actuator through the drilling control system. For example, when the extension and retraction displacement amount of the first quadrant is +3.5 millimeters, the corresponding bending deformation control amount is 17.5 degrees. When the drilling tool enters the obstacle avoidance adjustment section, such as when the position deviation increases to 0.9 meters, exceeding the pre-set threshold value of 0.8 meters, the system automatically switches to the obstacle avoidance adjustment mode. At this time, the system first calculates the optimal obstacle avoidance direction according to the distribution law of the weak connection zone. Considering that the current weak connection zone is located at an azimuth angle of 45°, the system determines the optimal obstacle avoidance direction as an azimuth angle of 225° (i.e. the opposite direction of the weak zone).

[0138] Based on the calculated optimal obstacle avoidance direction 225°, the circumferential arrangement position of the mud nozzles is adjusted. The standard drill bit is equipped with 8 circumferentially evenly distributed mud nozzles, and the system sets 3 nozzles within the range of 225°±30° (i.e. nozzles located in the interval of 195°-255°) as low-pressure zones, 3 nozzles within the range of 45°±30° (i.e. nozzles located in the interval of 15°-75°) as high-pressure zones, and the remaining nozzles maintain standard pressure. The mud injection pressure parameters of each nozzle are precisely adjusted. The pressure of the high-pressure zone nozzles is set to 18 MPa, the standard pressure zone is set to 12 MPa, and the low-pressure zone is set to 6 MPa. This pressure distribution forms an obstacle avoidance thrust towards the 225° azimuth angle, with a value of about 2.5 kN, which is sufficient to guide the drill bit to follow the predetermined trajectory to avoid obstacles. The system issues these parameters as adjustment instructions to the mud pump control system.

[0139] After executing the adjustment instructions, the system collects the motion trajectory data of the drilling tool through the position sensor and gyroscope built-in the drilling tool. For example, after 10 minutes of adjustment, the drilling tool position is updated to (126.2, 78.9, -56.7) meters, and the new position deviation is reduced to 0.3 meters. At the same time, the system collects the drilling tool force data, including axial force 12.6 kN, torque 420 N·m, and lateral force 1.8 kN. By analyzing these data to evaluate the motion stability, three parameters are mainly investigated: moment fluctuation rate less than 8%, drilling speed stability higher than 92%, and vibration amplitude less than the critical value 0.5g. The current measured moment fluctuation rate is 5.2%, the drilling speed stability is 96.3%, and the vibration amplitude is 0.32g, all of which meet the stability requirements.

[0140] According to the new position deviation of 0.3 meters (less than the threshold of 0.8 meters) and the motion stability meeting the requirements, it is decided to maintain the current adjustment parameters and continue drilling. If the new position deviation is not reduced or the motion is unstable, the system will re-input these data into the calculation unit, adjust the drill bit extension displacement, drill pipe bending deformation, and mud injection pressure parameters, and execute the above steps in a loop until the drilling tool maintains stable motion along the predetermined obstacle avoidance trajectory.

[0141] Based on the technical scheme, dynamic deviation sensing and adaptive adjustment control of the drilling tool in a complex stratum environment can be realized, so that the drilling tool can be accurately pushed along the obstacle avoidance trajectory while maintaining a stable posture. The prior art generally cannot respond to the deviation and stratum structure changes in the drilling process in real time, and the adjustment means is single and the feedback is lagging, which can easily cause the drill bit to deviate from the path or the drilling tool to be abnormally stressed. The present application introduces a joint criterion of position deviation and weak connection band distribution rule, dynamically divides the drilling stage, and respectively matches the applicable control strategy to maintain stable movement in the stable drilling stage and quickly generate optimal adjustment instructions in the obstacle avoidance adjustment stage. At the same time, combined with the three regulation and control modes of drill bit extension displacement, drill pipe deformation and mud injection, a real-time closed-loop adjustment mechanism is constructed to ensure that the adjustment process has continuous feedback and stability evaluation capability. The scheme starts from improving the continuity of path tracking and the accuracy of obstacle avoidance adjustment, effectively improves the problems of response lag and low control accuracy of the prior art, and realizes the cooperative adaptive optimization of the drilling tool posture and path.

[0142] In a second aspect, the present application provides an electronic device, comprising:

[0143] a processor;

[0144] a memory for storing processor-executable instructions;

[0145] The processor is configured to invoke the instructions stored in the memory to execute the method described above.

[0146] In a third aspect, the present application provides a computer-readable storage medium having stored thereon computer program instructions, which, when executed by a processor, implement the method described above.

[0147] The present application can be a method, device, system and / or computer program product. The computer program product can include a computer readable storage medium having computer readable program instructions loaded thereon for executing various aspects of the present application.

[0148] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to part or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for intelligent control of drilling parameters in the construction of bored piles in cobblestone strata, characterized in that, include: A ring-shaped acoustic wave emitting array is set at the end of the drill bit to collect the reflection signal of the sound wave between the pebbles. The reflected signal is subjected to spectrum analysis and waveform superposition processing to extract the attenuation characteristics and reflection characteristics of the sound wave in different media and to draw a pebble distribution density map in the three-dimensional space in front of the drill bit. Based on the pebble distribution density map, weak connecting zones between adjacent pebbles are identified as passable areas. The weak connecting zones in the passable areas are classified according to their width to obtain weak connecting zones of different width levels. The center line of each level of weak connecting zone is extracted to establish a multi-level weak connecting zone network that includes the center line and lateral connection relationships. In the multi-level weak connection zone network, a dynamic coordinate system is established with the drill string center of gravity as the origin, and the distance and direction angle from the drill string center of gravity to the center line of each weak connection zone are calculated; based on the distance, direction angle and lateral connection cost, a pass evaluation function of the logical weight model is constructed, wherein the lateral connection cost decreases as the width of the weak connection zone increases; The passage evaluation function is used to perform a bidirectional search in a multi-level weak connection network: while expanding the search space forward from the current position to the target position, the search space is expanded backward from the target position to the current position; the optimal expansion direction is determined based on the evaluation function values ​​of the two search directions, and the weak connection sequence with the smallest passage evaluation function value is selected as the obstacle avoidance trajectory; Attitude control points are set on the obstacle avoidance trajectory, and the spacing between adjacent attitude control points is proportional to the width of the corresponding weak connection zone; in the area where the width of the weak connection zone changes, transition control points are added by minimizing the local curvature change rate to ensure a continuous and smooth transition of the trajectory. For each attitude control point, calculate the extension and retraction displacement of each quadrant of the drill bit and the bending deformation of the drill rod, and adjust the discharge direction and injection pressure of the mud nozzles in each quadrant to form a directional softening channel so that the drill string passes through each control point in sequence. The system collects real-time data on the spatial position and attitude angle of the drill bit, extracts the strength and orientation characteristics of the weak connection zone in the current drilling area, calculates the positional deviation between the drill bit and each attitude control point, and adjusts the drill bit extension and retraction displacement, drill rod bending deformation, and mud injection pressure parameters in real time based on the positional deviation and the distribution pattern of the weak connection zone, so that the drill bit keeps moving along the obstacle avoidance trajectory.

2. The method according to claim 1, characterized in that, The process involves collecting reflected signals of sound waves among pebbles, performing spectral analysis and waveform superposition on the reflected signals, extracting the attenuation and reflection characteristics of sound waves in different media, and drawing a pebble distribution density map within a three-dimensional spatial range in front of the drill bit. Wavelet decomposition is performed on the reflected signal to obtain energy components of different frequency bands. The amplitude attenuation rate and phase delay value of each frequency band signal are extracted. Based on the difference in the attenuation degree of soil and pebbles on sound waves of different frequencies, the change in acoustic impedance of the medium along the propagation path is calculated. Based on the abrupt change point of the acoustic impedance change, the spatial coordinates of the medium interface on the acoustic wave propagation path are determined. The spatial coordinates are used as the target position. The received reflected signals are delayed and superimposed using a spatial beamforming method. The time delay compensation of each reflected signal is dynamically adjusted according to the phase delay value so that the echo signal at the target position is enhanced in phase. The three-dimensional space in front of the drill bit is divided into multiple layers of concentric circular grid voxels. For each voxel location, the probability value of the pebble at the voxel location is calculated based on the echo signal intensity, phase delay value, and acoustic impedance change. A three-dimensional distance weighted interpolation algorithm is used to process the probability values ​​of the existence of pebbles into a continuous form, thereby generating a pebble distribution density map within the three-dimensional space in front of the drill bit.

3. The method according to claim 1, characterized in that, Based on the pebble distribution density map, weak connecting zones between adjacent pebbles are identified as passable areas, including: Multi-scale density threshold segmentation was performed on the pebble distribution density map, and the optimal segmentation threshold was determined by an adaptive iterative method to obtain a binary density image of the pebble region. The boundary contour of pebbles is extracted from the binary density image. A morphological operator with adjustable structural radius is used to perform a closing operation to eliminate voids inside the boundary. An edge-preserving morphological opening operation is used to smooth the boundary, resulting in a regularized pebble region. An anisotropic distance field is constructed for the regularized pebble region. The distance weight is adaptively adjusted based on the density gradient information to calculate a directional distance gradient map. The continuous region with the largest gradient value is extracted from the distance gradient map as the initial boundary zone. Density sampling is performed along the center line of the initial boundary zone to obtain a density sampling sequence. Based on the density sampling sequence, region growing is performed. Continuous regions with density values ​​lower than the dynamic threshold are expanded and marked as weak connectivity zones. Morphological optimization is performed on the weak connectivity zones to obtain the final passable region.

4. The method according to claim 1, characterized in that, For each attitude control point, the extension and retraction displacement of the drill bit in each quadrant and the bending deformation of the drill pipe are calculated. The discharge direction and injection pressure of the mud nozzles in each quadrant are adjusted to create directional softening channels, allowing the drill string to sequentially pass through each control point, including: A local coordinate system is established at the attitude control point, and the drill bit is divided into four quadrants: up, down, left, and right. The extension and retraction displacement of each quadrant of the drill bit is calculated. The bending deformation of the drill rod is calculated based on the elastic deformation theory of the drill rod. The extension and retraction correction coefficient of each quadrant is obtained by calculating the ratio of the extension and retraction displacement to the bending deformation. Collect the rock hardness difference between each quadrant of the drill bit and the well wall, calculate the rock breaking power ratio of each quadrant, and calculate the initial softening weight of each quadrant by multiplying the rock breaking power ratio by the stretching correction coefficient. Based on the initial softening weight, the contact pressure between the drill bit and the well wall in each quadrant, and the bending deformation of the drill pipe, the mud nozzles in each quadrant are initially adjusted to obtain the initial injection state parameters of each quadrant. The spatial deviation between the drill string position and the attitude control point is collected in real time. The initial softening weight is corrected by combining the initial injection state parameters to obtain the real-time softening weight. Based on the real-time softening weight, the injection pressure and liquid discharge direction of the mud nozzles in each quadrant are dynamically adjusted to form a directional softening channel, so that the drill string passes through each attitude control point in sequence.

5. The method according to claim 1, characterized in that, Real-time acquisition of the spatial position and attitude angle data of the drill string; extraction of the strength and orientation characteristics of the weak connection zone in the current drilling area; calculation of the positional deviation between the drill string and each attitude control point, including: A motion sensor group is set along the axial direction on the drill body to collect raw motion data; a strain sensor array is set around the installation point of the motion sensor group to establish a local deformation vector field of the drill; the radial displacement and axial displacement of the sensor installation point are calculated by decomposing the deformation vector field, and the radial displacement and axial displacement are eliminated from the raw motion data to obtain the compensated real-time spatial position and attitude angle of the drill. Multiple sets of acoustic transducers are arranged circumferentially at the drill bit tip. Each set of transducers includes a transmitting unit and a receiving unit. Adjacent transducer sets are controlled to alternately transmit detection signals and collect reflected signals. The reflected signals are segmented and phase-superimposed to enhance boundary features, and a correspondence between the circumferential scanning angle of the drill bit and the amplitude of the reflected signal is constructed. Based on the correspondence, the boundary and density changes of the connecting zone are identified, and the spatial distribution range and orientation characteristics of the weak connecting zone are determined by combining the compensated real-time spatial position of the drill bit. The corrected real-time spatial position and attitude angle of the drill bit are projected onto the coordinate system of each attitude control point through spatial coordinate transformation. The force influence coefficient of the drill bit at each control point is established based on the strength and orientation characteristics of the weak connection zone. The projected coordinates are corrected, and the vector difference between each corrected coordinate and the corresponding attitude control point is calculated to obtain the positional deviation between the drill bit and each attitude control point.

6. The method according to claim 1, characterized in that, Based on the distribution patterns of positional deviations and weak connection zones, the drill bit extension / retraction displacement, drill pipe bending deformation, and mud injection pressure parameters are adjusted in real time to ensure the drill string maintains its movement along the obstacle avoidance trajectory. This includes: Based on the magnitude of the positional deviation and the distribution pattern of the weak connection zone, the drilling process is divided into a stable drilling section with a deviation less than the preset deviation threshold and an obstacle avoidance and adjustment section with a deviation greater than the preset deviation threshold. When the drill bit is in the stable drilling section, the telescopic displacement of each quadrant of the drill bit is calculated based on the position deviation, and the bending deformation between adjacent drill pipe sections is determined according to the distribution law of the weak connection zone. An adjustment command is output according to the set ratio relationship between the telescopic displacement and the bending deformation. When the drill string is in the fast obstacle avoidance section, the optimal obstacle avoidance direction is calculated according to the distribution law of the weak connection zone. Based on the optimal obstacle avoidance direction, the circumferential arrangement position of the mud nozzle is determined. The mud injection pressure parameters of each nozzle are adjusted according to the circumferential arrangement position to form the obstacle avoidance propulsion force. The adjustment command is output according to the obstacle avoidance propulsion force. Collect the drill string movement trajectory after executing the adjustment command, calculate the new position deviation, collect the drill string force data, evaluate the movement stability, and use the new position deviation and movement stability as evaluation parameters; Based on the evaluation parameters, if the new positional deviation decreases and the motion stability meets the requirements, the current adjustment parameters are maintained; otherwise, the new positional deviation and motion stability are returned as inputs, and the drill bit extension and retraction displacement, drill pipe bending deformation, and mud injection pressure parameters are recalculated until the drill string maintains its movement along the obstacle avoidance trajectory.

7. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 6.

8. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 6.

Images