Intelligent drilling method and system for a face of a petroleum drilling tool

Abstract

The application provides an intelligent drilling guide method and system for a petroleum drilling tool face, and relates to the technical field of intelligent control. The method comprises the following steps: according to a dynamic adjustment instruction, determining a first target point, a second target point and a third target point in a process of tracking a target azimuth angle; constructing a dynamic triangle through the first target point, the second target point and the third target point, and taking the barycentric coordinates of the dynamic triangle as an azimuth angle correction value; correcting the forward and reverse angle and the period parameter of the drill string according to the azimuth angle correction value to obtain a corrected dynamic adjustment instruction; and controlling the action of the drill string based on the corrected dynamic adjustment instruction, so that the downhole tool face continuously approaches the target azimuth angle through closed-loop iterative adjustment. The application can improve the tool face control efficiency and stability under complex working conditions.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to an intelligent drilling guidance method and system for oil drilling tool faces. Background Technology

[0002] In the field of oil drilling, especially in directional and horizontal well drilling operations, precise control of the tool face azimuth is a key technology to ensure that the wellbore trajectory meets design requirements. The tool face azimuth determines the orientation of the downhole drilling tools, and its control accuracy directly affects drilling efficiency, trajectory deviation, and operational safety.

[0003] In traditional drilling operations, tool face adjustment mainly relies on manual operation. The driller manually controls the angle and cycle of the top drive oscillation based on experience. Therefore, the following problems exist:

[0004] On the one hand, manual adjustments are subject to operational delays and accuracy errors, making it difficult to respond in real time to complex downhole conditions (such as changes in formation lithology and fluctuations in frictional torque). This leads to frequent deviations of the wellbore trajectory from the target, increasing subsequent correction costs and the risk of stuck pipe. On the other hand, in conventional sliding drilling technology, the pressure drag caused by high drill string frictional torque is particularly prominent, making it difficult to effectively transmit drilling pressure to the drill bit, resulting in a significant decrease in drilling speed. Although existing top drive torsion pendulum technology can reduce friction by periodically rotating the drill string, the setting of torsion pendulum parameters (such as forward and reverse rotation angles and periods) depends on manual experience. When the parameters are unreasonable, "over-adjustment" or "insufficient drag reduction" can easily occur, affecting both directional accuracy and exacerbating mechanical load fluctuations. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an intelligent drilling guidance method and system for oil drilling tool faces, which can improve the efficiency and stability of tool face control under complex working conditions.

[0006] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0007] In a first aspect, an intelligent drilling guide method for an oil drilling tool face is provided, the method comprising:

[0008] Step S1: Acquire the azimuth data of the downhole tool face in real time and record the historical azimuth offset simultaneously;

[0009] Step S2: Process the azimuth data in real time and generate a dynamic adjustment command in combination with the preset target tool face angle. The dynamic adjustment command includes the forward and reverse rotation angles of the drill string and the period parameters.

[0010] Step S3: Based on the dynamic adjustment command generated in step S2, determine the first target point, the second target point, and the third target point during the target azimuth tracking process; construct a dynamic triangle using the first target point, the second target point, and the third target point, and use the centroid coordinates of the dynamic triangle as the azimuth correction value;

[0011] Step S4: Correct the forward and reverse rotation angles and period parameters of the drill string according to the azimuth correction value to obtain the corrected dynamic adjustment command;

[0012] Step S5: Control the drill string movement based on the corrected dynamic adjustment command, so that the downhole tool face continuously approaches the target azimuth angle through closed-loop iterative adjustment.

[0013] Secondly, an intelligent drilling guide system for oil drilling tool faces includes:

[0014] The acquisition module is used to acquire the azimuth data of the downhole tool face in real time and simultaneously record the historical azimuth offset.

[0015] The generation module is used to process the azimuth data in real time and generate dynamic adjustment instructions in combination with the preset target tool face angle. The dynamic adjustment instructions include the forward and reverse rotation angles of the drill string and the period parameters.

[0016] The correction module is used to determine the first target point, the second target point, and the third target point during the target azimuth tracking process according to the generated dynamic adjustment instructions; construct a dynamic triangle using the first target point, the second target point, and the third target point, and use the centroid coordinates of the dynamic triangle as the azimuth correction value;

[0017] The processing module is used to correct the forward and reverse rotation angles and period parameters of the drill string according to the azimuth correction value, so as to obtain the corrected dynamic adjustment command.

[0018] The control module is used to control the drill string movement based on the corrected dynamic adjustment commands, so that the downhole tool face can continuously approach the target azimuth angle through closed-loop iterative adjustment.

[0019] Thirdly, a computing device, comprising:

[0020] One or more processors;

[0021] A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to implement the method.

[0022] Fourthly, a computer-readable storage medium storing a program that, when executed by a processor, implements the method.

[0023] The above-described solution of the present invention has at least the following beneficial effects:

[0024] By acquiring and preprocessing azimuth data in real time through downhole sensor arrays, the data source input to the algorithm is ensured to be real-time, continuous, and reliable. Historical offsets are recorded simultaneously to provide a data basis for subsequent dynamic compensation, avoiding adjustment lag or deviation accumulation caused by relying solely on real-time data, and improving the system's ability to analyze the trend of tool face changes.

[0025] By using a hybrid neural network (LSTM+CNN) to extract the spatiotemporal features of azimuth data and combining them with drilling parameters to predict the dynamic response of the drill string, the limitations of traditional manual experience in setting parameters are overcome. Through dynamic adjustment commands generated by multi-objective optimization (accuracy, load, energy consumption), it can adapt to complex downhole conditions (such as formation hardness and mud resistance) and achieve a balance between control accuracy and equipment safety.

[0026] The target orientation is dynamically corrected by "offset compensation point". The proportional compensation mechanism of historical offset is used to offset the lag or overshoot caused by friction, inertia and other factors during the tool surface adjustment process in advance. The correction value based on the centroid of the triangle integrates the real-time position, target position and historical deviation, so that the adjustment direction is closer to the actual dynamic needs, significantly improving the tracking accuracy and reducing oscillation.

[0027] The drill string's forward and reverse rotation angles and cycles are dynamically adjusted based on the corrected azimuth angle. The response speed is dynamically optimized by combining historical offset frequencies and current operating conditions (e.g., shortening the cycle to cope with rapid changes and extending the cycle to reduce load). Through multi-objective constraint verification (accuracy threshold, load safety range, energy consumption limit), equipment overload or ineffective adjustments caused by unreasonable parameters are avoided, ensuring long-term stable operation of the system.

[0028] Through a closed-loop feedback mechanism of "data acquisition - algorithm processing - parameter correction - execution control", the tool face azimuth angle is fully automatically tracked, and it can respond to formation changes in real time without human intervention. The iterative adjustment process continuously reduces the deviation, ensuring that the wellbore trajectory meets the design requirements with high precision. At the same time, it reduces the safety risks such as stuck drill and wellbore collapse caused by delays or lack of experience in manual operation, which can improve drilling efficiency and operational safety. Attached Figure Description

[0029] Figure 1 This is a schematic flowchart of an intelligent drilling guide method for oil drilling tool faces provided by an embodiment of the present invention.

[0030] Figure 2 This is a schematic diagram of an intelligent drilling guide system for an oil drilling tool face provided in an embodiment of the present invention. Detailed Implementation

[0031] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0032] like Figure 1 As shown, an embodiment of the present invention proposes an intelligent drilling guidance method for oil drilling tool faces, the method comprising the following steps:

[0033] Step S1: Acquire the azimuth data of the downhole tool face in real time and record the historical azimuth offset simultaneously;

[0034] Step S2: Process the azimuth data in real time and generate a dynamic adjustment command in combination with the preset target tool face angle. The dynamic adjustment command includes the forward and reverse rotation angles of the drill string and the period parameters.

[0035] Step S3: Based on the dynamic adjustment command generated in step S2, determine the first target point, the second target point, and the third target point during the target azimuth tracking process; construct a dynamic triangle using the first target point, the second target point, and the third target point, and use the centroid coordinates of the dynamic triangle as the azimuth correction value;

[0036] Step S4: Correct the forward and reverse rotation angles and period parameters of the drill string according to the azimuth correction value to obtain the corrected dynamic adjustment command;

[0037] Step S5: Control the drill string movement based on the corrected dynamic adjustment command, so that the downhole tool face continuously approaches the target azimuth angle through closed-loop iterative adjustment.

[0038] In this embodiment of the invention, tool face azimuth data is acquired in real time by a downhole sensor array and preprocessed to ensure that the data source input to the algorithm has high precision and low latency. Historical azimuth offsets are recorded synchronously to establish a time-series database of tool face changes, providing historical deviation data for offset compensation point calculation in step S3. This avoids blind adjustments caused by relying solely on real-time data and improves the system's ability to learn tool face fluctuation patterns. Based on a hybrid architecture of Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN), the system deeply mines the time-series features (such as periodic fluctuations) and spatial correlation features (such as the impact of formation changes on the tool face) of the azimuth data, breaking through the limitations of traditional manual experience-based parameter setting and achieving adaptive generation of adjustment commands. By combining multi-objective optimization such as azimuth tracking accuracy, drill string mechanical load balance, and energy consumption threshold, the generated dynamic adjustment commands (forward and reverse rotation angles, periodic parameters) can balance control accuracy and equipment reliability in complex downhole conditions (such as high-friction formations and mud resistance changes), avoiding overload or energy waste caused by single-parameter optimization. By introducing "offset compensation points" based on historical offsets, proportional compensation is performed on the reverse extension lines of the real-time position point and the theoretical target point. This proactively corrects adjustment lags or overshoots caused by factors such as drill string inertia and downhole friction, making the tool face adjustment direction closer to actual dynamic needs. A dynamic triangle is constructed by the real-time position (first target point), the theoretical target (second target point), and the historical offset compensation (third target point), using the centroid coordinates as correction values. This integrates the current state, target orientation, and historical experience, effectively suppressing oscillations during tool face tracking and improving the stability and accuracy of trajectory control.

[0039] The drill string's forward and reverse rotation angles and cycle parameters are dynamically adjusted based on the azimuth correction value. For example, the response cycle is adaptively adjusted based on the historical offset fluctuation frequency (shortening the cycle during rapid changes and lengthening the cycle during stable periods) to avoid efficiency losses or control delays caused by "one-size-fits-all" adjustments. The corrected parameters are verified through a multi-objective optimization model to ensure that the azimuth tracking accuracy meets the standards, the drill string load is within a safe range, and the energy consumption does not exceed the equipment's rated power. This avoids mechanical failures (such as torque overload) or ineffective adjustments caused by unreasonable parameters at the algorithm level, improving the long-term stability of the system. Through a closed-loop feedback mechanism of "data acquisition - intelligent processing - execution correction," the tool face azimuth angle can be automatically tracked without human intervention, responding in real time to formation changes (such as frictional changes caused by lithological abrupt changes), solving the delay problem of traditional manual adjustments, and significantly shortening the tool face correction time. The iterative adjustment process continuously reduces the deviation between the actual tool face and the target angle, ensuring that the wellbore trajectory strictly meets the design requirements; at the same time, it avoids operational errors caused by insufficient human experience (such as excessive twisting causing stuck drill bit), reduces safety risks such as wellbore collapse and equipment damage, and significantly improves drilling efficiency and operational safety under complex working conditions.

[0040] In a preferred embodiment of the present invention, step S1: real-time acquisition of azimuth data of the downhole tool face and synchronous recording of historical azimuth offset, including:

[0041] The tool face azimuth data is collected in real time by downhole sensor array, and the azimuth data is preprocessed to generate a preprocessed azimuth data stream.

[0042] Historical azimuth data is stored for a preset time window length, and the historical azimuth offset is calculated.

[0043] In this embodiment of the invention, a three-axis gyroscope, accelerometer, and magnetic compass are integrated into the downhole measurement while drilling (MWD) tool to construct a multi-dimensional sensor array. Raw tool face azimuth data (unit: degrees, accuracy ±0.1°) is acquired in real time at a frequency of at least 10Hz. A moving average filtering algorithm (window length 50ms) is used to remove high-frequency electromagnetic interference and mechanical vibration noise, preserving the true azimuth change trend. Threshold detection (e.g., deviation > 5° for three consecutive sampling points is considered abnormal) is used, and linear interpolation of data from different time points is employed to repair abnormal values. The raw attitude data (Eulerian angles) output by the sensors is converted into tool face azimuth in the geographic coordinate system to eliminate coordinate system deviation caused by drill string rotation. The preprocessed azimuth data is sorted by timestamp to generate a continuous azimuth data stream (format: timestamp-azimuth value), which serves as the input data source for subsequent algorithms.

[0044] In this embodiment of the invention, a circular buffer is established in the memory of the control system to store historical azimuth angle data with a preset time window length (e.g., 300 seconds). When new data is written, the oldest outdated data is automatically overwritten to ensure that the historical records of the most recent 5 minutes are always retained. The difference between the current azimuth angle and the preset target angle (△θ = current value - target value) is calculated, and the absolute value |△θ| is taken as the real-time offset amplitude.

[0045] For all data within the time window, calculate the change in azimuth angle (Δθ) between each time point and the previous time point. t =θ t -θ t-1 ), forming a historical offset sequence {△θ1, △θ2, ..., △θ n The tool surface fluctuation is used to analyze the frequency and amplitude characteristics of the fluctuation; the mean (reflecting the long-term trend), standard deviation (reflecting the fluctuation amplitude), and maximum offset (reflecting the extreme deviation) of the historical offset are calculated in real time, which serve as the basis for calculating the offset compensation point in step S3.

[0046] In this embodiment of the invention, multi-sensor fusion and multi-level preprocessing (filtering, outlier repair, coordinate system transformation) effectively suppress the impact of complex downhole environments (high temperature, high pressure, strong electromagnetic interference) on data acquisition, ensuring that the azimuth data error of the input algorithm is ≤0.5°, providing a reliable data source for precise control. Real-time generation of continuous data streams and the use of a circular buffer to manage historical data avoid information gaps caused by traditional discrete sampling, enabling the system to capture high-frequency fluctuations on the tool face (such as rapid angle changes caused by drill string vortex).

[0047] In a preferred embodiment of the present invention, step S2: the azimuth data is processed in real time, and a dynamic adjustment command is generated by combining it with a preset target tool face angle. The dynamic adjustment command includes the forward and reverse rotation angles of the drill string and period parameters, including:

[0048] The real-time azimuth data stream, historical azimuth offset, and preset target angle output in step S1 are input into a pre-trained artificial intelligence model. The artificial intelligence model is a hybrid architecture based on long short-term memory network and convolutional neural network, used to learn the mapping relationship between dynamic changes in azimuth and drill string control parameters.

[0049] The spatiotemporal features of the azimuth data are extracted by the artificial intelligence model and combined with the current drilling parameters to predict the real-time dynamic response of the drill string, so as to obtain the prediction result.

[0050] An initial dynamic adjustment command is generated based on the prediction results. The initial dynamic adjustment command includes the start angle threshold for forward and reverse rotation of the drill string, the rotation angle increment, and the rotation cycle parameter. The start angle threshold for forward and reverse rotation of the drill string is dynamically corrected by the amplitude and direction of the historical offset.

[0051] The initial dynamic adjustment command and the target angle constraints are optimized in a multi-objective manner. The optimization conditions include azimuth tracking accuracy, drill string mechanical load balance and energy consumption threshold of adjustment action. Finally, a feasible solution that satisfies the constraints is generated as the dynamic adjustment command.

[0052] In this embodiment of the invention, the real-time azimuth data stream, historical azimuth offset, and preset target angle output in step S1 are input into a pre-trained artificial intelligence model. This model is based on a hybrid architecture of Long Short-Term Memory Network (LSTM) and Convolutional Neural Network (CNN). The construction and training process of this model is as follows:

[0053] A large amount of historical drilling data, including azimuth data, drill string control parameters, and drilling condition information, was collected, cleaned, normalized, and divided into training, validation, and test sets.

[0054] Model building:

[0055] CNN layer: Convolutional operations are performed on the real-time azimuth data stream using convolutional kernels of different sizes (e.g., 3, 5, 7). Each convolutional layer contains 32 convolutional kernels. After convolution, a max pooling layer (window size 2, stride 2) and a ReLU activation function are added to extract local spatial features of the azimuth data.

[0056] LSTM layer: The feature sequence output from the CNN layer is input into an LSTM layer containing 64 LSTM units to learn the dynamic change pattern of azimuth data over time.

[0057] Fully connected layer: The output of the LSTM layer is concatenated with the current drilling parameters and then input into the fully connected layer (containing 128 neurons). A Dropout layer is added (Dropout rate 0.2). Finally, the real-time dynamic response prediction result of the drill string is output through the output layer.

[0058] Using the mean squared error loss function and the Adam optimizer (initial learning rate 0.001), the training set data is input into the model in batches (64 samples per batch) for training. The learning rate is adjusted according to the performance of the validation set. An early stopping strategy is adopted to prevent overfitting. The model performance is evaluated using the test set, and indicators such as root mean square error and mean absolute error are calculated. If the performance is not ideal, the model is further optimized by adjusting the hyperparameters. The model obtained through the above construction and training process can learn the mapping relationship between the dynamic changes of azimuth angle and the drill string control parameters.

[0059] By combining current drilling parameters, such as drilling pressure, rotation speed, and mud discharge, as additional input information, the extracted spatiotemporal features are fused with the drilling parameters through a fully connected layer to predict the real-time dynamic response of the drill string under different control parameters, such as torque changes and displacement, and thus obtain the prediction results.

[0060] Based on the prediction results, the initial values ​​of the starting angle threshold, rotation angle increment, and rotation cycle parameters for forward and reverse drill string rotation are determined. For example, when a large azimuth deviation is predicted, the rotation angle increment and starting angle threshold are appropriately increased, and the starting angle threshold is dynamically corrected based on the amplitude and direction of historical offsets. If the amplitude of historical offsets is large and the direction is stable, the starting angle threshold is appropriately increased to reduce unnecessary adjustments; if historical offsets fluctuate significantly, the starting angle threshold is decreased to improve the system's response speed.

[0061] By extracting the spatiotemporal features of azimuth data through deep learning models, the dynamic variation patterns of azimuth can be captured more accurately, thereby generating more precise dynamic adjustment commands and improving the control accuracy of the tool face azimuth. Combined with current drilling parameters, the real-time dynamic response of the drill string is predicted, enabling the system to adaptively adjust the control parameters of the drill string according to different drilling conditions. This enhances the system's adaptability and flexibility. The multi-objective optimization process considers the mechanical load balance of the drill string, avoiding excessive torque and stress during adjustment, reducing equipment wear and damage, and extending equipment lifespan. Dynamically correcting the start-up angle threshold reduces unnecessary adjustment actions, decreasing the frequency of equipment start-ups and shutdowns, further protecting the equipment. Incorporating energy consumption thresholds for adjustment actions into the multi-objective optimization optimizes adjustment commands, ensuring the drill string reaches the target azimuth while minimizing energy consumption, thus reducing drilling costs. The automatic generation of dynamic adjustment commands based on artificial intelligence algorithms reduces reliance on human experience, lowers the impact of human factors on the drilling process, and improves the automation level and reliability of drilling operations.

[0062] In a preferred embodiment of the present invention, the initial dynamic adjustment command and target angle constraints are subjected to multi-objective optimization. The optimization conditions include azimuth tracking accuracy, drill string mechanical load balance, and energy consumption threshold of the adjustment action. Finally, a feasible solution satisfying the constraints is generated as the dynamic adjustment command, including:

[0063] Based on the forward and reverse angle thresholds, rotation angle increments, and period parameters in the initial dynamic adjustment command, a multi-objective evaluation function is constructed, which includes azimuth tracking accuracy error, drill string axial torque fluctuation value, and single adjustment energy consumption. The weight coefficients of each objective are dynamically allocated according to the real-time downhole working conditions.

[0064] The multi-objective evaluation function is iteratively solved using a constrained particle swarm optimization algorithm. The constraints include: the azimuth tracking accuracy error does not exceed a preset threshold, the mechanical load fluctuation of the drill string is within a safe range, and the energy consumption of a single adjustment is lower than the rated power of the equipment.

[0065] During the iteration process, the candidate solution set with the smallest azimuth tracking accuracy error is retained first, and the mechanical load fluctuation value and the final command parameter combination of the comprehensive energy consumption score are selected from the candidate solution set to generate the preliminary optimization command;

[0066] The initial optimization command is verified by downhole dynamic response simulation. The tool face azimuth change trajectory after the initial optimization command is executed is simulated by a preset drill string dynamics model. If the azimuth converges to the target angle in the simulation results and meets the load and energy consumption constraints, it is determined to be a feasible solution.

[0067] If the simulation verification fails, the weight coefficients of the multi-objective evaluation function are adjusted according to the error direction, and new optimization instructions are generated iteratively until a dynamic adjustment instruction that satisfies all constraints is output, thus obtaining the final optimized dynamic adjustment instruction.

[0068] In this embodiment of the invention, within the feasible solution space, a set of particles is randomly generated according to the range of variables in the problem. The position vector of each particle consists of the forward and reverse angle thresholds, rotation angle increments, and period parameters in the initial dynamic adjustment command. For example, assuming the forward and reverse angle threshold range is [-180°, 180°], the rotation angle increment range is [-10°, 10°], and the period parameter range is [0.5, 5] seconds, then the position vector of each particle consists of values ​​randomly generated within these ranges. By randomly generating particles, a wide search can be performed throughout the feasible solution space, providing diverse initial solutions for the subsequent optimization process and increasing the possibility of finding the global optimum.

[0069] In this embodiment of the invention, the position vector of each particle is substituted into the multi-objective evaluation function for calculation. The multi-objective evaluation function includes objectives such as azimuth tracking accuracy error, drill string axial torque fluctuation value, and single adjustment energy consumption. For example, for a particle, the corresponding azimuth tracking accuracy error (the difference between the current azimuth and the target azimuth) is first calculated based on its position vector. Then, the drill string axial torque fluctuation value is calculated based on the drill string dynamics model, and the single adjustment energy consumption is calculated based on the adjustment action parameters. Then, based on the weight coefficients of each objective (dynamically allocated according to the real-time downhole conditions), these objective values ​​are weighted and summed to obtain the fitness value of the particle. The fitness value, as an indicator of particle quality, can comprehensively reflect the performance of each particle in multi-objective optimization. By calculating the fitness value, particles can be compared and selected.

[0070] In this embodiment of the invention, the process of constructing the drill string dynamics model is as follows:

[0071] Collect basic parameters of the drill string: drill string length, diameter, material elastic modulus, density, joint friction coefficient, etc.

[0072] Real-time operating data: current drill pressure, rotation speed, mud density, formation friction coefficient (obtained through historical data or measurement while drilling), current azimuth and target angle of the tool face;

[0073] Environmental parameters: downhole temperature, pressure (affecting material mechanical properties), mud viscosity (affecting fluid resistance).

[0074] Based on the lumped mass method, the drill string is discretized into multiple units, each of which considers the balance of axial force, torque, and bending moment. When the drill string contacts the wellbore, the tangential friction force and normal force are calculated based on the friction coefficient and contact angle to simulate the constraint effect of the formation on the drill string. Inertial force and damping force are added to consider the centrifugal force, Coriolis force, and viscous damping effect generated by the mud flow when the drill string rotates.

[0075] After inputting the adjustment command parameters (forward and reverse rotation angle, period), the model calculates the torque change curve when the drill string rotates: based on the rotation angle increment and period, combined with the formation friction, the torque required for each rotation step and the torque fluctuation range are predicted. By integrating the drill string rotation angle, combined with the inertial navigation principle (gyroscope and accelerometer data), the change trend of the tool face angle over time under different adjustment parameters is calculated.

[0076] Key output indicators: peak torque fluctuation (reflecting mechanical load), azimuth convergence speed (reflecting tracking accuracy), and energy consumption (calculated based on motor power and action time).

[0077] The model is trained using historical drilling data (such as actual torque and azimuth changes under known adjustment parameters) to correct unknown parameters such as friction coefficient and damping coefficient; error thresholds are set (such as torque simulation error ≤5%, azimuth trajectory error ≤1°) to ensure that the model prediction accuracy meets engineering requirements.

[0078] In practical applications, the following examples can be used to achieve this:

[0079] Drill string basic parameters and discretization, basic parameter collection:

[0080] Drill string length , outer diameter , inner diameter Material density Elastic modulus , Joint friction coefficient μ;

[0081] Real-time operating conditions: drilling pressure P (N), rotational speed n ( It needs to be converted to angular velocity. , ), mud density (kg / m 3 ), formation friction coefficient ;

[0082] Environmental parameters: downhole temperature T (°C), pressure p (Pa), mud viscosity .

[0083] Lumped mass discretization:

[0084] The drill string is divided into Unit 1, the 1st Units:

[0085] Unit length: ;

[0086] Unit mass: ,in Cross-sectional area (m 2 );

[0087] Moment of inertia: ,in, This indicates the outer diameter of the unit (m). If the drill string has a constant diameter, then... ; This indicates the inner diameter of the unit; if the drill string has a constant diameter, then... .

[0088] Axial force at the top of each unit: , This represents the axial force at the top of element i, with an initial value. (Drilling pressure) The axial distributed load (N / m) of element i includes the drill string's own weight and mud buoyancy.

[0089] Torque balance:

[0090] Considering formation friction and internal torque transmission: ,in, Indicates the torque (N·m) at the top of unit i; This represents the distributed torque (N·m / m) within the unit, caused by the rotational resistance of the drill string; This represents the frictional torque between the unit and the well wall.

[0091] Bending moment balance:

[0092] Consider lateral loads (such as wellbore contact forces): ,in, Indicates the bending moment (N·m) at the top of unit i; This represents the lateral contact force of element i.

[0093] Normal contact force:

[0094] When the drill string contacts the wellbore: ,in, The normal force (N) is represented by ; k represents the formation stiffness coefficient (N / m), which is determined by the formation lithology. The radial displacement (m) of unit i represents the amount of compression of the gap between the drill string and the well wall.

[0095] Tangential friction ,in, This represents the formation friction coefficient (dimensionless). This represents the tangential velocity of the element (m / s). The sign function determines that the friction direction is opposite to the motion direction.

[0096] Centrifugal force (rotational inertial force) ,in, This represents the distance (m) from the unit centroid to the drill string rotation axis, under bending conditions. ; This indicates the rotational angular velocity.

[0097] Coriolis force (radial inertial force) ,in, This indicates the radial velocity of the unit (m / s).

[0098] Mud viscosity damping force ,in, The damping coefficient (N·s / m) is positively correlated with mud viscosity and unit surface area. This represents the velocity of the unit relative to the mud (m / s).

[0099] Adjust command response calculation:

[0100] Based on the rotation angle increment and cycle : ,in, The base torque (N·m) is determined by drilling pressure and frictional resistance. This represents the torque fluctuation amplitude (N·m), which is related to formation heterogeneity. This represents the phase angle, initialized to 0.

[0101] By integrating the rotational angular velocity and combining it with inertial navigation data: ;in, Indicates the initial azimuth angle; This represents the wellbore inclination angle, which is the angle between the wellbore axis and the plumb line, and is the effective component affecting the change in azimuth angle. ω(t) is the tool face azimuth angle at time t, representing the angle of the tool face relative to the initial reference direction when the drill string rotates; ω(t) is the angular velocity of the drill string rotation, representing the rotational speed of the drill string around its own axis (which varies with time t).

[0102] single adjustment of energy consumption ,in, Instantaneous power (W) is calculated by multiplying torque and angular velocity; where, It represents the energy consumption (joules) of a single adjustment process, reflecting the total energy consumption of the motor or power system when the drill string rotates to adjust the tool face. It is used to evaluate the energy efficiency of drilling operations and optimize adjustment parameters (such as cycle and angle) to reduce energy consumption. The torque is expressed in N·m; t0 and t1 represent the start and end times of the adjustment process, respectively, corresponding to the cycle in the adjustment command. .

[0103] Model training and calibration:

[0104] Parameter correction (taking friction coefficient as an example)

[0105] Using historical measured torque With simulated torque That is, theoretical torque:

[0106] ;

[0107] in, Indicates the correction factor (0 < <1), avoid over-adjustment; error threshold setting:

[0108] Torque simulation error: ;

[0109] Azimuth trajectory error: .

[0110] in, This represents the corrected formation friction coefficient (dimensionless), used to update the friction calculation when the drill string contacts the wellbore / formation in the model; This represents the formation friction coefficient before correction (initial value or value from the previous iteration). The initial value can be set using empirical values ​​of formation lithology (e.g., 0.3 for sandstone, 0.5 for shale), or initialized based on historical data from similar working conditions. =0.1 indicates that the correction amount is 10% of the deviation ratio each time, gradually approaching the true value. If the value is too small, it will lead to slow convergence, and if it is too large, it may cause parameter oscillation (e.g., exceeding 1 may cause the friction coefficient to change in the opposite direction). Indicates the measured azimuth angle of the tool face; This represents the simulated azimuth angle, i.e., the theoretical azimuth angle trajectory; the threshold of 1° represents the maximum allowable angular deviation in directional drilling.

[0111] Based on the particle's fitness value and historical best position, the particle swarm optimization algorithm's update formula is used to update the particle's position and velocity. Specifically, the velocity update formula for each particle considers the particle's current velocity, the difference between its historical best position and current position, and the difference between the global best position and current position. Then, the particle's position is updated based on the updated velocity. For example, if a particle's historical best position has a better fitness value in a certain direction, it will tend to move in that direction when updating its velocity, thus guiding the particle to search for a better solution space. By continuously updating the particle's position and velocity, the particle can gradually move in the solution space, tending towards a better solution. This swarm intelligence-based search method can fully utilize information sharing and cooperation among particles, accelerating convergence speed and improving optimization efficiency.

[0112] The process checks whether the position vector of each particle meets constraints such as azimuth tracking accuracy error not exceeding a preset threshold, drill string mechanical load fluctuation within a safe range, and single adjustment energy consumption lower than the equipment's rated power. For example, if a particle's azimuth tracking accuracy error exceeds the preset threshold, or the drill string mechanical load fluctuation exceeds the safe range, or the single adjustment energy consumption exceeds the equipment's rated power, then that particle does not meet the constraints. This ensures that the solutions generated during the optimization process are all within the practically feasible range, avoiding infeasible solutions that could lead to system failure or performance degradation. By checking the constraints, the reliability and safety of the system can be guaranteed.

[0113] For particles that do not meet the constraints, certain correction strategies are employed to make them satisfy the constraints. For example, if the azimuth tracking accuracy error of a particle is too large, the error can be reduced by adjusting its forward and reverse angle thresholds or rotation angle increments; if the mechanical load fluctuation of the drill string exceeds the safe range, the period parameter or other relevant parameters can be adjusted to reduce load fluctuation; if the energy consumption of a single adjustment is too high, the parameters of the adjustment action can be optimized to reduce energy consumption. Specific correction methods can be designed according to the characteristics of the problem and the actual situation to ensure that all particles in the particle swarm satisfy the constraints, so that the optimization process can be carried out within the feasible solution space, thereby improving the feasibility and practicality of the optimization results.

[0114] The process is repeated until a preset termination condition is met. This termination condition can be reaching the maximum number of iterations, the fitness value converging to a certain accuracy, or finding an optimal solution that meets specific requirements. For example, when the number of iterations reaches a set maximum, or when the fitness value of the optimal solution changes less than a certain threshold in multiple consecutive iterations, the optimization process is considered converged, and iteration stops. Through continuous iteration, the particle swarm can continuously optimize, gradually approaching the global optimum or a solution that meets specific requirements. After meeting the termination condition, the obtained optimization result can achieve a good balance among multiple objectives, improving the overall performance of the system. After each iteration, a set of candidate solutions with the smallest azimuth tracking accuracy error is selected from all particles. This can be achieved by comparing the azimuth tracking accuracy error value of each particle, retaining the particle with the smallest error value and its corresponding solution to form a candidate solution set. It is crucial to ensure that the important objective of azimuth tracking accuracy is always considered during the optimization process, prioritizing the retention of solutions that perform best in this objective.

[0115] For the retained candidate solution set, each candidate solution is comprehensively scored based on the drill string mechanical load fluctuation value and energy consumption. The comprehensive score can be obtained by weighted summation, which assigns corresponding weights according to the importance of mechanical load fluctuation value and energy consumption, and adds the scores of the two to obtain the comprehensive score. Then, the candidate solution with the highest comprehensive score is selected as the initial optimization instruction. Under the premise of ensuring azimuth tracking accuracy, the mechanical load fluctuation value and energy consumption are comprehensively considered so that the generated initial optimization instruction can achieve good performance on multiple objectives, thereby improving the overall efficiency and economy of the system. The embodiments of the present invention can realize multi-objective optimization based on particle swarm optimization algorithm, improve the system performance in terms of azimuth tracking accuracy, drill string mechanical load balance and adjustment action energy consumption threshold, while ensuring that the solution results meet the constraints, thereby improving the reliability and safety of the system.

[0116] The specific process for determining the above comprehensive score is as follows:

[0117] Collect the peak torque fluctuation values ​​corresponding to all candidate solutions in the current iteration, determine the maximum and minimum values, and normalize the mechanical load fluctuation values ​​through minimum-maximum normalization to obtain the normalized load values;

[0118] Statistically analyze the energy consumption values ​​of all candidate solutions, determine the upper limit (energy consumption corresponding to the rated power of the equipment) and the lower limit (theoretical minimum energy consumption), and normalize the energy consumption values ​​through minimum-maximum normalization to obtain the normalized energy consumption values;

[0119] If the formation friction suddenly increases (judged by sudden changes in mud pressure or torque sensor data), increase the weight of mechanical load (e.g., mechanical load fluctuation value = 0.6, single adjustment energy consumption = 0.4) to avoid overloading. If the equipment has low power or needs to operate for a long time, increase the weight of energy consumption (e.g., mechanical load fluctuation value = 0.4, single adjustment energy consumption = 0.6) to reduce energy consumption. Under normal operating conditions, use a balanced weight (mechanical load fluctuation value = single adjustment energy consumption = 0.5). For each candidate solution, calculate the normalized load value and the normalized energy consumption value. Obtain a comprehensive score by weighted summation based on the normalized load value and the normalized energy consumption value. Select the candidate solution with the highest score as the initial optimization instruction to ensure a balance between load and energy consumption.

[0120] In this embodiment of the invention, multiphysics coupling modeling is used to accurately simulate the dynamic response of the drill string under different adjustment parameters, predicting torque fluctuations and azimuth changes in advance, thus avoiding overshoot or lag problems in actual execution. Model parameters are calibrated based on historical data to adapt to complex downhole environments (such as high friction and abrupt lithological changes), improving the engineering reliability of the simulation results. Normalization eliminates the influence of different dimensional indicators, making load and energy consumption comparable. A dynamic weight allocation strategy, combined with real-time operating conditions, flexibly balances equipment protection and energy efficiency optimization. Through downhole dynamic response simulation verification, the feasibility and effectiveness of preliminary optimization commands can be assessed in advance, avoiding unforeseen problems in practical applications and improving the reliability and stability of the system.

[0121] In a preferred embodiment of the present invention, based on the forward and reverse rotation angle thresholds, rotation angle increments, and period parameters in the initial dynamic adjustment command, a multi-objective evaluation function is constructed, including azimuth tracking accuracy error, drill string axial torque fluctuation value, and single adjustment energy consumption. The weight coefficients of each objective are dynamically allocated according to the real-time downhole operating conditions, including:

[0122] Real-time acquisition of downhole operating parameters, including drilling pressure, drill string rotation speed, formation lithology hardness coefficient, and mud flow resistance, serves as the basis for weight allocation;

[0123] The initial weighting ratios of azimuth tracking accuracy error, torque fluctuation, and energy consumption in the multi-objective evaluation function are calculated based on the aforementioned operating parameters.

[0124] The initial weight ratio is dynamically corrected by a pre-trained fuzzy logic rule base to obtain the corrected weight coefficient. The rule base is constructed based on the mapping relationship between working condition parameters and weight adjustment in historical drilling data.

[0125] The corrected weighting coefficients are integrated with the azimuth tracking accuracy error, drill string axial torque fluctuation value, and single adjustment energy consumption to form weighted parameter terms corresponding to each objective. A multi-objective evaluation function matching the current downhole conditions is constructed by combining the weighted parameter terms.

[0126] In this embodiment of the invention, sensors are used to collect parameters such as drill pressure, drill string rotation speed, formation lithology hardness coefficient, and mud flow resistance in real time. For example, drill pressure is measured by a pressure sensor, drill string rotation speed is obtained by a rotation speed sensor, formation lithology hardness coefficient can be determined based on geological exploration data or relevant information from measurements while drilling, and mud flow resistance can be calculated from mud performance parameters and wellbore conditions.

[0127] It provides an accurate real-time data foundation for weight allocation, enabling weights to be dynamically adjusted according to actual downhole conditions, thereby more accurately reflecting the importance of each target under the current conditions.

[0128] Based on the collected operating parameters, the initial weighting ratios of azimuth tracking accuracy error, torque fluctuation, and energy consumption are calculated using a pre-set algorithm. For example, the weight of torque fluctuation can be determined based on the formation lithology hardness coefficient and mud flow resistance; when the formation lithology hardness is high or the mud flow resistance is high, the weight of torque fluctuation is appropriately increased. The weight of azimuth tracking accuracy error can be determined based on drill pressure and drill string speed; when the drill pressure is high or the drill string speed is high, the weight of azimuth tracking accuracy error is appropriately increased. The weight of energy consumption is determined based on the relationship between energy consumption and other parameters, as well as the energy consumption requirements of the operation.

[0129] The weight ratios of each objective were initially determined based on the operating parameters, taking into account the degree of influence of different operating conditions on each objective, thus providing a reasonable starting point for subsequent dynamic corrections.

[0130] The calculated initial weight ratios are input into a pre-trained fuzzy logic rule base. The rule base dynamically adjusts the initial weight ratios based on the mapping relationship between operating parameters and weight adjustments in historical drilling data. For example, if historical data indicates that under a specific combination of formation lithology hardness coefficient and mud flow resistance, the weight of torque fluctuations should be increased by a certain percentage, then the rule base will adjust the initial weights of torque fluctuations accordingly based on the current operating parameters. In this way, the corrected weight coefficients are obtained.

[0131] By dynamically adjusting the initial weights using historical data and a fuzzy logic rule base, we can more flexibly adapt to different downhole working conditions, further improve the accuracy and rationality of weight allocation, and make the multi-objective evaluation function more in line with the actual situation.

[0132] The corrected weighting coefficients are then integrated with the azimuth tracking accuracy error, drill string axial torque fluctuation, and single adjustment energy consumption. For example, multiplying the azimuth tracking accuracy error by its corresponding weighting coefficient yields a weighted parameter for the azimuth tracking accuracy error; multiplying the drill string axial torque fluctuation by its weighting coefficient yields a weighted parameter for the torque fluctuation; and multiplying the single adjustment energy consumption by its corresponding weighting coefficient yields a weighted parameter for the energy consumption. These weighted parameter terms are then combined to form a multi-objective evaluation function that matches the current downhole conditions. For example, a weighted summation method can be used to combine these weighted parameter terms to obtain the final multi-objective evaluation function.

[0133] By integrating the modified weighting coefficients with each objective, the constructed multi-objective evaluation function can comprehensively consider the real-time downhole conditions and accurately evaluate the overall performance of different adjustment commands on multiple objectives such as azimuth tracking accuracy, drill string axial torque fluctuation value, and single adjustment energy consumption.

[0134] In another preferred embodiment of the present invention, the above content can be implemented in the following way when applied in a specific application:

[0135] Sensor data acquisition model

[0136] Real-time acquisition of operating condition parameter vectors via downhole sensors: ;

[0137] in, It represents the drilling pressure (N), which is measured by a pressure sensor and reflects the force exerted by the drill bit on the formation; The drill string rotation speed (rad / s) is obtained from a speed sensor and affects torque fluctuation and azimuth rate of change. It represents the lithological hardness coefficient of the formation (dimensionless), which is calibrated by logging while drilling (LWD) data or historical geological exploration data, and characterizes the drilling resistance of the formation; The resistance to mud flow (N) is represented by the mud density. viscosity And calculation of wellbore geometric parameters: ;

[0138] Where L is the well depth, The mud flow rate is... This is the ratio of the wellbore radius to the drill string radius.

[0139] Initial weight ratio calculation: Based on the weight allocation model for the working conditions, a nonlinear mapping function is used to calculate the initial weight vector.

[0140] in: ;in:

[0141] ;

[0142] in: The sigmoid function maps the input to the interval (0, 1); , , and These are calibration coefficients (determined through regression analysis of historical data). For drilling pressure; This serves as a reference value for drilling pressure. This refers to the drill string rotation speed; This serves as a reference value for the drill string rotation speed; is the formation hardness coefficient, dimensionless; This serves as a reference value for the formation hardness coefficient. This represents the resistance to mud flow. These are reference values ​​for the resistance to mud flow, and these reference values ​​are used to normalize the chemical condition parameters. , and The initial weights are azimuth accuracy, torque fluctuation, and energy consumption, respectively, and the sum of the three is 1.

[0143] Fuzzy logic dynamic correction, fuzzy rule base and weight correction model:

[0144] The following is a textual description of the specific calculation process of the fuzzy logic rule base and weight correction model:

[0145] Fuzzification process (input variable processing):

[0146] For each input variable (such as drilling pressure, drill string speed, formation lithology hardness coefficient, mud flow resistance), it is converted into a membership value in a fuzzy set. This means determining the degree of membership of each input variable in different fuzzy subsets (such as low, medium, and high).

[0147] Taking the formation lithology hardness coefficient as an example, three fuzzy subsets are defined: low, medium, and high. The membership degree of the formation lithology hardness coefficient in each subset is calculated by a specific function (such as a Gaussian membership function). This function determines the membership degree based on the input value and pre-set parameters (such as mean and standard deviation). For example, when the input formation lithology hardness coefficient is close to the mean of a certain fuzzy subset (such as "high"), its membership degree in that subset will be higher.

[0148] Fuzzy rule base construction:

[0149] A fuzzy rule base consists of a series of rules, each with a precondition and a corresponding conclusion. The precondition is a combined description of a fuzzy subset of the input variables, while the conclusion is a fuzzy subset description of each output variable (i.e., the azimuth tracking accuracy error weight correction, the drill string axial torque fluctuation value weight correction, and the single adjustment energy consumption weight correction).

[0150] For example, a rule might be: if the formation lithology hardness coefficient is "high" and the mud flow resistance is "high", then the weight correction for the drill string axial torque fluctuation value is "increase by 0.15", the weight correction for the azimuth tracking accuracy error is "decrease by 0.05", and the weight correction for single adjustment energy consumption is "decrease by 0.10". Here, "high", "increase by 0.15", etc. are all fuzzy concepts, and the specific numerical range is determined by the membership function defined above.

[0151] Fuzzy reasoning process:

[0152] For each rule, its activation degree must first be calculated. The activation degree represents the degree of matching between the current input operating condition parameters and the preconditions of the rule. The calculation method is to take the minimum value of the membership degree of each input variable in the preconditions in the corresponding fuzzy subset.

[0153] For example, for the rule mentioned above, it is necessary to find the membership degree of the formation lithology hardness coefficient in the "high" subset and the membership degree of the mud flow resistance in the "high" subset. Then, the minimum value of these two membership degrees is taken as the activation degree of the rule. This activation degree reflects the degree of conformity between the current working condition and the rule. The higher the activation degree, the stronger the applicability of the rule to the current working condition.

[0154] For the output of each rule, the membership function of the corresponding output variable is truncated using the calculated activation value. This means that the membership of the output variable in its fuzzy subset is adjusted according to the activation value, so that the fuzzy subset of the output is more consistent with the current working conditions.

[0155] Deblurring process:

[0156] The outputs of all rules are synthesized, and then the centroid method is used to calculate the final output value, which is the specific weight adjustment amount of each output variable. The calculation process of the centroid method is as follows: first, calculate the center value of the fuzzy subset of each rule's output; then, multiply the activation of each rule by the corresponding center value and sum them; finally, divide by the sum of the activations of all rules. This calculation result is the weight adjustment amount of each output variable, which is a specific value used for subsequent weight adjustments.

[0157] Weight adjustment constraints:

[0158] To ensure that the sum of the corrected weights is always 1, a normalization factor is introduced. This factor is calculated by adding the reciprocal of the sum of the current initial weights and their respective weight corrections. Then, this normalization factor is multiplied by the initial weight plus the weight correction of each output variable to obtain the final corrected weights. This ensures that after considering the weight corrections of all objectives, their sum is still 1, thus maintaining the rationality and effectiveness of the multi-objective evaluation function.

[0159] Through the above series of calculations, the fuzzy logic system can dynamically adjust the weights of each objective in the multi-objective evaluation function based on the real-time collected downhole operating parameters, so as to better adapt to different downhole operating conditions and achieve a more optimized drilling control strategy.

[0160] In a preferred embodiment of the present invention, step S3: according to the dynamic adjustment command generated in step S2, during the target azimuth tracking process, a first target point, a second target point, and a third target point are determined; a dynamic triangle is constructed using the first target point, the second target point, and the third target point, and the centroid coordinates of the dynamic triangle are used as the azimuth correction value, including:

[0161] The real-time azimuth angle of the current tool face is converted into the real-time position point in the polar coordinate system and taken as the first target point. With the wellhead center as the origin, the azimuth angle coordinate plane is constructed with a unit radius. The coordinates of the first target point corresponding to the real-time azimuth angle are (cosθ1, sinθ1), where θ1 is the real-time acquired tool face azimuth angle.

[0162] The preset target tool surface angle is converted into a theoretical position point in polar coordinates and used as the second target point. The coordinates of the theoretical position point are (cosθ). t sinθ t ), where θ t The preset target tool face angle;

[0163] The third target point, i.e. the offset compensation point, is calculated based on the historical azimuth offset. The direction of the line connecting the real-time position point and the theoretical position point is determined. The offset distance is determined proportionally according to the magnitude of the historical azimuth offset along the reverse extension of the line. Starting from the real-time position point, the line is extended in the opposite direction by a distance proportional to the magnitude of the historical offset to obtain the coordinates of the third target point. This proportional relationship is preset based on the mapping relationship between offset and adjustment amplitude in historical drilling data.

[0164] The coordinates of the three target points in the polar coordinate system are converted to the coordinates in the rectangular coordinate system. A dynamic triangle is constructed using the three rectangular coordinate points. The arithmetic mean of the coordinates of the three vertices of the dynamic triangle is calculated as the centroid coordinates. The centroid coordinates are converted into angle values ​​in the polar coordinate system, and the angle values ​​are used as azimuth correction values.

[0165] In this embodiment of the invention, the specific calculation process of step S3 above can be implemented in the following way:

[0166] Step 1: Determine the first target point (real-time location):

[0167] Coordinate plane setting: With the wellhead center as the origin, construct an azimuth coordinate plane (polar coordinate system) with a unit radius. All angles are measured with reference to this origin. Convert the real-time acquired tool face azimuth (denoted as θ1, for example, the actual angle that the current tool face is pointing to) into coordinate points in the polar coordinate system. In polar coordinates, the rectangular coordinates corresponding to any angle on the unit circle are (cosθ, sinθ). Therefore, the coordinates of the first target point corresponding to the real-time azimuth θ1 are (cosθ1, sinθ1). This point represents the actual position of the tool face on the azimuth plane.

[0168] Step 2: Determine the second target point (theoretical location).

[0169] Target angle conversion: Converts the preset target tool face angle (denoted as θ) t For example, the tool surface angle required by the design is also converted to coordinate points in the polar coordinate system. According to the rules for converting polar coordinates to rectangular coordinates, the target angle θ t The corresponding coordinates of the second target point are (cosθ) t sinθ t This point represents the theoretical target location that the tool face needs to track, which is the ideal state for directional drilling.

[0170] Step 3: Determine the third target point (offset compensation point).

[0171] Analyze historical offsets: Review the historical azimuth offsets recorded in step S1 (i.e., the difference between the actual azimuth and the target angle), and calculate the offset amplitude (such as the average offset amplitude or the maximum offset amplitude) over a period of time before the current moment.

[0172] Determine the direction and distance of compensation:

[0173] Connection direction: Connect the first target point (real-time position) and the second target point (theoretical position) to form a straight line. The offset compensation point should be extended in the opposite direction of this straight line (i.e., starting from the real-time position point and extending in a direction away from the theoretical position point, opposite to the connection direction). Based on the mapping relationship between "offset magnitude" and "effective adjustment range" in historical drilling data (for example, when the historical average offset is 10°, a compensation adjustment of 5° is required to offset the lag), a proportional coefficient (such as 30% of the offset amplitude) is preset.

[0174] Calculate the coordinates of the compensation point: Starting from the real-time location point, extend a distance along the reverse extension line. This distance is proportional to the magnitude of the historical offset. For example, if the historical average offset magnitude is Δθ, then the compensation distance is k×Δθ (k is a preset proportional coefficient), and the coordinates of the third target point are finally obtained.

[0175] Step 4: Construct a dynamic triangle and calculate the centroid coordinates

[0176] Coordinate unification transformation: The polar coordinates (angle-radius) of the three target points are uniformly converted into coordinates (x, y) in the rectangular coordinate system. The three rectangular coordinate points (first target point, second target point, and third target point) are used as vertices to form a dynamic triangle. The shape of the triangle changes with the real-time azimuth angle and historical offset.

[0177] Calculate the centroid: The centroid coordinates are the arithmetic mean of the coordinates of the three vertices, i.e.:

[0178] The x-coordinate of the center of gravity = (x of the first target point + x of the second target point + x of the third target point) ÷ 3;

[0179] The y-coordinate of the center of gravity = (y of the first target point + y of the second target point + y of the third target point) ÷ 3;

[0180] The center of gravity represents the combined position of the three target points, integrating the current state, goal orientation, and historical compensation.

[0181] Step 5: Convert the barycenter coordinates to azimuth correction values

[0182] Polar coordinate transformation: Convert the rectangular coordinates (x, y) of the centroid to angle values ​​(i.e., azimuth) in the polar coordinate system. Calculate the angle between the line connecting the centroid and the origin using the arctangent function. This angle is the azimuth correction value.

[0183] Correction logic: This correction value takes into account the current tool face position (first point), the target position (second point), and the compensation position caused by historical offset (third point), and is used to adjust the rotation direction and amplitude of the drill string so that the tool face angle gradually approaches the target through closed-loop adjustment.

[0184] In this embodiment of the invention, the first point reflects the current state, the second point clarifies the target orientation, and the third point compensates in advance for the adjustment lag caused by drill string inertia or friction through historical offsets, avoiding overshoot or oscillation. The center of gravity, as the average position of the three, can smooth the deviation of different target points, making the correction direction closer to the actual dynamic needs, and improving the stability and accuracy of trajectory control. The compensation distance of the third target point is set based on the historical offset pattern, enabling the system to have self-learning ability and adapt to different downhole conditions (such as the adjustment delay of high friction formations).

[0185] In a preferred embodiment of the present invention, step S4: correcting the forward and reverse rotation angles and period parameters of the drill string according to the azimuth correction value to obtain the corrected dynamic adjustment command, including:

[0186] Calculate the azimuth correction value and the angular deviation between the current real-time azimuth of the tool face;

[0187] Based on the circular characteristics of the azimuth angle, the minimum angle difference in the clockwise direction is taken as the effective adjustment deviation;

[0188] Based on the preset forward and reverse angle increment range in the dynamic adjustment command, and combined with the effective adjustment deviation, the forward and reverse direction of the drill string and the corrected rotation angle are determined: if the correction value is greater than the real-time azimuth angle and the deviation is within the range of 0-180 degrees, it is determined to be a forward adjustment, and the rotation angle is the weighted average of the deviation value and the initial rotation angle increment; if the deviation exceeds 180 degrees, 360 degrees minus the deviation value is taken as the reverse adjustment angle, and the direction is counterclockwise.

[0189] Based on the fluctuation frequency of historical azimuth offset and the rate of change of the current dynamic triangle centroid coordinates, the initial period parameters are corrected. Specifically, when the rate of change of the centroid coordinates exceeds a preset threshold, the adjustment period is shortened to improve the response speed, and when the offset fluctuation tends to stabilize, the period is extended to reduce the mechanical load, so as to obtain the corrected period parameters.

[0190] In this embodiment of the invention, the specific calculation process of step S4 above can be as follows:

[0191] The first step is to calculate the angle deviation:

[0192] Initial deviation calculation: Subtract the azimuth correction value obtained in step S3 (representing the ideal angle after considering the current state, target, and historical compensation) from the real-time acquired tool face azimuth angle (the current actual angle) to obtain the initial angle deviation. For example, if the correction value is 45° and the real-time azimuth angle is 30°, the initial deviation is +15°; if the correction value is 350° and the real-time azimuth angle is 10°, the initial deviation is -20° (or 340°, which requires subsequent processing of circumferential characteristics).

[0193] Step 2: Address circumferential characteristics and determine effective adjustment deviations.

[0194] Circular angle characteristics: Since the azimuth angle is a 360° cycle angle, direct subtraction may result in a deviation of more than 180° (e.g., the deviation between 350° and 10° is 340°, but the actual minimum adjustment angle is 20°, which is the shortest path clockwise or counterclockwise).

[0195] Minimum angle difference calculation: The minimum angle difference in the clockwise direction is taken as the effective adjustment deviation, and the specific rules are as follows:

[0196] If the absolute value of the initial deviation is ≤180°, then this deviation is directly used as the effective adjustment deviation;

[0197] If the absolute value of the initial deviation is greater than 180°, then subtract the absolute value of the deviation from 360° to obtain the minimum angle difference in the clockwise direction (for example, when the deviation is -340°, the effective adjustment deviation is 20°, and the direction is clockwise).

[0198] Step 3: Determine the forward and reverse directions and the corrected rotation angle.

[0199] Determine the forward and reverse direction:

[0200] Forward adjustment: If the correction value is greater than the real-time azimuth angle and the effective adjustment deviation is within the range of 0-180° (inclusive), then it is determined to be the forward direction (i.e., clockwise rotation).

[0201] Reverse adjustment: If the correction value is less than the real-time azimuth angle, or the effective adjustment deviation exceeds 180° (in this case, take 360° minus the deviation value as the actual adjustment angle, and the direction is counterclockwise).

[0202] Calculate the rotation angle:

[0203] When rotating forward, the rotation angle is taken as the weighted average of the effective adjustment deviation value and the initial rotation angle increment in the dynamic adjustment command (for example, the deviation value accounts for 60% of the weight, and the initial increment accounts for 40% of the weight; the specific weight can be dynamically set according to the historical adjustment effect), ensuring that the adjustment range takes into account both the current deviation and the preset parameters.

[0204] When reversing, the rotation angle is 360° minus the effective adjustment deviation value (e.g., if the deviation is 340°, the reversal angle is 20°), and the direction is counterclockwise to avoid taking a long detour for adjustment (e.g., from 350° to 10°, reversing by 20° is more efficient than rotating clockwise by 340°).

[0205] Step 4: Adjust the period parameters

[0206] Analyze historical offsets and current rate of change:

[0207] Historical offset fluctuation frequency: Review the historical azimuth offset data recorded in step S1, calculate the recent offset fluctuation frequency (such as the number of times the offset changes by more than a certain threshold per unit time), and reflect the stability of the tool face angle.

[0208] Centroid coordinate change rate: Calculates the rate of change of the position of the centroid coordinates of the dynamic triangle within two adjacent adjustment cycles (i.e., the change range of the x and y components of the centroid coordinates per unit time), reflecting the dynamic trend of the current tool face adjustment.

[0209] Periodic adjustment strategy:

[0210] In a fast-response scenario: if the rate of change of the center of gravity coordinates exceeds the preset threshold (e.g., angle change > 5° per unit time), or the frequency of historical offset fluctuations is high (e.g., fluctuations more than 3 times per minute), it indicates that the tool face angle is unstable and the adjustment cycle needs to be shortened (e.g., from the initial cycle of 10 seconds to 5 seconds) to improve the system response speed and correct deviations in a timely manner.

[0211] Stable scenario: If the rate of change of the center of gravity coordinate is low (e.g., angle change < 2° per unit time) and the historical offset fluctuation tends to be stable (e.g., offset change < 1° for 5 consecutive cycles), it indicates that the current adjustment is close to the target. Extend the adjustment cycle (e.g., from 10 seconds to 15 seconds) to reduce the mechanical load caused by frequent drill string movements and reduce equipment wear.

[0212] In this embodiment of the invention, circumferential characteristic processing avoids ineffective large-angle rotations (such as converting a 340° deviation into a 20° reversal), thereby improving adjustment efficiency; the weighted average rotation angle takes into account both the current deviation and preset parameters, balancing adjustment accuracy and mechanical load, and dynamically adjusts the cycle according to historical fluctuation patterns and the current rate of change, responding promptly under rapidly changing operating conditions and reducing the frequency of action under stable operating conditions, thereby improving system robustness and equipment lifespan.

[0213] In another preferred embodiment of the present invention, step S5 above: controlling the drill string movement based on the modified dynamic adjustment command, so that the downhole tool face continuously approaches the target azimuth angle through closed-loop iterative adjustment, may include:

[0214] The corrected drill string motion parameters are obtained from step S4, including the forward and reverse direction (clockwise / counterclockwise), the corrected rotation angle (e.g., 20° clockwise rotation), and the adjustment period (e.g., 5 seconds). These parameters are converted into control signals (e.g., motor drive commands) for the drill string drive system, driving the drill string to rotate in the specified direction and angle, and completing the action within a preset period. For example, if the corrected parameter is "15° counterclockwise rotation, period 8 seconds," then the motor is controlled to rotate counterclockwise at a constant speed, and the action stops after 8 seconds.

[0215] The current toolface azimuth angle (denoted as θ2) is acquired in real time using an inertial navigation system (such as MWD measurement-while-drilling tool). This data reflects the actual angle after the drill string moves. The newly acquired azimuth angle θ2 is compared with the target toolface angle θ. t Compare and calculate the current deviation Δθ = θ² - θ t (Considering the characteristics of a circle, take the smallest angle difference within the range of 0-360°). For example, if the target angle is 90° and the current angle is 85°, the deviation is +5°; if the current angle is 350° and the target angle is 10°, the minimum deviation is 20° (counterclockwise).

[0216] Step 3: Determine whether to proceed to the next round of adjustment.

[0217] Accuracy verification: Compare the current deviation Δθ with the preset accuracy threshold (e.g., 1°):

[0218] If △θ≤th threshold, it is considered that the tool surface has approached the target azimuth angle, the closed-loop adjustment is terminated, and a stable tracking state is entered;

[0219] If Δθ > threshold, it means that further adjustment is needed, and the next iteration should begin.

[0220] Data recording: The adjusted azimuth angle θ2, deviation Δθ, actual rotation angle, period parameters, etc. are stored in the historical database to provide historical offset reference for subsequent adjustments (such as the compensation calculation of the third target point in step S3).

[0221] Step 4: Closed-loop iterative adjustment (if the deviation does not meet the target)

[0222] Restart the dynamic adjustment process: Use the current azimuth angle θ2 as the new real-time position point (first target point), and the target angle θ t Still using the theoretical position point (second target point), calculate the third target point (offset compensation point) by combining the latest historical offset (including the current deviation △θ), and repeat steps S3-S4:

[0223] Construct a new dynamic triangle and calculate the centroid coordinates as the new azimuth correction value;

[0224] Adjust the forward and reverse angles and cycle parameters based on the new correction values, and generate the next dynamic adjustment command.

[0225] Iterative optimization logic: Each iteration is based on the latest real-time data and historical offset patterns, dynamically adjusting the compensation magnitude and response cycle to form a closed loop of "execution-measurement-correction-re-execution" to gradually reduce the deviation.

[0226] When the azimuth deviation collected for N consecutive times (e.g., 3 times) is ≤ the preset threshold, or the rate of change of the centroid coordinate within the adjustment cycle approaches 0, it is determined that the tool face has stabilized near the target azimuth, and the active adjustment is stopped, switching to real-time monitoring mode. After the adjustment is terminated, the azimuth of the tool face still needs to be monitored in real time. If the deviation suddenly increases due to factors such as stratum disturbance (e.g., exceeding twice the threshold), a new round of closed-loop adjustment is automatically triggered to ensure that the tool face remains stable at the target angle.

[0227] In this embodiment of the invention, the tool face angle is dynamically calibrated through a cycle of "execution command → real-time feedback → deviation calculation → parameter correction," avoiding the blindness of single adjustments and adapting to complex downhole environments (such as frictional changes and drill string vibration). The historical offset of each iteration is accumulated and recorded to optimize the compensation amplitude (such as the offset distance of the third target point) and cycle parameters of subsequent adjustments, enabling the system to have self-optimization capabilities, reducing manual intervention, and ensuring that the adjustment process is neither too frequent (reducing mechanical wear) nor too infrequent (such as restarting the closed loop when the deviation suddenly increases), thus balancing control accuracy and system stability.

[0228] like Figure 2 As shown, embodiments of the present invention also provide an intelligent drilling guide system for oil drilling tool faces, comprising:

[0229] The acquisition module is used to acquire the azimuth data of the downhole tool face in real time and simultaneously record the historical azimuth offset.

[0230] The generation module is used to process the azimuth data in real time and generate dynamic adjustment instructions in combination with the preset target tool face angle. The dynamic adjustment instructions include the forward and reverse rotation angles of the drill string and the period parameters.

[0231] The correction module is used to determine the first target point, the second target point, and the third target point during the target azimuth tracking process according to the generated dynamic adjustment instructions; construct a dynamic triangle using the first target point, the second target point, and the third target point, and use the centroid coordinates of the dynamic triangle as the azimuth correction value;

[0232] The processing module is used to correct the forward and reverse rotation angles and period parameters of the drill string according to the azimuth correction value, so as to obtain the corrected dynamic adjustment command.

[0233] The control module is used to control the drill string movement based on the corrected dynamic adjustment commands, so that the downhole tool face can continuously approach the target azimuth angle through closed-loop iterative adjustment.

[0234] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent steering method for a petroleum drilling tool face, characterized in that, The method comprises: Step S1: Real-time acquisition of azimuth angle data of downhole tool face and synchronous recording of historical azimuth angle deviation, comprising: real-time acquisition of tool face azimuth angle data by downhole sensor array, preprocessing of azimuth angle data, and generation of preprocessed azimuth angle data stream; storing historical azimuth angle data with a preset time window length and calculating historical azimuth angle deviation; Step S2: Real-time processing of the azimuth angle data, combined with a preset target tool face angle to generate a dynamic adjustment instruction, the dynamic adjustment instruction including the positive and negative rotation angle and cycle parameters of the drill string; Step S3: determining the first target point, the second target point and the third target point in the process of tracking the target azimuth angle according to the dynamic adjustment instruction generated in step S2; constructing a dynamic triangle through the first target point, the second target point and the third target point, and taking the barycentric coordinates of the dynamic triangle as the azimuth correction value, including: taking the current tool face real-time azimuth angle as the first target point in the polar coordinate system, taking the wellhead center as the coordinate origin, taking a unit radius to construct an azimuth coordinate plane, the first target point coordinate corresponding to the real-time azimuth angle is (cosθ1, sinθ1), wherein θ1 is the tool face azimuth angle collected in real time; taking the preset target tool face angle as the second target point in the polar coordinate system, the theoretical position point coordinate is (cosθ t , sinθ t ), wherein θ t is the preset target tool face angle; calculating the third target point, i.e., the offset compensation point, based on the historical azimuth angle offset, determining the direction of the line connecting the real-time position point and the theoretical position point, extending in the opposite direction of the line, determining the offset distance according to the proportional relationship of the amplitude of the historical azimuth angle offset, taking the real-time position point as the starting point, extending in the opposite direction of the line by a distance proportional to the amplitude of the historical azimuth angle offset, obtaining the third target point coordinate, the proportional relationship is pre-set according to the mapping relationship between the offset and the adjustment amplitude in the historical drilling data; converting the coordinates of the three target points in the polar coordinate system into the coordinates of the rectangular coordinate system, constructing a dynamic triangle with the three rectangular coordinate points, calculating the arithmetic mean of the coordinates of the three vertices of the dynamic triangle as the barycentric coordinates, converting the barycentric coordinates into the angle value in the polar coordinate system, and taking the angle value as the azimuth correction value; Step S4: Correcting the positive and negative rotation angle and cycle parameters of the drill string according to the azimuth angle correction value to obtain a corrected dynamic adjustment instruction; Step S5: Controlling the drill string action based on the corrected dynamic adjustment instruction to make the downhole tool face continuously approach the target azimuth angle through closed-loop iterative adjustment.

2. The intelligent steering method of a petroleum drilling tool face according to claim 1, wherein, Step S2: Real-time processing of the azimuth angle data, combined with a preset target tool face angle to generate a dynamic adjustment instruction, the dynamic adjustment instruction including the positive and negative rotation angle and cycle parameters of the drill string, comprising: inputting the real-time azimuth angle data stream output by step S1, the historical azimuth angle deviation, and the preset target angle into a pre-trained artificial intelligence model, the artificial intelligence model being a hybrid architecture based on long short-term memory network and convolutional neural network, and being used to learn the mapping relationship between azimuth angle dynamic change and drill string control parameters; extracting the spatiotemporal features of the azimuth angle data through the artificial intelligence model, and predicting the real-time dynamic response of the drill string combined with the current drilling parameters to obtain a prediction result; generating an initial dynamic adjustment instruction according to the prediction result, the initial dynamic adjustment instruction including a start angle threshold of drill string positive and negative rotation, a rotation angle increment, and a rotation cycle parameter, wherein the start angle threshold of drill string positive and negative rotation is dynamically corrected by the amplitude and direction of the historical deviation; performing multi-objective optimization of the initial dynamic adjustment instruction and the target angle constraint condition, the multi-objective optimization conditions including azimuth angle tracking accuracy, drill string mechanical load balance, and energy consumption threshold of adjustment action, and finally generating a feasible solution satisfying the constraint as the dynamic adjustment instruction.

3. The intelligent steering method of a petroleum drilling tool face according to claim 2, wherein, performing multi-objective optimization of the initial dynamic adjustment instruction and the target angle constraint condition, the multi-objective optimization conditions including azimuth angle tracking accuracy, drill string mechanical load balance, and energy consumption threshold of adjustment action, and finally generating a feasible solution satisfying the constraint as the dynamic adjustment instruction, comprising: based on the positive and negative rotation angle threshold, the rotation angle increment, and the cycle parameter in the initial dynamic adjustment instruction, constructing a multi-objective evaluation function including azimuth angle tracking accuracy error, drill string axial torque fluctuation value, and single adjustment energy consumption, wherein the weight coefficients of each target are dynamically allocated according to the real-time downhole working condition; using a particle swarm optimization algorithm with constraints to iteratively solve the multi-objective evaluation function, the constraint conditions including: the azimuth angle tracking accuracy error being not more than a preset threshold, the drill string mechanical load fluctuation value being in a safe interval, and the single adjustment energy consumption being lower than the rated power of the equipment; In the iteration process, the candidate solution set with the minimum azimuth tracking accuracy error is preferentially retained, and the final instruction parameter combination with the minimum mechanical load fluctuation value and energy consumption comprehensive score is screened in the candidate solution set to generate a preliminary optimized instruction; The downhole dynamic response simulation verification is performed on the preliminary optimized instruction, the tool face azimuth change trajectory after the preliminary optimized instruction is executed is simulated through the pre-set drill string dynamics model, and if the azimuth converges to the target angle and meets the load and energy consumption constraints in the simulation result, it is determined as a feasible solution; If the simulation verification fails, the weight coefficients of the multi-objective evaluation function are adjusted according to the error direction, a new optimized instruction is generated through re-iteration, and the dynamic adjustment instruction meeting all the constraint conditions is output until the final optimized dynamic adjustment instruction is obtained.

4. The intelligent steering method of a petroleum drilling tool face according to claim 3, wherein, Based on the positive and negative rotation angle threshold, the rotation angle increment and the period parameter in the initial dynamic adjustment instruction, a multi-objective evaluation function including the azimuth tracking accuracy error, the drill string axial torque fluctuation value and the single adjustment energy consumption is constructed, wherein the weight coefficients of each target are dynamically allocated according to the downhole real-time working condition, including: Real-time acquisition of downhole working condition parameters, including drilling pressure, drill string rotation speed, formation lithology hardness coefficient and mud flow resistance, as the basis for weight allocation; The initial weight proportion of the azimuth tracking accuracy error, the torque fluctuation and the energy consumption in the multi-objective evaluation function is calculated according to the working condition parameters; The initial weight proportion is dynamically corrected through the pre-trained fuzzy logic rule base to obtain the corrected weight coefficients, and the rule base is constructed based on the mapping relationship between the working condition parameters and the weight adjustment in the historical drilling data; The corrected weight coefficients are fused with the azimuth tracking accuracy error, the drill string axial torque fluctuation value and the single adjustment energy consumption respectively to form the weighted parameter items corresponding to each target, and the multi-objective evaluation function matched with the current downhole working condition is constructed through the combination of each weighted parameter item.

5. The intelligent steering method of a petroleum drilling tool face according to claim 4, wherein, Step S4: correcting the positive and negative rotation angle and the period parameter of the drill string according to the azimuth correction value to obtain the corrected dynamic adjustment instruction, including: Calculating the angle deviation of the azimuth correction value and the current tool face real-time azimuth; According to the circumferential characteristics of the azimuth, the minimum angle difference in the clockwise direction is taken as the effective adjustment deviation; According to the pre-set positive and negative rotation angle increment range in the dynamic adjustment instruction, the drill string positive and negative rotation direction and the corrected rotation angle are determined by combining the effective adjustment deviation: if the correction value is greater than the real-time azimuth and the deviation is within 0-180 degrees, it is determined as positive rotation adjustment, and the rotation angle is the weighted average of the deviation value and the initial rotation angle increment; if the deviation exceeds 180 degrees, 360 degrees minus the deviation value is taken as the negative rotation angle, and the direction is counterclockwise; Based on the fluctuation frequency of the historical azimuth offset and the change rate of the current dynamic triangle barycenter coordinates, the initial period parameter is corrected, that is, when the change rate of the barycenter coordinates exceeds the pre-set threshold, the adjustment period is shortened to improve the response speed, and when the offset fluctuation tends to be stable, the period is lengthened to reduce the mechanical load, to obtain the corrected period parameter.

6. An intelligent drill steering system for a petroleum drilling tool face, the system implementing the method of any one of claims 1 to 5, characterized in that, including: An acquisition module is configured to acquire azimuth data of a downhole tool face in real time and record historical azimuth offset simultaneously; The generating module is configured to perform real-time processing on the azimuth angle data, and generate a dynamic adjustment instruction in combination with a preset target tool face angle, the dynamic adjustment instruction including a positive and negative rotation angle and a period parameter of the drill string; The correction module is configured to determine a first target point, a second target point, and a third target point in a process of tracking the target azimuth angle according to the generated dynamic adjustment instruction, construct a dynamic triangle through the first target point, the second target point, and the third target point, and take a barycentric coordinate of the dynamic triangle as an azimuth angle correction value; The processing module is configured to correct the positive and negative rotation angle and the period parameter of the drill string according to the azimuth angle correction value, so as to obtain a corrected dynamic adjustment instruction; The control module is configured to control the drill string to act based on the corrected dynamic adjustment instruction, so that the downhole tool face continuously approaches the target azimuth angle through closed-loop iterative adjustment.

7. A computing device, comprising: The method comprises: one or more processors; a storage device configured to store one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a program, and the program is executed by the processor to implement the method according to any one of claims 1 to 5.

Images