An adaptive integrated directional cavitation control system and method

Abstract

The application relates to the technical field of downhole directional drilling, and discloses a self-adaptive integrated directional hole-forming control system and method, which comprises a collection module, a modeling module, a self-adaptive control module and a path planning module; the collection module collects the posture parameters of a drilling tool, formation response parameters, the running state parameters of a cutter arm actuating mechanism and the radial distance of a hole wall of a hole-forming cavity in real time; the modeling module constructs a continuous distribution profile of formation hardness along a drilling hole track and a three-dimensional morphological model of the cavity; the self-adaptive control module continuously optimizes hole-forming parameters through prediction; and the path planning module generates a safe crossing path which can avoid pipe sticking. The application realizes integrated intelligent control of directional drilling and hole-forming, improves hole-forming precision, construction safety and adaptability to complex strata, and provides an efficient, intelligent and safe solution for gas extraction in a coal mine.

Description

Technical Field

[0001] This invention relates to the field of downhole directional drilling technology, specifically to an adaptive integrated directional cavity-making control system and method. Background Technology

[0002] In underground gas drainage operations in coal mines, directional cavity creation refers to the process of radially expanding the borehole wall using a cutter arm actuator after directional drilling has reached a predetermined position, thereby forming a cavity of a certain size and shape. This expands the pressure relief range of the coal seam and improves the conductivity of the gas flow channel. This technology is a key means to achieve coal seam permeability enhancement and improve gas drainage efficiency.

[0003] Existing directional drilling systems typically consist of a drilling rig, drill rod, drill bit, and cutter arm actuator. They generally employ a segmented operation mode: first, the entire directional borehole is drilled; then, the drill string is pulled out of the hole, and a new drill string assembly with a drilling tool (mechanical cutter or water jet) is lowered back into the hole to create a cavity at a predetermined depth. Some improved solutions integrate the directional measurement system and the drilling tool into the same drill string assembly, enabling drilling and cavity creation to be completed in a single run. This relies on fixed drilling parameters set by ground operators based on experience, and on indirect parameters such as ground pressure and torque to judge the drilling progress. After cavity creation, the drill string is pulled back through the cavity using the original borehole trajectory.

[0004] The above-mentioned existing technologies have the following problems:

[0005] Existing technologies employ segmented operation modes that are cumbersome and inefficient. Furthermore, during secondary drilling, the cavity-creating tools are difficult to precisely align with the predetermined position of the original directional borehole, resulting in significant deviations that negatively impact gas extraction. Integrated drill string assemblies use pre-set fixed values ​​for cavity-creating parameters, which cannot be adaptively adjusted in real-time according to formation conditions. This not only leads to low cavity-creating efficiency but also inconsistent cavity-creating effects or excessive formation damage. The lack of real-time perception of cavity morphology during cavity-creating operations prevents timely dynamic correction, resulting in difficulty in precisely controlling the cavity shape and causing problems such as substandard cavity creation or localized borehole collapse. Moreover, the size and shape of the enlarged cavity formed after cavity creation are unknown. When subsequent drill strings pass through the cavity, the lack of morphological reference can easily lead to deviation, stuck drill bits, or loss of trajectory control, affecting drilling quality and construction safety. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the present invention aims to provide an adaptive integrated directional cavity-making control system and method. The system includes a data acquisition module, a modeling module, an adaptive control module, and a path planning module. The data acquisition module collects real-time data on drill bit attitude parameters, formation response parameters, cutter arm actuator operating status parameters, and the radial distance of the cavity wall. The modeling module constructs a continuous distribution profile of formation hardness along the borehole trajectory and a three-dimensional morphological model of the cavity. The adaptive control module generates and continuously optimizes cavity-making parameters through prediction. The path planning module generates a safe crossing path to avoid drill bit jamming. This invention achieves integrated intelligent control of directional drilling and cavity making, improving cavity-making accuracy, construction safety, and adaptability to complex formations, providing an efficient, intelligent, and safe solution for underground gas extraction in coal mines.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] In a first aspect, the present invention provides an adaptive integrated directional cavitation control system, comprising:

[0009] The data acquisition module is used to acquire in real time the attitude parameters of the drill bit, formation response parameters, and operating status parameters of the cutter arm actuator, as well as the radial distance of the hole wall of the cavity.

[0010] The modeling module is used to invert and construct a continuous distribution profile of formation hardness along the borehole trajectory based on the formation response parameters, and to generate a three-dimensional morphological model of the cavity based on the radial distance of the borehole wall.

[0011] An adaptive control module is used to generate corresponding cavity-making parameters based on the predicted hardness value corresponding to the preset cavity-making position in the continuous distribution profile and in combination with preset mapping rules. When the drill bit reaches the cavity-making position, it controls the cutter arm actuator to start the cavity-making operation according to the cavity-making parameters, and corrects the cavity-making parameters based on real-time formation response parameters until the radial distance and shape of the borehole wall reach the preset cavity-making target.

[0012] The path planning module is used to generate a passage path through the cavity based on the three-dimensional morphological model after the cavity is created, and to compare the attitude parameters with the passage path to generate a closed-loop deviation adjustment command to complete the passage through the cavity.

[0013] Preferably, the specific steps for inverting and constructing a continuous distribution profile of formation hardness along the borehole trajectory include:

[0014] The first mechanical specific energy is calculated based on the formation response parameters at each sampling point during drilling; the formation response parameters include drilling torque, drilling pressure, and mechanical drilling speed.

[0015] The first mechanical specific energy is fitted with a sliding window local polynomial along the drilling trajectory to obtain the specific energy fitting function, and the values ​​at each sampling point in the specific energy fitting function are extracted as the drill bit wear factor.

[0016] The second mechanical specific energy of each sampling point is obtained by performing a difference calculation based on the first mechanical specific energy and the drill bit wear factor; a continuous distribution profile of formation hardness along the borehole trajectory is generated based on the second mechanical specific energy.

[0017] Preferably, the specific steps for generating the three-dimensional morphological model of the cavity include:

[0018] A two-dimensional grid is constructed with the drilling depth as the first dimension and the circumferential angle as the second dimension. The radial distance of the borehole wall at each sampling point is filled into the corresponding grid cell, and the grid cells that are not sampled are marked as blind zones.

[0019] Calculate the radial distance gradient change rate along the borehole depth and the radial distance curvature change rate along the circumferential angle for each sampled grid cell; if the radial distance gradient change rate or the radial distance curvature change rate exceeds a preset abrupt change threshold, mark the corresponding grid cell as the cavity abnormal boundary and use the radial distance between the borehole walls on the adjacent two sides as a discontinuity constraint condition.

[0020] Using the sampled grid cells and the discontinuity constraint as boundary conditions, the radial distance of the hole wall in the blind zone is calculated using the radial basis function interpolation method to form a complete two-dimensional grid.

[0021] Preferably, the specific steps for generating the three-dimensional morphological model of the cavity further include:

[0022] Based on the complete two-dimensional mesh, the radial distance of the hole wall at all circumferential angles for each drilling depth is combined with the corresponding circumferential angle coordinates to generate a two-dimensional closed cavity cross-sectional profile.

[0023] The cavity cross-sectional contours are arranged in order of drilling depth, and surface fitting and smoothing processes are performed to generate a point cloud of the cavity surface within a continuous drilling depth range.

[0024] Extract the geometric center at each drilling depth from the point cloud on the cavity surface, calculate the centerline offset trajectory of the cavity, identify and mark the area where the radial distance of the hole wall exceeds the preset diameter expansion threshold as the diameter expansion area; fuse the centerline offset trajectory, the diameter expansion area and the point cloud on the cavity surface to generate a three-dimensional morphological model of the cavity.

[0025] Preferably, the specific steps for constructing the mapping rules include:

[0026] Acquire historical hole-making operation data. Each historical hole-making operation data includes the second mechanical specific energy before hole-making, the cutter arm opening speed and rotation speed during hole-making, as well as the cutter arm wear per unit depth and mechanical drilling speed.

[0027] Using the second mechanical specific energy as the input variable, the cutter arm opening speed and the rotation speed as the variables to be optimized, and the objectives of minimizing the cutter arm wear per unit feed and maximizing the mechanical drilling speed, an optimization function is constructed.

[0028] Based on the second mechanical specific energy, clustering is performed to divide the hardness intervals, and the optimization function is solved in each hardness interval to obtain the optimized cavity-forming parameter set;

[0029] Calculate the efficiency index of each group of cavity-creating parameters in the optimized cavity-creating parameter set, select the cavity-creating parameter corresponding to the maximum value of the efficiency index as the target cavity-creating parameter of the hardness range, and each hardness range and the corresponding target cavity-creating parameter constitute the mapping rule; the efficiency index is the weighted ratio of the mechanical drilling rate and the wear of the cutter arm per unit depth.

[0030] Preferably, the hole-forming parameters are corrected based on real-time formation response parameters, and the specific steps include:

[0031] The target cavity-creating parameters are obtained according to the mapping rules, the cutter arm actuator is controlled to start the cavity-creating operation, and a first response model of the hydraulic system is established; the target cavity-creating parameters include at least the target cutter arm opening speed and the target rotation speed;

[0032] Based on the first response model, the hole-making operation is started, and the angle lag deviation is obtained by comparing the real-time opening angle with the target angle of the target cutter arm opening speed at the corresponding time.

[0033] Calculate the energy difference between the real-time mechanical specific energy and the predicted hardness value. If the energy difference exceeds the preset deviation threshold range, use the angle hysteresis deviation to correct the first response model to obtain the second response model.

[0034] The target cutter arm opening speed is adjusted according to the energy difference to obtain a first opening speed; and the predicted response time of the current control cycle is determined based on the second response model, and the control command corresponding to the first opening speed is issued in advance of the predicted response time.

[0035] Preferably, the hole-forming parameters are corrected based on real-time formation response parameters, and the specific steps further include:

[0036] When the deviation direction of the cutter arm contact force relative to the preset pressure range is opposite to the deviation direction of the real-time mechanical energy relative to the predicted hardness value, the first difference between the drilling torque and the target torque and the second difference between the cutter arm contact force and the target pressure are input into the preset decoupling control equation, and the first compensation amount of the cutter arm opening speed and the second compensation amount of the rotation speed are output.

[0037] The second opening speed is obtained by summing the first compensation amount with the current blade arm opening speed, and the second rotation speed is obtained by summing the second compensation amount with the current rotation speed.

[0038] Preferably, the hole-forming parameters are corrected based on real-time formation response parameters, and the specific steps further include:

[0039] The radial distance between the borehole wall and the radial distance at the same drilling depth in the preset cavity-making target are calculated to obtain the radial distance deviation, and the rate of change of the radial distance deviation is calculated.

[0040] If the radial distance deviation or the rate of change of the radial distance deviation exceeds the preset shape control threshold, the third compensation amount of the cutter arm opening speed and the fourth compensation amount of the rotation speed are calculated by the PID control algorithm.

[0041] The third opening speed is obtained by summing the third compensation amount with the current blade arm opening speed, and the third rotation speed is obtained by summing the fourth compensation amount with the current rotation speed.

[0042] Preferably, generating a passage path through the cavity includes the following steps:

[0043] Extract the stacking height sequence at the bottom of the cavity from the three-dimensional morphological model, and calculate the gravitational sag of the drill bit in the cavity based on the unit length weight and stiffness distribution of the drill bit.

[0044] The stacking height sequence is superimposed with the gravity sag to generate a compensation offset; the centerline offset trajectory is moved upward along the vertical direction by the compensation offset to generate the crossing path.

[0045] Secondly, the present invention provides an adaptive integrated directional cavitation control method, comprising:

[0046] Real-time acquisition of drill bit attitude parameters, formation response parameters, and cutter arm actuator operating status parameters, as well as the radial distance of the hole wall of the cavity;

[0047] Based on the formation response parameters, a continuous distribution profile of formation hardness along the borehole trajectory is constructed by inversion, and a three-dimensional morphological model of the cavity is generated based on the radial distance of the borehole wall.

[0048] Based on the predicted hardness value corresponding to the preset cavity-making position in the continuous distribution profile, the corresponding cavity-making parameters are generated in combination with the preset mapping rules; when the drill bit reaches the cavity-making position, the cutter arm actuator is controlled to start the cavity-making operation according to the cavity-making parameters, and the cavity-making parameters are corrected based on the real-time formation response parameters until the radial distance and shape of the borehole wall reach the preset cavity-making target.

[0049] After the cavity is created, a path for traversing the cavity is generated based on the three-dimensional morphological model. The attitude parameters are compared with the path for traversing the cavity to generate a closed-loop deviation adjustment command, thus completing the traversal of the cavity.

[0050] Compared with the prior art, the beneficial effects achieved by the present invention are as follows:

[0051] This invention utilizes formation response parameter inversion to construct a continuous distribution profile of formation hardness along the borehole trajectory and generates a three-dimensional morphological model of the cavity, achieving digital characterization of unknown formations and existing cavities, overcoming the shortcomings of blind drilling and cavity creation in traditional methods. Secondly, it generates cavity-creating parameters based on the predicted formation hardness values ​​and adjusts them in real time, significantly improving control accuracy and avoiding over-cutting or under-cutting caused by fixed parameters. This enhances the cavity-creating effect and reliability under complex geological conditions and reduces equipment wear and tear costs. Then, it generates the optimal crossing path through the cavity based on the three-dimensional morphological model, reducing the risk of stuck drill and improving downhole operation safety. This invention integrates directional drilling and cavity creation, saving time and labor, and improves the accuracy of cavity-creating parameter control, adaptability to complex geological conditions, and construction safety through adaptive control. Attached Figure Description

[0052] Figure 1 A schematic diagram of an adaptive integrated directional cavitation control system provided in an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of a downhole integrated drilling tool assembly provided in an embodiment of the present invention;

[0054] Figure 3 The flowchart illustrates an adaptive integrated directional hole-creation control method provided in an embodiment of the present invention.

[0055] Figure descriptions: 1. Directional drill bit; 2. Measuring sub; 3. Morphological detection sub; 4. Control sub; 5. Hydraulic switching valve assembly; 6. Cutter arm actuator. Detailed Implementation

[0056] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features therein are detailed descriptions of the technical solution of the present invention, rather than limitations thereof. It should be noted that the terms "first," "second," etc., involved in the embodiments of the present invention are only used to distinguish features of the same type and should not be construed as indicating relative importance, quantity, order, etc.

[0057] Example 1:

[0058] like Figure 1 As shown, this embodiment provides an adaptive integrated directional cavitation control system, including:

[0059] The data acquisition module is used to acquire in real time the attitude parameters of the drill bit, formation response parameters, and operating status parameters of the cutter arm actuator, as well as the radial distance of the hole wall of the cavity.

[0060] The modeling module is used to invert and construct a continuous distribution profile of formation hardness along the borehole trajectory based on the formation response parameters, and to generate a three-dimensional morphological model of the cavity based on the radial distance of the borehole wall.

[0061] An adaptive control module is used to generate corresponding cavity-making parameters based on the predicted hardness value corresponding to the preset cavity-making position in the continuous distribution profile and in combination with preset mapping rules. When the drill bit reaches the cavity-making position, it controls the cutter arm actuator to start the cavity-making operation according to the cavity-making parameters, and corrects the cavity-making parameters based on real-time formation response parameters until the radial distance and shape of the borehole wall reach the preset cavity-making target.

[0062] The path planning module is used to generate a passage path through the cavity based on the three-dimensional morphological model after the cavity is created, and to compare the attitude parameters with the passage path to generate a closed-loop deviation adjustment command to complete the passage through the cavity.

[0063] To address the problems of high blindness, poor geological adaptability, and high crossing risks in existing hole-making operations, this invention establishes a control system integrating data perception, intelligent decision-making, and execution feedback. This embodiment provides an adaptive integrated directional hole-making control system applied to directional drilling for underground gas drainage in coal mines. The system includes an underground integrated drilling tool assembly and a surface collaborative control platform, wherein the underground integrated drilling tool assembly is as follows... Figure 2As shown, the system comprises a directional drill bit 1, a measurement sub 2, a morphological detection sub 3, a control sub 4, a hydraulic switching valve assembly 5, and a cutter arm actuator 6 connected sequentially. The control sub 4 integrates a microcontroller, non-volatile memory, and solenoid valve drive circuit, and is equipped with a modeling module, an adaptive control module, and a path planning module. The hydraulic switching valve assembly 5 is driven by a solenoid valve and can switch between the main path (directional drilling cycle) and the high-pressure cavity-building path. The cutter arm actuator 6 includes a radially opening and closing mechanical cutter arm and is equipped with a contact force sensor, a speed sensor, and an angle sensor. A surface-based collaborative control platform communicates with the downhole integrated drilling tool assembly to receive operational data from multiple cavity-building sections. Based on the operational data from the preceding cavity-building sections, it collaboratively optimizes the operational parameters of subsequent cavity-building sections, generates optimized operational parameters, and sends them to the control sub 4.

[0064] For the data acquisition module:

[0065] In this embodiment, the data acquisition module is jointly executed by sensors such as the measurement sub-section 2 and the shape detection sub-section 3 in the downhole integrated drilling tool assembly. The measurement sub-section 2 integrates a triaxial accelerometer, a triaxial magnetometer, and a depth encoder. The shape detection sub-section 3 has several ultrasonic sensors evenly distributed around its circumference for acquiring the radial distance of the borehole wall. Attitude parameters, including inclination angle, azimuth angle, and tool face angle, are acquired by the triaxial accelerometer and triaxial magnetometer. Specifically, the triaxial accelerometer measures the components of the gravitational field in the three axes of the drill string to obtain the inclination angle (the angle between the drill string axis and the vertical direction) and the tool face angle (the rotation angle of the directional drilling tool relative to the borehole height). The triaxial magnetometer measures the components of the geomagnetic field in the three axes of the drill string, and calculates the azimuth angle (the direction angle of the drill string on the horizontal plane) based on the inclination angle data. The depth encoder is used to acquire the borehole depth. Formation response parameters, including drilling torque, bit pressure, and mechanical drilling speed, are obtained by measuring and calculating these parameters using the drilling torque sensor, bit pressure sensor, and accelerometer in the measurement sub-section 2. Operating parameters include cutter arm contact force, opening angle, opening speed, and rotation angle, which are measured by contact force sensors, speed sensors, and angle sensors. All the collected data corresponds one-to-one with the drilling depth and is stored in the non-volatile memory of control section 4. The acquisition frequency is set by those skilled in the art according to control response requirements.

[0066] For the modeling module:

[0067] In this embodiment, the main functions of the modeling module are: first, to invert and construct a continuous distribution profile of formation hardness along the borehole trajectory based on the formation response parameters collected during directional drilling; and second, to generate a three-dimensional morphological model of the cavity based on the radial distance data of the borehole wall collected after cavity creation. The former provides predicted hardness values ​​for the adaptive control module, and the latter provides geometric references for the path planning module to traverse the path.

[0068] In a preferred embodiment, during directional drilling, drilling torque and drilling pressure are not only affected by formation hardness but also easily interfered with by drill bit wear. This can lead to distorted hardness assessments in complex formations such as soft coal seams and alternating layers of soft and hard rock. For example, in soft coal seams, excessive drilling pressure may cause the drilling torque to increase abnormally due to the drill bit penetrating too deeply, misjudging the formation as hard. After drill bit wear, even when drilling in the same formation, the torque will gradually increase, creating the illusion of "artificially increased" formation hardness. This embodiment quantifies the interaction between the drill bit and the formation by calculating mechanical specific energy. Mechanical specific energy characterizes the mechanical energy required to break a unit volume of rock; a higher value generally indicates a harder formation, and this embodiment eliminates the interference of drill bit wear.

[0069] Furthermore, the specific steps for inverting and constructing a continuous distribution profile of formation hardness along the borehole trajectory include:

[0070] S201. Calculate the first mechanical specific energy based on the formation response parameters at each sampling point during drilling;

[0071] S202. The first mechanical specific energy is fitted with a sliding window local polynomial along the drilling trajectory to obtain the specific energy fitting function, and the value at each sampling point in the specific energy fitting function is extracted as the drill bit wear factor, which corresponds to the low frequency component.

[0072] S203. Perform a difference calculation based on the first mechanical specific energy and the drill bit wear factor to obtain the second mechanical specific energy of each sampling point; and generate a continuous distribution profile of formation hardness along the borehole trajectory based on the second mechanical specific energy.

[0073] Specifically, the mechanical specific energy is calculated using the Teale model formula based on the real-time collected formation response parameters. The first mechanical specific energy obtained at this time includes both formation hardness information and the cumulative effect of drill bit wear. In order to separate the drill bit wear factor, the first mechanical specific energy is regarded as a one-dimensional signal along the borehole trajectory. The rapid fluctuations caused by local changes in formation hardness are manifested as random fluctuations in a short depth range, which are high-frequency components. The trend drift caused by slow-changing factors such as drill bit wear is manifested as the specific energy value slowly and monotonically increasing with the increase of borehole depth, which are low-frequency components. In this embodiment, a sliding window is set along the drilling trajectory direction. The window length is set by those skilled in the art based on the effective life of the drill bit, and can be set to 20 sampling points. The sliding step size is set to 1 to 5 sampling points. A first-order polynomial function or a second-order polynomial function can be used for fitting. This invention is not limited to a specific polynomial. For example, the fitting process is as follows: the current sampling point and the first mechanical specific energy in the previous sliding window constitute a local dataset. A quadratic polynomial function is selected for fitting. The coefficients of the quadratic polynomial function are solved by the least squares method to minimize the sum of squared residuals between the function value and the measured value. The polynomial function obtained by fitting is the specific energy fitting function at the window position. For each sampling point, the value of the specific energy fitting function at that sampling point is the low-frequency component, which is used as the drill bit wear factor at that sampling point. Since the window is sliding, each sampling point will be covered by multiple windows. In this embodiment, the fitting value when each sampling point is taken as the center of the window is used as the low-frequency component estimate of that sampling point. It should be understood that the sliding window polynomial fitting method can be replaced by other real-time trend separation methods, such as the exponentially weighted moving average method, the Kalman filter method, and the recursive least squares method. As long as the method can extract the slowly changing term representing drill bit wear from the real-time arriving first mechanical specific energy sequence and retain the rapidly fluctuating term representing formation hardness, it belongs to the equivalent implementation of this invention. By introducing a sliding window local polynomial fitting to extract low-frequency components as drill bit wear factors, the energy loss caused by equipment operating conditions can be effectively separated, improving the accuracy and reliability of the continuous distribution profile of formation hardness, providing accurate predicted hardness values ​​for the adaptive control module, and avoiding overly conservative or overly aggressive hole-making parameters due to wear misjudgment.

[0074] In one optional implementation, since drill bit wear is an irreversible cumulative process, the drill bit wear factor should show a monotonically increasing trend along the drilling depth direction. However, due to measurement noise and fitting errors, the actual calculated drill bit wear factor sequence may show a local decrease, so it needs to be corrected. The specific process is as follows: after obtaining the drill bit wear factor of a new sampling point, it is compared with the drill bit wear factor of the previous sampling point. If the drill bit wear factor of the new sampling point is less than the drill bit wear factor of the previous sampling point, then the value of the previous drill bit wear factor is forcibly taken as the drill bit wear factor of the new sampling point to ensure that it conforms to the physical law.

[0075] After determining the drill bit wear factor at each sampling point, the original first mechanical specific energy is calculated by subtracting the corresponding drill bit wear factor to obtain the second mechanical specific energy, which can characterize the true change in formation hardness. Based on this, using borehole depth as the abscissa and second mechanical specific energy as the ordinate, a continuous and smooth distribution profile of formation hardness is generated using interpolation or fitting methods suitable for non-uniform sampling data. Taking cubic spline interpolation as an example, a smooth curve with second-order continuous differentiability is generated by fitting the discrete second mechanical specific energy sequence. The specific process is as follows: First, all sampling points are sorted in ascending order of borehole depth to form a depth sequence and the corresponding second mechanical specific energy sequence. Second, a cubic spline interpolation function is constructed. This function is a cubic polynomial in each group of adjacent depth sub-intervals. The function value at each borehole depth sampling point is the corresponding second mechanical specific energy, and the second derivative at both ends of the sequence is zero. Then, the coefficients of the cubic polynomial in each depth sub-interval are solved. The linear equation system of its second derivative can be solved using the three-moment method or the chasing method to obtain the cubic polynomial expression in each depth sub-interval. The cubic polynomials of all depth sub-intervals are spliced ​​in ascending order of borehole depth to form a smooth curve, which is the continuous distribution profile of formation hardness along the borehole trajectory. It should be understood that cubic spline interpolation is not the only option in this embodiment. When accuracy requirements are not high or computational resources are limited, piecewise linear interpolation or conformal piecewise cubic interpolation (such as PCHIP) can be used instead. If a certain smoothing effect on measurement noise is desired, smoothing spline method or local weighted regression method can be used, but attention should be paid to the selection of smoothing parameters to avoid over-smoothing and loss of true stratigraphic variation details. Any method that can generate a continuous distribution profile of subsurface hardness based on discrete second mechanical energy sampling points and meet the real-time query requirements is an equivalent embodiment of the present invention.

[0076] In a preferred embodiment, after the cavity is created, the cavity has a complex three-dimensional geometric shape. The radial distance data of the borehole wall collected by the morphology detection section 3 is a set of three-dimensional spatial sampling points discretely distributed along the drilling trajectory, covering only a limited circumferential angle range, with a large number of blind spots between the sampling points. The modeling module needs to reconstruct a complete, high-fidelity three-dimensional morphological model of the cavity from these sparse, discrete sampling data, while preserving discontinuous features such as collapse and accumulation, to provide accurate cavity geometric information for the path planning module.

[0077] Furthermore, the specific steps for generating the three-dimensional morphological model of the cavity include:

[0078] S211. Construct a two-dimensional grid with the drilling depth as the first dimension and the circumferential angle as the second dimension. Fill the corresponding grid cell with the radial distance of the hole wall of each sampling point. Mark the unsampled grid cells as blind areas.

[0079] S212. Calculate the radial distance gradient change rate along the borehole depth and the radial distance curvature change rate along the circumferential angle for each sampled grid cell; if the radial distance gradient change rate or the radial distance curvature change rate exceeds a preset abrupt change threshold, mark the corresponding grid cell as the cavity abnormal boundary and use the radial distance between the borehole walls on the adjacent two sides as a discontinuity constraint condition.

[0080] S213. Using the sampled grid cells and the discontinuity constraint as boundary conditions, the radial distance of the hole wall in the blind zone is calculated using the radial basis function interpolation method to form a complete two-dimensional grid;

[0081] S214. Based on the complete two-dimensional mesh, according to the radial distance of the hole wall at all circumferential angles for each drilling depth, combined with the corresponding circumferential angle coordinates, a two-dimensional closed cavity cross-sectional profile is generated.

[0082] S215. Arrange the cavity cross-sectional contours in order of drilling depth, and perform surface fitting and smoothing processing to generate a cavity surface point cloud within a continuous drilling depth range.

[0083] S216. Extract the geometric center at each drilling depth from the point cloud on the surface of the cavity, calculate the centerline offset trajectory of the cavity, identify and mark the area where the radial distance of the hole wall exceeds the preset diameter expansion threshold as the diameter expansion area; fuse the centerline offset trajectory, the diameter expansion area and the point cloud on the surface of the cavity to generate a three-dimensional morphological model of the cavity.

[0084] This implementation constructs a two-dimensional mesh, mapping discrete sampling points onto a regular mesh to identify unsampled blind spots, providing clear data missing identifiers for subsequent interpolation. Traditional 3D reconstruction methods typically assume the cavity walls are continuous and smooth, using smooth interpolation to fill blind spots, thus smoothing out real geological discontinuities such as collapse zones and accumulation zones. This invention identifies anomalous boundaries by calculating the radial distance gradient change rate and radial distance curvature change rate, and uses the radial distance values ​​of the borehole walls on both sides of the boundary as discontinuity constraints, ensuring that the interpolation results maintain a step change at anomalous boundaries, truly reflecting the physical morphology of the cavity. Furthermore, when generating the traversal path in subsequent processes, not only the geometric contour of the cavity is needed, but also key feature parameters and risk area identifiers, such as the location of collapse / accumulation zones, are required to construct a 3D morphological model of the cavity containing complete information, providing crucial reference for path planning.

[0085] Specifically, the process of constructing the three-dimensional morphological model is as follows:

[0086] The first step is to construct a two-dimensional grid with the borehole depth as the first dimension and the circumferential angle as the second dimension. The grid resolution for the borehole depth is set according to the sampling interval of the morphological detection section 3, and the grid resolution for the circumferential angle is set according to the circumferential distribution density of the sensor. The radial distance value of the borehole wall of each sampling point is filled into the corresponding grid cell for the borehole depth and circumferential angle. Some grid cells do not have corresponding sampling values ​​and are marked as blind zones.

[0087] The second step involves identifying anomalous boundaries by calculating the radial distance gradient rate of change along the borehole depth and the radial distance curvature rate of change along the circumferential angle for each sampled grid cell. The radial distance gradient rate is the ratio of the difference in radial distance between adjacent depths to the depth interval, while the radial distance curvature rate is the ratio of the second difference in the difference in radial distance between adjacent angles to the square of the angle interval. Preset abrupt change thresholds include gradient rate of change thresholds and curvature rate of change thresholds, which are the sum of the maximum rate of change and the safety margin under normal formation conditions. Typically, the gradient rate of change threshold is set to 20 mm / m, and the curvature rate of change threshold is set to 0.3 rad / m. When the gradient rate of change or the curvature rate of change exceeds the corresponding abrupt change threshold, the grid cell is determined to be located on the cavity anomalous boundary. At this time, the radial distance values ​​of the borehole walls on both sides of the grid cell are marked as discontinuous constraints, meaning that during interpolation, a jump must be maintained on both sides of the anomalous boundary, and a smooth transition is not allowed. This preserves the true geological features and improves the model's ability to predict actual crossing risks.

[0088] The third step involves calculating the borehole wall radial distance of the blind zone grid cells using radial basis function interpolation, under the combined constraints of the sampled grid cells and the discontinuity constraint. Specifically, the positions and borehole wall radial distances of all sampled grid cells are used as the reference points for interpolation. Each sampled grid cell is represented by a two-dimensional coordinate vector consisting of the corresponding borehole depth coordinates and circumferential angle coordinates. The borehole wall radial distance of each grid cell marked as a blind zone is the value to be estimated. In this embodiment, a multiple quadratic function is selected as the radial basis function, which has the characteristics of global smoothness and adjustable shape parameters, and can better fit the continuous variation trend of the geological body. The shape parameters are preset by those skilled in the art based on the spatial distribution density of the sampling points. When calculating the value to be estimated for the blind zone, all sampled grid cells are first traversed. Calculate the Euclidean distance between each sampled grid cell and the current blind zone. Substitute this Euclidean distance into a multiple quadratic function to obtain the influence weight of each sampled grid cell on the current blind zone. Then, multiply the influence weights of all sampled grid cells by their corresponding radial distance values, sum them up, and divide by the sum of the influence weights to obtain the initial estimate of the current blind zone. It should be noted that, considering the existence of discontinuity constraints, when calculating the blind zone estimate, only sampled grid cells located on the same side of the anomaly boundary can be weighted to retain the discontinuity feature. For blind zone grid cells that are exactly on the anomaly boundary, take the sampled values ​​from the adjacent cells on both sides that belong to the same connected region as the blind zone grid cell and extrapolate them. After filling all blind zone grid cells, a complete two-dimensional grid data field containing all grid cells is obtained.

[0089] The fourth step involves extracting the radial distance values ​​of all circumferential angles for each borehole depth based on a complete two-dimensional mesh. These values ​​are then arranged in order of circumferential angles and combined with the corresponding circumferential angle coordinates to reconstruct the two-dimensional closed cavity cross-sectional profile at that borehole depth. This cavity cross-sectional profile reflects the actual shape of the cavity on the corresponding depth plane. If the radial distance of the borehole wall is evenly distributed at each angle, the profile is approximately circular. If there is local collapse or accumulation, the profile will show corresponding depressions or bulges.

[0090] The fifth step involves creating an independent cavity cross-sectional profile for each borehole depth. This profile is discrete along the borehole depth direction, with depth intervals between adjacent cavity cross-sectional profiles. Therefore, a surface fitting and smoothing process is needed to obtain a parametric surface model that can calculate the radial distance of the borehole wall at any borehole depth and any circumferential angle. Those skilled in the art can select different surface fitting algorithms based on the data characteristics, such as B-spline surface fitting or Bezier surface fitting. This invention is not limited to a specific surface fitting algorithm. Taking B-spline surface fitting as an example, the specific fitting steps are as follows: First, determine two parameter directions: the borehole depth direction and the circumferential angle direction. To simplify the calculation, each borehole depth... The cavity cross-sectional profile at 0° is resampled at equal angles. Within the range of 0° to 360°, the corresponding radial distance of the hole wall is extracted according to the angle interval preset by those skilled in the art, so that all drilling depths have the same number of sampling points in the circumferential angle direction. The drilling depth, circumferential angle, and radial distance of the hole wall together constitute regular three-dimensional data. At the same time, the order and number of control points of the B-spline surface are determined by those skilled in the art. Usually, an order of 3 is selected, which can ensure the smoothness of the surface and avoid excessive computation. The number of control points is determined according to the complexity of the cavity cross-sectional profile and the sampling density in the drilling depth direction. The first and last control points in the circumferential angle direction are set the same to ensure that the fitted surface achieves smooth closure at 0° and 360°.Secondly, an approximation least squares fitting algorithm is used to find a set of control point grids that minimizes the sum of squared weighted distances from the fitted surface to all data points. The weighting coefficients during the fitting process are set by those skilled in the art; for data points corresponding to actual sampled grid cells, the weighting coefficients are set larger, while for data points obtained through blind zone interpolation, the weighting coefficients are set smaller. In the borehole depth direction, node vectors are determined by accumulating chord lengths, with the repetition of the two endpoints of the node vectors set to the order plus one. In the circumferential angle direction, since the circumferential angle is closed at both ends, a node vector setting method for closed B-spline surfaces is used to force the first and last control points of the circumferential angle to be equal, and the node vectors are appropriately adjusted to ensure the continuity of the borehole wall radial distance and its first derivative along the circumferential direction at 0° and 360°. After determining the node vectors, the B-spline basis function values ​​at each data point are calculated, establishing a system of linear equations between the data points and unknown control points. Each equation in the system corresponds to a data point, and the coefficient of the equation is the basis function value of that data point multiplied by the corresponding weighting coefficient. Then, the least squares criterion is used to solve the linear equations. The process involves obtaining the radial distance of the borehole wall from the control point grid. Finally, dense sampling is performed in both the borehole depth direction and the circumferential angle direction. The sampling interval is set by those skilled in the art, and is much smaller than the depth interval of the cavity cross-sectional profile. For each depth sample value, its position in the node vector in the borehole depth direction is calculated, and the control point index corresponding to a non-zero basis function value is determined. Then, all control point indices in the circumferential angle direction of the control point grid are multiplied by the corresponding basis function values, and accumulated to obtain an intermediate control curve. Along the intermediate control curve, the radial distance of the borehole wall is calculated at each angle sample value, thus outputting a three-dimensional spatial point for each depth sample value and each angle sample value. The three-dimensional spatial points of all depth and angle sample values ​​are combined to generate a cavity surface point cloud continuously distributed in both the depth and circumferential angle directions. Each point in the cavity surface point cloud is uniquely determined by the three-dimensional coordinates of the borehole depth, circumferential angle, and radial distance of the borehole wall. Furthermore, the spacing between adjacent points in the depth direction is much smaller than the original sampling spacing, realizing the transformation from a discrete cross-sectional profile to a continuous surface.

[0091] Step 6: Due to the large amount of data in the cavity surface point cloud, which is not conducive to quickly obtaining important decisions, it is necessary to further extract the key geometric features. The specific process is as follows: First, convert the polar coordinates of all points on the cavity cross-sectional profile at each drilling depth to rectangular coordinates, and calculate the average value to obtain the geometric center at that drilling depth. The geometric center coordinates include lateral offset and vertical offset. Then, connect all geometric centers sequentially along the drilling depth direction to form the cavity center offset trajectory. Second, scan the entire cavity surface point cloud, identify areas where the radial distance from the borehole wall exceeds the preset expansion threshold, mark them as expansion areas, and record their spatial positions. The preset expansion threshold is set by those skilled in the art based on the target size of the cavity design. The expansion area represents a dangerous area where the cavity is over-expanded or the surrounding rock is broken. The drill should avoid or reduce the speed of passage when passing through it. Finally, add the cavity centerline offset trajectory and expansion area identifier as additional attributes to the data structure of the cavity surface point cloud to generate a three-dimensional morphological model of the cavity, which is stored in the memory of control section 4. When the path planning module needs to generate a crossing path, it can directly query the radial distance of the borehole wall, the centerline offset, and whether it is located in the enlarged diameter zone at any drilling depth, thereby planning a safe crossing path.

[0092] Due to limitations in sensor field of view or data transmission delays, data sampling blind spots often exist, and abrupt changes may occur at the cavity boundaries. Direct interpolation can lead to model distortion. Therefore, this implementation method accurately locates boundary abrupt change points by calculating gradients and curvature change rates, and uses these as constraints for radial basis function interpolation. This solves the problem of constructing continuous high-precision models from discrete and incomplete data. The complete two-dimensional mesh preserves real geological features such as collapses and accumulations, providing a high-fidelity data foundation for subsequent cavity cross-sectional contour and surface point cloud reconstruction, thereby improving the path planning module's ability to perceive cavity obstacles. In addition, the cavity cross-sectional contour provides the cross-sectional shape of the cavity at various depths, the cavity surface point cloud constructs a continuous three-dimensional surface, and the centerline offset trajectory and expansion zone quantify the geometric features of the cavity, providing key navigation parameters for path planning. The generated three-dimensional morphological model not only visually displays the cavity morphology but also embeds the offset trajectory and danger zone markers required for traversal path planning, enabling the path planning module to avoid collapse zones, accumulation zones, and excessive expansion zones in advance, significantly improving traversal safety.

[0093] For the adaptive control module:

[0094] In this embodiment, the adaptive control module is responsible for generating and dynamically adjusting the cavity-forming parameters based on the continuous distribution profile of formation hardness provided by the modeling module, until the cavity shape meets the target cavity-forming requirements.

[0095] Furthermore, the specific steps for constructing the mapping rules include:

[0096] S301. Obtain historical hole-making operation data. Each historical hole-making operation data includes the second mechanical specific energy before hole-making, the cutter arm opening speed and rotation speed during hole-making, as well as the cutter arm wear per unit advance and mechanical drilling speed.

[0097] S302. Using the second mechanical specific energy as the input variable, the opening speed of the cutter arm and the rotation speed as the variables to be optimized, and with the goal of minimizing the wear of the cutter arm per unit feed and maximizing the mechanical drilling speed, construct an optimization function.

[0098] S303. Based on the second mechanical specific energy, cluster and divide the hardness intervals, and solve the optimization function in each hardness interval to obtain the optimized cavity-forming parameter set;

[0099] S304. Calculate the efficiency index of each group of cavity-forming parameters in the optimized cavity-forming parameter set, select the cavity-forming parameter corresponding to the maximum value of the efficiency index as the target cavity-forming parameter of the hardness range, and each hardness range and the corresponding target cavity-forming parameter constitute the mapping rule; the efficiency index is the weighted ratio of the mechanical drilling speed and the wear of the cutter arm per unit depth.

[0100] Specifically, the mapping rule aims to establish a quantitative relationship between formation hardness and optimal hole-making parameters to improve hole-making efficiency and reduce costs. It is constructed based on historical hole-making operation data. The second mechanical specific energy before hole-making represents the original formation hardness at the hole-making location. The wear on the cutter arm per unit depth is obtained by dividing the difference in cutter arm mass obtained before and after hole-making by the length of the hole-making section. The mechanical drilling rate is obtained by dividing the depth of penetration by the hole-making time. An optimization function is constructed with minimizing cutter arm wear per unit depth and maximizing the mechanical drilling rate as two optimization objectives. There is an inherent contradiction between these two objectives: increasing the cutter arm opening speed or rotation speed increases the mechanical drilling rate but exacerbates cutter arm wear; conversely, decreasing the speed extends the cutter arm's lifespan but reduces operational efficiency. Therefore, this implementation method seeks a set of optimization variables to achieve a Pareto optimal balance between the two objectives. Specifically, cluster analysis is performed on the second mechanical specific energy, dividing historical hole-making operation data into several hardness intervals. This implementation uses the K-means clustering algorithm, and the number of clusters is determined by those skilled in the art based on the distribution range of formation hardness values. For example, it is divided into five hardness intervals: extremely soft, soft, medium, hard, and extremely hard. Within each hardness interval, historical data and a genetic algorithm are used to solve the optimization function. The specific process is as follows: each set of variables to be optimized is encoded as an individual, and the fitness function is the minimization of cutter arm wear per unit depth and the maximization of mechanical drilling speed. Through selection, crossover, and mutation operations, iterative evolution is performed to finally obtain the Pareto optimal solution within that hardness interval, which constitutes the optimized hole-making parameters. The system employs a multi-stage mapping method. Based on an efficiency index, a unique solution is selected as the target cavity-building parameter for the corresponding hardness range. The efficiency index is defined as the weighted ratio of mechanical drilling speed to cutter arm wear per unit depth. The weighting coefficient is set by those skilled in the art according to project priorities. If efficiency is prioritized in the current cavity-building task, the weight of mechanical drilling speed is increased; if cutter arm life is prioritized, such as in hard formations or long-distance multi-stage cavity building, the weight of cutter arm wear per unit depth is increased. The cutter arm opening speed and rotation speed corresponding to the maximum efficiency index are selected as the target cavity-building parameters for that hardness range. A mapping rule table is constructed by associating all hardness ranges with their corresponding target cavity-building parameters and stored in the non-volatile memory of control section 4. This implementation method enables the adaptive control module to quickly look up the table based on the predicted hardness value when facing unknown formations, obtaining target cavity-building parameters that balance efficiency and lifespan. This achieves optimal matching of cavity-building parameters under different hardness ranges, improving cavity-building efficiency while reducing cutter arm wear.

[0101] In a preferred embodiment, during the cavity creation process, the opening and closing of the cutter arm actuator 6 depends on the pressure build-up and release process of the downhole hydraulic system. When the control system determines that the cutter arm opening speed needs to be adjusted based on the real-time collected formation response parameters, there is a hydraulic response delay from the issuance of the command to the actual action of the cutter arm. During this delay, continued rotation and cutting of the drill string may lead to overcutting or undercutting effects. Overcutting refers to the phenomenon where, when the formation hardens, the hydraulic response delay causes the cutter arm actuator 6 to fail to decelerate in time, resulting in excessive torque, cutter arm overload, or even breakage. Undercutting refers to the phenomenon where, when the formation softens, the hydraulic response delay causes the cutter arm actuator 6 to fail to accelerate in time, resulting in reduced cavity creation efficiency and irregular cavity shape. Therefore, after obtaining the target cavity creation parameters, time-series compensation is actively performed using formation response parameters to reduce the response delay and adapt to rapidly changing formations.

[0102] Furthermore, the hole-forming parameters are corrected based on real-time formation response parameters, specifically including the following steps:

[0103] S311. Obtain the target cavity-creating parameters according to the mapping rules, control the cutter arm actuator to start the cavity-creating operation, and establish the first response model of the hydraulic system; the target cavity-creating parameters include at least the target cutter arm opening speed and the target rotation speed;

[0104] S312. Start the cavity-making operation based on the first response model, and compare the real-time opening angle with the target angle of the target cutter arm opening speed at the corresponding time to obtain the angle lag deviation.

[0105] S313. Calculate the energy difference between the real-time mechanical specific energy and the predicted hardness value. If the energy difference exceeds the preset deviation threshold range, use the angle hysteresis deviation to correct the first response model and obtain the second response model.

[0106] S314. Adjust the target cutter arm opening speed according to the energy difference to obtain a first opening speed; and determine the predicted response time of the current control cycle based on the second response model, and issue the control command corresponding to the first opening speed in advance by the predicted response time.

[0107] Specifically, in this embodiment, the preset cavity-forming position is pre-set by those skilled in the art. Based on the drilling depth corresponding to the preset cavity-forming position, a continuous distribution profile is queried to obtain the corresponding second mechanical energy, which is the predicted hardness value. This is then combined with the aforementioned mapping rules to obtain the target cavity-forming parameters. Simultaneously, a first response model of the hydraulic system is established using a micro-motion testing method. The specific process is as follows: when the drill bit reaches the preset cavity-forming position and the cutter arm actuator 6 is fully closed, a short-time pulse width modulation signal command is sent to the hydraulic switching valve group 5 via the solenoid valve drive circuit, causing the cutter arm actuator 6 to perform a micro-opening and immediate retraction action. The micro-opening amplitude is set to within five percent of the maximum opening angle, ensuring sufficient detection by the angle sensor without substantial contact with the borehole wall. Simultaneously, a high-speed timer is started to record the actual opening angle value of the cutter arm fed back by the angle sensor. The moment when the actual opening angle value first exceeds the static noise threshold of the cutter arm closed state is recorded as the start time of the action. This is subtracted from the command issuance time to obtain the delay time of this test. Multiple tests are performed, and the average delay time is obtained, constituting the first response model.

[0108] Based on the above implementation method, when issuing the hole-making parameter adjustment command, the command is sent in advance according to the average delay time of the first response model to start hole making. Since the delay time will drift during the actual hole making process, the actual opening angle of the cutter arm is compared with the target angle obtained by interpolation according to the target cutter arm opening speed in real time to calculate the angle lag deviation. The angle lag deviation reflects the error of the current delay relative to the estimated value of the first response model. The real-time mechanical specific energy of the hole-making location is compared with the predicted hardness value to calculate the specific energy difference. If the specific energy difference exceeds the preset deviation threshold range, it indicates that the actual formation hardness deviates greatly from the expected judgment, and the cutter arm opening speed needs to be adjusted. The preset deviation threshold range is determined by those skilled in the art based on the statistical distribution of specific energy fluctuations in historical hole-making operations, and is usually set to ±10% to ±15% of the predicted hardness value. First, the first response model is corrected using the angle lag deviation to obtain the second response model. The specific process is as follows: if the angle lag deviation is positive, the delay time of the first response model is increased by one. A correction amount proportional to the angular lag deviation is applied. If the angular lag deviation is negative, the delay time of the first response model is reduced by a correction amount proportional to the angular lag deviation. Simultaneously, the magnitude of the energy difference exceeding a preset deviation threshold is multiplied by a preset proportional coefficient to obtain the speed correction amount. The target cutter arm opening speed is then superimposed with the speed correction amount to obtain the first opening speed. In this embodiment, the proportional coefficients for calculating each correction amount are determined by those skilled in the art based on the empirical transfer relationship between the change in cutter arm opening speed and the change in mechanical energy, so as to generate the gain value required for the energy response amount per unit speed adjustment. The delay time of the second response model is the corrected predicted response time. The issuance time of the control command for the first opening speed is advanced by this predicted response time, so that the actual movement of the cutter arm can be synchronized with the change in formation hardness. In this embodiment, the control cycle is the same as the data acquisition cycle in section 3 of the morphological detection. By actively compensating for the dynamic lag of the hydraulic system, the actual action timing of the cutter arm actuator is synchronized with the changes in formation hardness, eliminating overcutting or undercutting caused by hydraulic system delays, improving the real-time performance and effectiveness of parameter correction during cavity creation, and avoiding correction commands lagging behind formation changes.

[0109] For example, when the drill bit reaches the preset cavity-making position and the cutter arm actuator is fully closed, a micro-motion test is performed. This involves issuing a PWM command with a pulse width of 100ms to slightly open the cutter arm to 3% of its maximum opening angle, then immediately retracting it. If the maximum opening angle is 60°, the slight opening angle is 1.8°. The command issuance time is recorded as 0ms. The high-speed timer repeats this process three times, recording the times when the angle sensor first detects the angle change as 215ms, 225ms, and 220ms respectively. The average delay time is 220ms, constituting the first response model. After the cavity creation begins, at the current time t=2s, the target blade arm opening speed is 1.0mm / s. Assuming the conversion factor between angle and opening distance is 1° / mm, the cumulative opening angle of the target from the start of cavity creation to the current time should be 2.0°. However, the actual opening angle collected by the angle sensor is 1.85°, resulting in an angle lag deviation of 2.0 - 1.85 = 0.15°, which translates to a time deviation of 0.15s, or 150ms. Assuming the real-time mechanical specific energy is 0.6MJ / m³,... 3 The predicted hardness value is 0.45 MJ / m. 3 Therefore, the energy difference is 0.6 - 0.45 = +0.15 MJ / m 3 The preset deviation threshold range is ±12% of the predicted hardness value, i.e., ±0.054 MJ / m. 3 If the actual specific energy difference is 0.15 > 0.054, exceeding the preset deviation threshold, the first response model needs to be corrected. The predicted response time of the second response model is 220 + 150 = 370 ms, and the magnitude of the specific energy difference exceeding the preset deviation threshold is 0.15 - 0.054 = 0.096 MJ / m². 3 Assuming the proportionality coefficient is 0.5 (mm / s) / (MJ / m 3 If the speed correction is 0.096 × 0.5 = 0.048 mm / s, and the first opening speed is 1.0 - 0.048 = 0.952 mm / s, then the control command for the blade arm opening speed of 0.952 mm / s is issued 370 ms in advance.

[0110] In a preferred embodiment, during the hole-making process, the effects of the cutter arm opening speed and rotation speed on the formation response parameters are coupled. Increasing the cutter arm opening speed increases the contact force between the cutter arm and the borehole wall, and simultaneously increases the drilling torque; while increasing the rotation speed increases the drilling torque, but may decrease the contact force between the cutter arm and the borehole wall. In actual working conditions, it is difficult to determine which hole-making parameter should be adjusted in a limited way. For example, when the formation hardens, it is necessary to reduce the cutter arm opening speed to decrease torque, but low contact force requires increasing either the cutter arm opening speed or the rotation speed to increase cutting force, resulting in opposite control command directions. Adjusting both parameters simultaneously may cause system oscillations. Therefore, it is necessary to eliminate the coupling conflict of multiple variables when adjusting the hole-making parameters.

[0111] Furthermore, the hole-forming parameters are corrected based on real-time formation response parameters, and the specific steps also include:

[0112] S321. When the deviation direction of the cutter arm contact force relative to the preset pressure range is opposite to the deviation direction of the real-time mechanical energy relative to the predicted hardness value, the first difference between the drilling torque and the target torque and the second difference between the cutter arm contact force and the target pressure are input into the preset decoupling control equation, and the first compensation amount of the cutter arm opening speed and the second compensation amount of the rotation speed are output.

[0113] S322. The first compensation amount is summed with the current blade arm opening speed to obtain the second opening speed, and the second compensation amount is summed with the current rotation speed to obtain the second rotation speed.

[0114] In this embodiment, the coupling conflict state is identified by determining whether the direction of contact force deviation is opposite to the direction of mechanical energy deviation. Specifically, the preset pressure range of the contact force is predetermined by those skilled in the art based on the predicted hardness value and the target cavity-forming parameters. The direction of contact force deviation represents the deviation trend of the real-time tool arm contact force relative to the preset pressure range. When the real-time tool arm contact force is less than the lower limit of the preset pressure range, the contact force deviation direction is negative; when it is greater than the upper limit of the preset pressure range, the contact force deviation direction is positive. Similarly, the direction of mechanical energy deviation represents the deviation trend of the real-time mechanical energy relative to the predicted hardness value. When the real-time mechanical energy is less than the predicted hardness value, it is negative; otherwise, it is positive. If the current contact force deviation direction is opposite to the mechanical energy deviation direction, it indicates that a control command conflict has occurred, and decoupling control needs to be initiated. In this embodiment, the target torque and target pressure are determined based on the predicted hardness value and target cavity-forming parameters combined with experimental testing. The decoupling control equation is established based on the partial derivative coefficients of the cutter arm opening speed and rotation speed with respect to the drilling torque and cutter arm contact force, respectively. Each partial derivative coefficient is calibrated through offline testing. The specific process is as follows: the drill string assembly is fixed on the test bench to simulate typical formation conditions. First, the cutter arm opening speed is kept constant, the rotation speed is changed, and the changes in drilling torque and cutter arm contact force are measured. The change in torque is divided by the change in rotation speed to obtain the partial derivative coefficient of torque with respect to rotation speed, and the change in contact force is divided by the change in rotation speed to obtain the partial derivative coefficient of contact force with respect to rotation speed. Then, the rotation speed is kept constant, the cutter arm opening speed is changed, and the changes in drilling torque and contact force are measured. Similarly, the partial derivative coefficients of torque with respect to opening speed and contact force with respect to opening speed are obtained, which together form a 2×2 coefficient matrix. It is important to note that it is necessary to simulate strata of different hardness for test calibration. When fixing a cavity-building parameter, different benchmark operating points that can be stably operated should be selected in combination with the working conditions and hardware constraints. Repeat the test multiple times to obtain multiple sets of partial derivative coefficients. In actual use, select a set of partial derivative coefficients that are close to the predicted hardness value and the current cavity-building parameter as the parameter configuration of the decoupled control equation to be called.

[0115] Based on the above implementation method, the first difference between the real-time drilling torque and the target torque, and the second difference between the real-time cutter arm contact force and the target pressure are input into the decoupling control equation. Through the inverse transformation of the coefficient matrix, the first compensation amount for the cutter arm opening speed and the second compensation amount for the rotation speed are output. These two compensation amounts can simultaneously eliminate the deviation of drilling torque and cutter arm contact force without generating coupling conflicts. The first compensation amount is added to the current cutter arm opening speed to obtain the second opening speed, and the second compensation amount is added to the current rotation speed to obtain the second rotation speed, which is then output to the cutter arm actuator 6. During the cavity creation process, the effects of the cutter arm opening speed and rotation speed on the drilling torque and cutter arm contact force are coupled. Decoupling is achieved when there are multi-variable coupling conflicts, avoiding mutual cancellation of commands, ensuring the stability and convergence of cavity creation parameter correction, preventing control system oscillation, and improving the smoothness of the cavity creation process and the consistency of cavity morphology.

[0116] For example, assume the predicted hardness value is 0.45 MJ / m. 3 The real-time mechanical specific energy is 0.62 MJ / m 3 If the mechanical specific energy deviation direction is positive, and the preset pressure range is 2.0~2.5MPa, while the actual cutter arm contact force is 1.8MPa, then the contact force deviation direction is negative. Since the two deviation directions are opposite, it is determined to be a coupling conflict state. The target torque is 800 N·m, and the real-time drilling torque is 850 N·m, so the first difference ΔT = 850 - 800 = 50 N·m; the target pressure is 2.2MPa, and the real-time contact force is 1.8MPa, so the second difference ΔP = 1.8 - 2.2 = -0.4MPa. Assume the partial derivative coefficients of the decoupling control equations are a... 11 =-0.05mm / (s·N·m), a 12 =0.03mm / (s·MPa), a 21 =-2rpm / (N·m), a 22 =-1.5rpm / MPa, the decoupling control equation is Δv1=a 11 ×ΔT+a 12 ×ΔP, Δv2=a 21 ×ΔT+a 22 Substituting ×ΔP into specific values, we obtain the first compensation amount Δv1 = -2.512 mm / s and the second compensation amount Δv2 = -99.4 rpm. The current tool arm opening speed is added to the first compensation amount to obtain the second opening speed, and the current rotation speed is added to the second compensation amount to obtain the second rotation speed. Amplitude limiting is then performed according to physical conditions.

[0117] In a preferred embodiment, during the cavity creation process, not only should the cavity creation parameters be adjusted in real time according to the formation response parameters, but also adjustments should be made based on the feedback of the cavity morphology. During the process, uneven cutting of the cavity in the circumferential or axial direction may occur, forming local protrusions or step-like abrupt changes. These morphological defects cannot be perceived by the formation response parameters, but they will directly affect the gas extraction effect and the safety of subsequent drilling tool crossing.

[0118] Furthermore, the hole-forming parameters are corrected based on real-time formation response parameters, and the specific steps also include:

[0119] S331. Perform a difference calculation between the radial distance of the borehole wall and the radial distance of the target at the same drilling depth in the preset cavity-making target to obtain the radial distance deviation, and calculate the rate of change of the radial distance deviation.

[0120] S332. If the radial distance deviation or the rate of change of the radial distance deviation exceeds the preset shape control threshold, the third compensation amount of the blade arm opening speed and the fourth compensation amount of the rotation speed are calculated by the PID control algorithm.

[0121] S333. Summing the third compensation amount with the current blade opening speed to obtain the third opening speed, and summing the fourth compensation amount with the current rotation speed to obtain the third rotation speed; until the radial distance deviation is less than the preset size difference threshold and the radial distance deviation change rate is less than the preset smoothing threshold, the hole-making is determined to be completed.

[0122] In this embodiment, the preset cavity-creating target is designed and determined by those skilled in the art based on cavity-creating requirements, including the cavity radius required for gas extraction at different drilling depths, i.e., the target radial distance. The radial distance deviation is obtained by subtracting the real-time borehole wall radial distance from the target radial distance. Simultaneously, based on the radial distance deviation within the historical sampling interval, the radial distance deviation change rate, which characterizes the smoothness of the cavity's axial contour, is calculated. Specifically, the preset shape control thresholds include a radial distance deviation threshold and a radial distance deviation change rate threshold. When the real-time cavity parameter, i.e., the radial distance deviation or the radial distance deviation change rate, exceeds either shape control threshold, it indicates that the cavity shape deviates from the design requirements, and shape correction needs to be initiated. In this embodiment, the input of the PID controller is the cavity parameter exceeding the shape control threshold, and the output is the third compensation amount of the cutter arm opening speed and the fourth compensation amount of the rotation speed. The proportional coefficient, integral coefficient, and derivative coefficient are determined by those skilled in the art based on the system response characteristics and can be obtained by combining the Ziegler-Nichols critical proportionality method with step response testing. Furthermore, the three coefficients for calculating the third and fourth compensation amounts are independent of each other. The calculated third compensation amount is summed with the current cutter arm opening speed to obtain the third opening speed, and the fourth compensation amount is summed with the current rotation speed to obtain the third rotation speed, which is then output to the cutter arm actuator 6. This correction action continues until the following termination conditions are met simultaneously: the absolute value of the radial distance deviation is less than the preset size difference threshold, and the absolute value of the radial distance deviation change rate is less than the preset smoothing threshold, indicating that the radial distance and contour smoothness of the cavity have met the design requirements, the cavity creation is determined to be complete, and the cavity creation operation is stopped; the preset size difference threshold and the preset smoothing threshold are determined by those skilled in the art when designing the cavity creation target to ensure the quality of cavity creation. By continuously monitoring the radial distance of the borehole wall and the smoothness of the cavity, obvious sharp edges or unevenness on the inner wall of the cavity are avoided, improving gas extraction efficiency and providing a smooth passage for the subsequent drilling tools to pass smoothly.

[0123] For the path planning module:

[0124] In a preferred embodiment, after the cavity is created, during the process of controlling the borehole to pass through the cavity, there may be accumulated material at the bottom of the cavity. The drill bit will naturally droop under the action of gravity. If it passes through the original borehole centerline, the drill bit is very likely to collide with the accumulated material or get stuck. Moreover, the drill bit posture will deviate due to the geometric irregularity of the cavity and the influence of the accumulated material. Therefore, it is necessary to plan a safe passage path based on the three-dimensional shape model of the cavity generated by the modeling module and guide the drill bit to pass through the cavity smoothly.

[0125] Further, generating a passage path through the cavity includes the following steps:

[0126] S401. Extract the stacking height sequence at the bottom of the cavity from the three-dimensional morphological model, and calculate the gravitational sag of the drill bit in the cavity based on the unit length weight and stiffness distribution of the drill bit.

[0127] S402. The stacking height sequence is superimposed with the gravity sag to generate a compensation offset; the centerline offset trajectory is moved upward along the vertical direction by the compensation offset to generate the crossing path.

[0128] This implementation method designs a safe crossing path based on two factors: the accumulation of material occupying the bottom space and the drill bit's tendency to sag towards the bottom. This eliminates the risk of stuck drill bit and ensures the smooth completion of integrated large-diameter directional drilling operations. Specifically, based on the above implementation method, it is known that the three-dimensional morphological model of the cavity contains radial distance information of the borehole wall, but it cannot directly distinguish between the accumulation material and the original borehole wall. This implementation method estimates the accumulation height indirectly. The specific process is as follows: at each drilling depth, the radial distance value within a certain circumferential angle range at the bottom of the cavity is extracted and compared with the target radial distance at that drilling depth. The circumferential angle range is usually 30° to the left and right, with the vertical downward reference. Since the accumulation material occupies the bottom space of the cavity, its surface will bulge inward, causing the measured radial distance value to be less than the target radial distance. Therefore, the accumulation height can be approximated as the target radial distance minus the measured radial distance value. By estimating sequentially at each drilling depth, a sequence of accumulation heights is obtained, reflecting the axial distribution characteristics of the accumulation material, including the maximum accumulation height and its corresponding borehole depth. Meanwhile, as the drill bit traverses the cavity, its two ends are supported by the normal borehole sections at the cavity inlet and outlet, while the middle section is suspended. Under the influence of gravity, the drill bit will sag downwards, and the amount of sag is non-uniformly distributed along the borehole trajectory. This embodiment uses beam bending theory to calculate the amount of sag due to gravity, based on... Figure 2It is known that the drill string assembly comprises different structural components, each with different geometric and material parameters. This results in a non-uniform distribution of bending stiffness and weight per unit length along the borehole trajectory. Therefore, this embodiment employs a segmented discretization and segment-by-segment integration method for specific calculations. First, the geometric and material parameters of each structural component are obtained, including outer diameter, inner diameter, length, density, and elastic modulus. The weight per unit length is calculated by multiplying the cross-sectional area of ​​each structural component by its density and then by the gravitational acceleration. Simultaneously, the moment of inertia of the circular cross-section is calculated using the formula for calculating the moment of inertia of the circular cross-section, and then multiplied by the elastic modulus to obtain the bending stiffness. To simplify calculations, the entire drill string assembly is simplified as a variable cross-section beam. The normal borehole sections at the cavity inlet and outlet are treated as two simple supports, with the span between the simple supports equal to the cavity length. The drill string assembly is discretized into multiple micro-segments, the length of which is set by those skilled in the art. This ensures that each component structure in the drill string assembly is divided into at least multiple segments, and each micro-segment contains only one component structure. The weight per unit length and bending stiffness of each micro-segment are the same as those of the corresponding component structure, and the gravity load is equal to the weight per unit length multiplied by the micro-segment length. Then, using the principle of numerical integration, the gravitational load on each micro-segment is multiplied by the lever arm from the center of gravity of that micro-segment to the current calculation point, and the results are accumulated to obtain the bending moment distribution. Then, using the piecewise integration method, the bending moment and bending stiffness within each micro-segment are treated as constants. According to the bending differential relationship in mechanics of materials, the change in the rotation angle of a micro-segment is equal to the bending moment on that micro-segment multiplied by the length of the micro-segment and then divided by the bending stiffness. Therefore, starting from the cavity inlet, the change in the rotation angle of each micro-segment is accumulated sequentially to obtain the rotation angle value of each point relative to the inlet. Similarly, the change in sag within each micro-segment is equal to the rotation angle at the starting point of that micro-segment multiplied by the length of the micro-segment, while the sag at the cavity outlet is approximately zero. By reverse-engineering the accumulation process, the initial rotation angle and the sag at each point are obtained, which is the final sag sequence.

[0129] In this embodiment, the sag sequence and the accumulation height sequence are added point by point to generate a compensation offset sequence. Each point on the centerline offset trajectory is moved upwards by the compensation offset distance in the vertical direction to obtain the crossing path. This ensures that the actual sag position of the drill string matches the pre-lifted path, thus safely avoiding the bottom accumulation. When the drill string crosses the cavity, the expected spatial coordinates corresponding to the crossing path are queried based on the real-time drilling depth. The positional deviation vector between the actual spatial coordinates and the expected spatial coordinates is calculated and decomposed into a normal deviation perpendicular to the crossing path and a tangential deviation tangential to the path. The normal deviation is the main error that needs to be eliminated by tool face angle adjustment. The required tool face angle adjustment and drill pressure adjustment are calculated using a PID control algorithm.

[0130] The specific calculation process for the tool face angle adjustment is as follows: Since downhole guiding tools can usually only apply lateral force in one plane through a bent screw or a rotary guide push block, it is necessary to project the three-dimensional deviation onto the direction of action of the tool face. Therefore, the normal deviation is used as the input of the PID control algorithm, and the output is the tool face angle adjustment. The proportional coefficient, integral coefficient, and derivative coefficient are determined by those skilled in the art based on the system response characteristics and can be obtained by tuning through the Ziegler-Nichols critical proportional method combined with step response test. The current tool face angle is superimposed with the tool face angle adjustment to obtain the corrected tool face angle. The guiding actuator is controlled to change the bias direction, thereby generating the required lateral force on the drill bit and gradually pulling the drill string back to the desired crossing path.

[0131] The specific calculation process of the drill pressure adjustment is as follows: Since the tangential deviation reflects the lead or lag of the actual spatial coordinates along the crossing path, and indirectly reflects the mismatch between the drill bit's travel speed and the desired speed, the PID control algorithm takes the weighted sum of the tangential deviation and the normal deviation as input and outputs the drill pressure adjustment. The proportional coefficient, integral coefficient, and derivative coefficient are determined by those skilled in the art based on the system response characteristics, and can be obtained by tuning through the Ziegler-Nichols critical proportional method combined with step response tests. The weighting coefficient of the weighted sum is set according to the lateral and vertical correction priorities, and the sum of the two is 1. The current drill pressure and the drill pressure adjustment are superimposed to obtain the corrected drill pressure, which is output to the drill pressure control actuator to change the axial thrust of the drill bit and assist in adjusting the bending shape of the drill bit.

[0132] Example 2:

[0133] like Figure 3 As shown, this embodiment provides an adaptive integrated directional hole-forming control method, including:

[0134] Real-time acquisition of drill bit attitude parameters, formation response parameters, and cutter arm actuator operating status parameters, as well as the radial distance of the hole wall of the cavity;

[0135] Based on the formation response parameters, a continuous distribution profile of formation hardness along the borehole trajectory is constructed by inversion, and a three-dimensional morphological model of the cavity is generated based on the radial distance of the borehole wall.

[0136] Based on the predicted hardness value corresponding to the preset cavity-making position in the continuous distribution profile, the corresponding cavity-making parameters are generated in combination with the preset mapping rules; when the drill bit reaches the cavity-making position, the cutter arm actuator is controlled to start the cavity-making operation according to the cavity-making parameters, and the cavity-making parameters are corrected based on the real-time formation response parameters until the radial distance and shape of the borehole wall reach the preset cavity-making target.

[0137] After the cavity is created, a path for traversing the cavity is generated based on the three-dimensional morphological model. The attitude parameters are compared with the path for traversing the cavity to generate a closed-loop deviation adjustment command, thus completing the traversal of the cavity.

[0138] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. An adaptive integrated directional cavitation control system, characterized in that, include: The data acquisition module is used to acquire in real time the attitude parameters of the drill bit, formation response parameters, and operating status parameters of the cutter arm actuator, as well as the radial distance of the hole wall of the cavity. The modeling module is used to invert and construct a continuous distribution profile of formation hardness along the borehole trajectory based on the formation response parameters, and to generate a three-dimensional morphological model of the cavity based on the radial distance of the borehole wall. The specific steps for generating the three-dimensional morphological model of the cavity include: A two-dimensional grid is constructed with the drilling depth as the first dimension and the circumferential angle as the second dimension. The radial distance of the borehole wall at each sampling point is filled into the corresponding grid cell, and the grid cells that are not sampled are marked as blind zones. Calculate the radial distance gradient change rate along the borehole depth and the radial distance curvature change rate along the circumferential angle for each sampled grid cell; if the radial distance gradient change rate or the radial distance curvature change rate exceeds a preset abrupt change threshold, mark the corresponding grid cell as the cavity abnormal boundary and use the radial distance between the borehole walls on the adjacent two sides as a discontinuity constraint condition. Using the sampled mesh cells and the discontinuity constraint as boundary conditions, the radial distance of the hole wall in the blind zone is calculated using the radial basis function interpolation method to form a complete two-dimensional mesh; Based on the complete two-dimensional mesh, the radial distance of the hole wall at all circumferential angles for each drilling depth is combined with the corresponding circumferential angle coordinates to generate a two-dimensional closed cavity cross-sectional profile. The cavity cross-sectional contours are arranged in order of drilling depth, and surface fitting and smoothing processes are performed to generate a point cloud of the cavity surface within a continuous drilling depth range. Extract the geometric center at each drilling depth from the point cloud on the cavity surface, calculate the centerline offset trajectory of the cavity, identify and mark the area where the radial distance of the hole wall exceeds the preset diameter expansion threshold as the diameter expansion area; fuse the centerline offset trajectory, the diameter expansion area and the point cloud on the cavity surface to generate a three-dimensional morphological model of the cavity; An adaptive control module is used to generate corresponding cavity-making parameters based on the predicted hardness value corresponding to the preset cavity-making position in the continuous distribution profile and in combination with preset mapping rules. When the drill bit reaches the cavity-making position, it controls the cutter arm actuator to start the cavity-making operation according to the cavity-making parameters, and corrects the cavity-making parameters based on real-time formation response parameters until the radial distance and shape of the borehole wall reach the preset cavity-making target. The specific steps for correcting the cavity-forming parameters based on real-time formation response parameters include: The target cavity-creating parameters are obtained according to the mapping rules, the cutter arm actuator is controlled to start the cavity-creating operation, and a first response model of the hydraulic system is established; the target cavity-creating parameters include at least the target cutter arm opening speed and the target rotation speed; Based on the first response model, the hole-making operation is started, and the angle lag deviation is obtained by comparing the real-time opening angle with the target angle of the target cutter arm opening speed at the corresponding time. Calculate the energy difference between the real-time mechanical specific energy and the predicted hardness value. If the energy difference exceeds the preset deviation threshold range, use the angle hysteresis deviation to correct the first response model to obtain the second response model. The target cutter arm opening speed is adjusted according to the energy difference to obtain a first opening speed; and the predicted response time of the current control cycle is determined based on the second response model, and the control command corresponding to the first opening speed is issued in advance of the predicted response time. The path planning module is used to generate a passage path through the cavity based on the three-dimensional morphological model after the cavity is created, and to compare the attitude parameters with the passage path to generate a closed-loop deviation adjustment command to complete the passage through the cavity. The specific steps for generating the passage path through the cavity include: Extract the stacking height sequence at the bottom of the cavity from the three-dimensional morphological model, and calculate the gravitational sag of the drill bit in the cavity based on the unit length weight and stiffness distribution of the drill bit. The stacking height sequence is superimposed with the gravity sag to generate a compensation offset; the centerline offset trajectory is moved upward along the vertical direction by the compensation offset to generate the crossing path.

2. The adaptive integrated directional cavitation control system as described in claim 1, characterized in that, The specific steps for constructing a continuous distribution profile of formation hardness along the borehole trajectory through inversion include: The first mechanical specific energy is calculated based on the formation response parameters at each sampling point during drilling; the formation response parameters include drilling torque, drilling pressure, and mechanical drilling speed. The first mechanical specific energy is fitted with a sliding window local polynomial along the drilling trajectory to obtain the specific energy fitting function, and the values ​​at each sampling point in the specific energy fitting function are extracted as the drill bit wear factor. The second mechanical specific energy of each sampling point is obtained by performing a difference calculation based on the first mechanical specific energy and the drill bit wear factor; a continuous distribution profile of formation hardness along the borehole trajectory is generated based on the second mechanical specific energy.

3. The adaptive integrated directional cavitation control system as described in claim 2, characterized in that, The specific steps for constructing the mapping rules include: Acquire historical hole-making operation data. Each historical hole-making operation data includes the second mechanical specific energy before hole-making, the cutter arm opening speed and rotation speed during hole-making, as well as the cutter arm wear per unit depth and mechanical drilling speed. Using the second mechanical specific energy as the input variable, the cutter arm opening speed and the rotation speed as the variables to be optimized, and the objectives of minimizing the cutter arm wear per unit feed and maximizing the mechanical drilling speed, an optimization function is constructed. Based on the second mechanical specific energy, hardness intervals are clustered and divided. The optimization function is solved in each hardness interval to obtain the optimized cavity-forming parameter set. Calculate the efficiency index of each group of cavity-creating parameters in the optimized cavity-creating parameter set, select the cavity-creating parameter corresponding to the maximum value of the efficiency index as the target cavity-creating parameter of the hardness range, and each hardness range and the corresponding target cavity-creating parameter constitute the mapping rule; the efficiency index is the weighted ratio of the mechanical drilling rate and the wear of the cutter arm per unit depth.

4. The adaptive integrated directional cavitation control system as described in claim 2, characterized in that, The specific steps of correcting the cavity-forming parameters based on real-time formation response parameters further include: When the deviation direction of the cutter arm contact force relative to the preset pressure range is opposite to the deviation direction of the real-time mechanical energy relative to the predicted hardness value, the first difference between the drilling torque and the target torque and the second difference between the cutter arm contact force and the target pressure are input into the preset decoupling control equation, and the first compensation amount of the cutter arm opening speed and the second compensation amount of the rotation speed are output. The second opening speed is obtained by summing the first compensation amount with the current blade arm opening speed, and the second rotation speed is obtained by summing the second compensation amount with the current rotation speed.

5. The adaptive integrated directional cavitation control system as described in claim 4, characterized in that, The specific steps of correcting the cavity-forming parameters based on real-time formation response parameters further include: The radial distance between the borehole wall and the radial distance at the same drilling depth in the preset cavity-making target are calculated to obtain the radial distance deviation, and the rate of change of the radial distance deviation is calculated. If the radial distance deviation or the rate of change of the radial distance deviation exceeds the preset shape control threshold, the third compensation amount of the cutter arm opening speed and the fourth compensation amount of the rotation speed are calculated by the PID control algorithm. The third opening speed is obtained by summing the third compensation amount with the current blade arm opening speed, and the third rotation speed is obtained by summing the fourth compensation amount with the current rotation speed.

6. An adaptive integrated directional cavitation control method, implemented based on an adaptive integrated directional cavitation control system according to any one of claims 1-5, characterized in that, include: Real-time acquisition of drill bit attitude parameters, formation response parameters, and cutter arm actuator operating status parameters, as well as the radial distance of the hole wall of the cavity; Based on the formation response parameters, a continuous distribution profile of formation hardness along the borehole trajectory is constructed by inversion, and a three-dimensional morphological model of the cavity is generated based on the radial distance of the borehole wall. Based on the predicted hardness value corresponding to the preset cavity-making position in the continuous distribution profile, the corresponding cavity-making parameters are generated in combination with the preset mapping rules; when the drill bit reaches the cavity-making position, the cutter arm actuator is controlled to start the cavity-making operation according to the cavity-making parameters, and the cavity-making parameters are corrected based on the real-time formation response parameters until the radial distance and shape of the borehole wall reach the preset cavity-making target. After the cavity is created, a path for traversing the cavity is generated based on the three-dimensional morphological model. The attitude parameters are compared with the path for traversing the cavity to generate a closed-loop deviation adjustment command, thus completing the traversal of the cavity.

Images