A real-time drilling path prediction method and system

Abstract

The application provides a real-time drilling path prediction method and system, the method inputs the current node coordinates, gamma value, predicted drilling step length and current design planning operation instruction into a prediction model to obtain a candidate point set with probability distribution, generates a plurality of candidate paths based on the coordinates and probability distribution of all candidate points to determine a preliminary prediction path, and further calculates the equivalent stress of the next predicted node of the drill rod based on the drill rod parameters, the current node posture information of the drill rod and the preliminary prediction path to finally generate a final prediction path in combination with the equivalent stress of the next predicted node of the drill rod. The above technical scheme can identify the bending overrun risk in advance, realizes the collaborative optimization of drilling accuracy and drill safety, and solves the problem that the existing technology ignores the stress of the drill rod and is prone to drill breakage.

Description

Technical Field

[0001] This application relates to the field of directional drilling technology, specifically to a real-time drilling path prediction method and system. Background Technology

[0002] In existing directional drilling or measurement while drilling (MWD) systems, drill bit path prediction directly determines hole quality, construction safety, and engineering efficiency. Traditional path prediction methods mainly rely on empirical estimation. Empirical estimation heavily depends on the driller's field experience, predicting drill bit position using historical construction data and manual projection software. This process is cumbersome, lacks real-time performance, and is susceptible to human error, making it difficult to adapt to complex formations and high-precision directional drilling requirements.

[0003] See Patent Publication No. CN201810253315 for a method to predict the optimal drilling path for horizontal wells. This method discloses predicting the direction of fracturing fractures using static geological data (such as rock mechanics parameters, fracture distribution, and stress field simulation) to determine the optimal path for the horizontal well. This method relies entirely on previously solidified exploration data, making it difficult to characterize hidden, discrete small geological structures and dynamic changes in formation parameters. It suffers from significant geological modeling errors, poor data timeliness, and inability to handle unforeseen formation changes, heterogeneity, and operational disturbances during drilling. It also lacks the ability for dynamic path correction and real-time control. When actual geological conditions deviate from the pre-modeled path, the applicability of the preset path decreases sharply, easily leading to problems such as wellbore deviation, path loss of control, substandard fracturing results, and rework. Meanwhile, the scheme does not consider the stress on the drill string assembly, the interaction between the drill bit and the rock, and the dynamic changes in drilling parameters. It lacks analysis and control of the mechanical behavior during the drilling process, and is prone to drill string wear, buckling, stuck drill, or even drill breakage in complex formations. It cannot take into account the comprehensive requirements of hole quality, construction safety and efficient mining, and is difficult to meet the engineering needs of high-precision and high-reliability directional drilling under complex geological conditions. Summary of the Invention

[0004] In view of the above problems, this application provides a real-time drilling path prediction method and system to solve the problem that existing prediction paths ignore "how the drill rod is stressed", which leads to the drill rod being prone to breakage in complex rock formations.

[0005] To achieve the above objectives, the inventors provide a real-time drilling path prediction method, which includes the following steps:

[0006] The drill pipe parameters, current node coordinates, attitude information, gamma value while drilling, expected drilling progress, and current design plan operation instructions are obtained. The operation instructions include expected roll angle and thrust.

[0007] The current node coordinates of the drill pipe, the gamma value while drilling, the expected drilling progress, and the current design plan operation commands are input into the drilling path prediction model to obtain a set of prediction candidate points; the set of prediction candidate points has There are 1 time steps, and each time step has 1 time step. There are 10 candidate points, each containing coordinates and its corresponding probability distribution;

[0008] Based on the coordinates and probabilities of all candidate points in the predicted candidate point set, several candidate paths are generated by sequentially combining them according to time steps, and the preliminary predicted paths are determined.

[0009] Based on the drill pipe parameters, the current node attitude information of the drill pipe, and the preliminary predicted path, calculate the equivalent stress of the next predicted node of the drill pipe.

[0010] A prediction path is generated based on the equivalent stress of the next prediction node of the drill pipe.

[0011] Furthermore, the training of the path prediction model includes the following steps:

[0012] A training sample set is constructed using thousands of sets of correlated data from historical drilling operations; each training sample includes:

[0013] Input feature sequence:

[0014] Operation command: Roll angle Propulsion ;

[0015] Sensor feedback: Pitch angle Heading angle Gamma value during drilling ;

[0016] Measured coordinates: ;

[0017] Dig deeper and make progress: ;

[0018] Output label sequence:

[0019] Measured path coordinates: ,…, ;

[0020] Construct a time series prediction model based on the Transformer architecture;

[0021] The feature sequence is input into a time-series prediction model based on the Transformer architecture; gamma data is used to correct the formation drag coefficient, and the time-series prediction model based on the Transformer architecture is trained to learn the turning efficiency of the drill bit under different formations; the difference between the output and the output label sequence is calculated using the loss function, and the model parameters are updated through backpropagation until the loss converges; the next prediction point set is obtained.

[0022] Furthermore, the preliminary predicted path is selected from several candidate paths based on the path with the smallest cumulative arc length.

[0023] Furthermore, the drill pipe parameters include the elastic modulus. Cross-sectional area Drill pipe outer diameter ;

[0024] The calculation of the equivalent stress at the next predicted node of the drill pipe, based on drill pipe parameters, the current node attitude information of the drill pipe, the coordinates of the next predicted node of the drill pipe, and the preliminary predicted path, includes the following steps:

[0025] Based on the coordinates of the next predicted node of the drill pipe and the preliminary predicted path, the attitude information and actual drilling progress of the next predicted node of the drill pipe are obtained.

[0026] The curvature of the predicted path is obtained by using the current node attitude information of the drill pipe, the actual drill progress, and the attitude information of the next predicted node of the drill pipe. ;

[0027] Then the equivalent stress at the next predicted node of the drill pipe :

[0028] ；

[0029] in, As the overall driving force.

[0030] Furthermore, the attitude information includes pitch angle. and heading angle ;

[0031] The curvature of the predicted path :

[0032] ;

[0033] in, Indicates the current node's pitch angle; Indicates the heading angle of the current node; Indicates the pitch angle of the next prediction node; Indicates the heading angle of the next prediction node. It indicates that progress has been made in drilling. The average pitch angle between the two nodes.

[0034] Furthermore, the step of generating the predicted path based on the equivalent stress of the next predicted node of the drill pipe includes the following steps:

[0035] Determine the equivalent stress at the next predicted node of the drill pipe. If the value exceeds a preset threshold, discard the initially predicted path and select a curvature path from several candidate paths. Curvature smaller than the initially predicted path The path that is predicted is used as the predicted path; otherwise, the preliminary predicted path is used as the predicted path.

[0036] Furthermore, an LLM decision layer is used to perform multi-objective optimization:

[0037] Equivalent stress at the next predicted node of the drill pipe When the threshold is exceeded, switch to safety priority mode and select a curvature from several candidate paths. Curvature smaller than the initially predicted path The path is used as the predicted path;

[0038] Equivalent stress at the next predicted node of the drill pipe When within the safe range, switch to correction priority mode and use the preliminary predicted path as the predicted path.

[0039] Furthermore, it also includes calculating the deviation vector between the next predicted node coordinates of the drill pipe and the target path based on the next predicted node coordinates of the predicted path; and calculating the roll angle correction.

[0040] Furthermore, the roll angle correction is based on the guide vector generated by the drill bit's fixed 1.25-degree eccentricity angle;

[0041] Calculate the roll angle correction based on the eccentricity direction. :

[0042] ；

[0043] in, This represents the deviation vector between the coordinates of the next predicted node of the drill pipe and the target path. This represents the formation coupling coefficient.

[0044] A real-time drilling path prediction system, applying the real-time drilling path prediction method, includes:

[0045] The drilling sensing end includes a triaxial IMU and a gamma detector; the triaxial IMU is used to acquire the pitch angle, roll angle and yaw angle of the current node in real time; the gamma detector is used to acquire the gamma value of the current node to reflect the formation lithology parameters.

[0046] The surface measurement end includes a hydraulic pressure sensor and an automatic section / meter counter. The hydraulic pressure sensor is used to collect propulsion pressure values ​​in real time, which are used to indirectly calculate the total propulsion force on the drill pipe. The automatic section / meter counter records the number of meters in real time, and accurately calculates the current drilling depth and coordinate mapping point by recording the number of drill rods lowered and the length of a single section.

[0047] The data processing center includes a mechanical calculation unit, a path estimation engine, and a decision unit. The FEA mechanical calculation unit is used to calculate the equivalent stress of the next predicted node of the drill pipe. The path estimation engine is used to obtain the next predicted point set. The decision unit generates a predicted path based on the equivalent stress of the next predicted node of the drill pipe.

[0048] The human-computer interaction terminal provides a visual decision-making interface, including a path deviation dashboard, a stress warning screen, and an expected roll angle command interface. The path deviation dashboard dynamically displays a deviation graph between the designed path and the actual predicted path. The stress warning screen uses different colors to indicate the current fatigue stress level of the drill pipe. The expected roll angle command interface displays the UI block of the "expected roll angle" command to guide manual angle adjustment.

[0049] Unlike existing technologies, the above-mentioned technical solution outputs a set of candidate points with probability distributions for multiple future steps through a drilling path prediction model. By introducing the candidate point set with probability distributions for multiple future steps, several candidate paths are generated and a preliminary predicted path is determined, thereby quantifying the impact of formation uncertainty on the path and avoiding the "blind confidence" of traditional single-point prediction. On this basis, the drill pipe parameters are integrated to calculate the equivalent stress of the drill pipe, which can identify the risk of exceeding the limit of drill pipe bending stress in advance during the path planning stage. This achieves the coordinated optimization of drilling path accuracy and drill bit safety, and improves the reliability and intelligence level of directional drilling under complex formation conditions.

[0050] The above description of the invention is merely an overview of the technical solution of this application. In order to enable those skilled in the art to better understand the technical solution of this application and to implement it based on the description and drawings, and to make the above-mentioned objectives and other objectives, features and advantages of this application easier to understand, the following description is provided in conjunction with the specific embodiments and drawings of this application. Attached Figure Description

[0051] The accompanying drawings are only used to illustrate the principles, implementation methods, applications, features, and effects of specific embodiments of the present invention and other related contents, and should not be considered as limitations on this application.

[0052] In the accompanying drawings of the instruction manual:

[0053] Figure 1 This is a flowchart illustrating the real-time drilling path prediction method described in a specific implementation. Detailed Implementation

[0054] To illustrate the possible application scenarios, technical principles, implementable specific solutions, and achievable objectives and effects of this application in detail, the following description, in conjunction with the listed specific embodiments and accompanying drawings, provides a detailed explanation. The embodiments described herein are merely illustrative of the technical solutions of this application and are therefore intended to limit the scope of protection of this application.

[0055] In this document, the term "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The term "embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment, nor does it specifically limit its independence or connection with other embodiments. In principle, in this application, as long as there are no technical contradictions or conflicts, the technical features mentioned in each embodiment can be combined in any way to form corresponding implementable technical solutions.

[0056] Unless otherwise defined, the technical terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the use of related terms herein is merely for the purpose of describing particular embodiments and is not intended to limit this application.

[0057] In the description of this application, the term "and / or" is used to describe the logical relationship between objects, indicating that three relationships can exist. For example, A and / or B means: A exists, B exists, and A and B exist simultaneously. Additionally, the character " / " in this document generally indicates that the preceding and following objects have an "or" logical relationship.

[0058] In this application, terms such as “first” and “second” are used only to distinguish one entity or operation from another, and do not necessarily require or imply any actual quantity, hierarchy or order relationship between these entities or operations.

[0059] Without further limitations, the use of terms such as “comprising,” “including,” “having,” or other similar open-ended expressions in this application is intended to cover non-exclusive inclusion, which does not exclude the presence of additional elements in a process, method, or product that includes the stated elements, such that a process, method, or product that includes a list of elements may include not only those defined elements but also other elements not expressly listed, or elements inherent to such a process, method, or product.

[0060] In this application, expressions such as "greater than", "less than", and "exceeding" are understood to exclude the stated number; expressions such as "above", "below", and "within" are understood to include the stated number. Furthermore, in the description of the embodiments of this application, "multiple" means two or more (including two), and similar expressions related to "multiple" are also understood in this way, such as "multiple groups" and "multiple times", unless otherwise explicitly specified.

[0061] In the description of the embodiments of this application, the space-related expressions used, such as "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "vertical," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential," indicate the orientation or positional relationship based on the orientation or positional relationship shown in the specific embodiments or drawings. They are only for the purpose of describing the specific embodiments of this application or for the reader's understanding, and do not indicate or imply that the device or component referred to must have a specific position, a specific orientation, or be constructed or operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0062] The processor described in the embodiments of this application can be implemented by hardware, firmware, software, or a combination thereof. It can be a circuit, one or more of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing device (DSPD), a programmable logic device (PLD), a field-programmable gate array (FPGA), a central processing unit (CPU), a controller, a microcontroller, or a microprocessor. It also includes other physical, biological, or chemical structures that can implement the same or equivalent functions as the processors listed above, such as biological neurons, quantum computing units, DNA computing units, etc., so that the processor can execute some or all of the steps in the computer program or method involved in the various embodiments of this application, or any combination of the steps mentioned therein.

[0063] The computer program involved in the embodiments can be stored in a computer device readable storage medium, which includes, but is not limited to, disks, magnetic tapes, magnetic cards, floppy disks, flash memory, optical disks, optical cards, read-only memory (ROM), random access memory (RAM), erasable programmable ROM (EPROM), and electrically erasable programmable ROM (EEPROM), etc., and also includes other biological, physical, or chemical structures that can achieve the same or equivalent functions as the storage media listed above, such as DNA, RNA, proteins, and other units with information storage capabilities. In specific embodiments, the storage medium involved can be one of the above-mentioned media types, or a combination of the above-mentioned media types. In different embodiments, the computer program involved in the embodiments can be centrally stored in a single medium, or distributed and stored in multiple media. The memory containing the computer device readable storage medium can be non-volatile memory or random access memory. These computer device readable storage media can be built into the device, or can be connected to the device involved in the embodiments as an external device or part of an external device. In some embodiments, the memory having a computer device readable storage medium is deployed locally; in other embodiments, the memory may be deployed remotely from the processor, for example, as a network-attached memory accessed via RF circuitry or an external port and a communication network, wherein the communication network may be the Internet, one or more intranets, a local area network (LAN), a wide area network (WLAN), a storage area network (SAN), or a suitable combination thereof, as long as computer device access to the memory is enabled. Furthermore, the computer program involved in the embodiments may be stored in plaintext / ciphertext form, or it may be designed as training data, integrated and recombined through model training and implicitly stored in the parameter states of a deep neural network or other machine learning model.

[0064] A real-time drilling path prediction method is proposed to address the problem that existing prediction methods only focus on "where the drill bit is" while neglecting "how the drill pipe is stressed," leading to drill pipe breakage in complex rock formations. The method outputs a set of candidate points with probability distributions for multiple future steps through a drilling path prediction model. By introducing these candidate point sets, several candidate paths are generated, and a preliminary predicted path is determined, thereby quantifying the impact of formation uncertainties on the path and avoiding the "blind confidence" of traditional single-point prediction. Based on this, drill pipe parameters are integrated to calculate the equivalent stress of the drill pipe, enabling early identification of potential risks of exceeding drill pipe bending stress limits during the path planning stage. This achieves synergistic optimization of drilling path accuracy and drill string safety, improving the reliability and intelligence level of directional drilling under complex formation conditions.

[0065] The following combination Figure 1 This paper provides a specific implementation method for a real-time drilling path prediction method, which includes the following steps:

[0066] S10. Obtain drill pipe parameters, current node coordinates, attitude information, gamma value while drilling, expected drilling progress, and current design plan operation instructions;

[0067] S20. Input the current node coordinates of the drill pipe, the gamma value while drilling, the expected drilling progress, and the current design plan operation commands into the drilling path prediction model to obtain a set of prediction candidate points; the set of prediction candidate points has There are 1 time steps, and each time step has 1 time step. There are 10 candidate points, each containing coordinates and its corresponding probability distribution;

[0068] S30. Based on the coordinates and probabilities of all candidate points in the predicted candidate point set, generate several candidate paths by combining them sequentially according to time steps, and determine the preliminary predicted paths.

[0069] S40. Based on the drill pipe parameters, the current node attitude information of the drill pipe, and the preliminary predicted path, calculate the equivalent stress of the next predicted node of the drill pipe.

[0070] S50. Generate a prediction path based on the equivalent stress of the next predicted node of the drill pipe.

[0071] In step S10, the drill pipe parameters can be obtained from the drill pipe manufacturer's technical specifications or through on-site measurements (such as standard material mechanics tests, caliper diameter measurement, and ultrasonic wall thickness measurement), including material parameters and structural parameters; the material parameters include the elastic modulus. Yield strength etc.; among which, elastic modulus The elastic modulus is a physical quantity that characterizes the ability of drill pipe materials to resist elastic deformation. It is defined as the ratio of stress to strain during the elastic deformation stage of the material. The higher the yield strength, the less likely the drill pipe is to undergo elastic bending under external forces; The critical stress value characterizing the transition of drill pipe material from elastic deformation to plastic deformation (i.e., permanent deformation) is defined as the stress value at which the drill pipe material transitions from elastic deformation to plastic deformation. When the equivalent stress inside the drill pipe exceeds the yield strength, the drill pipe will undergo irreversible plastic bending and even fracture. The structural parameters include cross-sectional area. Drill pipe outer diameter Drill pipe inner diameter The aforementioned cross-sectional area It refers to the cross-sectional area of ​​the drill pipe body perpendicular to the axis. It can be the original nominal cross-sectional area or the current actual remaining cross-sectional area updated in real time based on the wear detection data while drilling.

[0072] When the above cross-sectional area When the cross-section of the original drill pipe is used, it refers to the nominal design cross-sectional area of ​​the drill pipe before wear or corrosion, based on the outer diameter of the drill pipe. With inner diameter calculate:

[0073] ;

[0074] When the above cross-sectional area When considering the cross-sectional area after wear, the current actual remaining cross-sectional area of ​​the drill pipe, obtained through drilling-while-drilling measurement or offline testing, takes into account the wall thickness reduction caused by wear, corrosion, or mechanical damage. It can be calculated either based on the reduction in wall thickness on one side or directly from the current actual remaining wall thickness measured by an ultrasonic sensor. The expression obtained through direct calculation is as follows:

[0075] ;

[0076] in, The reduction in wall thickness on one side due to wear, corrosion, or mechanical damage can be calculated by using ultrasonic thickness gauges or electromagnetic ultrasonic (EMAT) probes to collect the current wall thickness value at multiple points along the circumference of the drill pipe during drilling. This value is then compared with the original wall thickness (factory parameters) of the drill pipe to determine the reduction on one side. .in, It can be acquired in real time during the drilling process using ultrasonic thickness sensors while drilling and electromagnetic ultrasonic (EMAT) probes.

[0077] In step S10, the current node coordinates are the three-dimensional spatial coordinates of the drill pipe's current position, which are obtained by vector decomposition recursion calculation using the initial reference coordinates (e.g., establishing a three-dimensional wellbore coordinate system with a known reference point at the wellhead or downhole as the origin and recording the initial drill pipe entry position coordinates), drill progress data, and drilling attitude data. The aforementioned attitude information is obtained by the three-axis IMU at the drilling sensing end. The three-axis IMU is used to acquire the pitch angle, roll angle, and yaw angle of the current node in real time. The drilling gamma value is acquired in real time by the gamma detector at the drilling sensing end, which measures the formation radioactivity intensity in real time. It is used to identify formation lithology, distinguish between soft and hard formations, and invert formation resistance parameters to reflect formation lithology parameters.

[0078] In step S20, the drilling path prediction model is based on (Long Short-Term Memory Network) or Using historical drilling data as training samples, the system learns the nonlinear mapping relationship between formation lithology, design operation commands, drilling input parameters, and drill bit path evolution. The drilling gamma value is used to characterize formation lithology and dynamically correct the formation resistance coefficient to adjust the drill bit turning efficiency prediction results. The predicted candidate point set possesses… There are 1 time steps, and each time step has 1 time step. There are 10 candidate points, each containing coordinates and its corresponding probability distribution.

[0079] The following time series prediction model based on the Transformer architecture is used as the basic architecture. Training the above path prediction model includes the following steps:

[0080] Thousands of sets of correlated data from historical drilling operations were collected as training samples, forming a training sample set. Each training sample includes:

[0081] Input feature sequence (past) (time step)

[0082] Operation command: Roll angle Propulsion ;

[0083] Sensor feedback: Pitch angle Heading angle Gamma value during drilling ;

[0084] Measured coordinates: ;

[0085] Dig deeper and make progress: (The actual length from the previous node to the current node);

[0086] Output label sequence (future) +n (time step)

[0087] Measured path coordinates: ,…, ;

[0088] The multivariate features at each time step (operation commands, sensor feedback, measured coordinates, and drilling progress) are mapped into fixed-dimensional embedding vectors. Among them, the gamma value while drilling Simultaneously, it functions as an independent feature channel and outputs the formation resistance feature code through a small fully connected subnetwork. Formation resistance feature code It will interact with multiple features in subsequent attention mechanisms.

[0089] Embedded sequence The input is fed into the Transformer encoder. The Transformer encoder consists of L identical layers stacked together (e.g., L=4), each layer containing:

[0090] Multi-Head Self-Attention: Allows the model to associate information from different time steps in different representation subspaces, especially capturing the long-range dependency between formation lithology (gamma) and drill bit steering efficiency.

[0091] Feedforward Neural Network (FFN): Two-layer linear transformation plus ReLU activation.

[0092] Residual connectivity and layer normalization.

[0093] Among them, the attention mechanism can learn automatically: several steps after the occurrence of high gamma values ​​(hard rock), the model should reduce its sensitivity to predicting lateral offset, that is, correct the "turning efficiency".

[0094] Use a Transformer decoder (or directly use the encoder output followed by a one-dimensional convolution) to generate the future. n The decoder takes the encoder output and the predicted preceding steps (autoregressive or non-autoregressive) as input for each future time step. n Output a value with a probability distribution. m There are 10 candidate points.

[0095] Training utilizes a loss function (such as negative log-likelihood (NLL)) to calculate the difference between the output and the output label sequence, and updates the model parameters through backpropagation until the loss converges; thus obtaining the next set of predicted points. Furthermore, auxiliary losses (such as L2 regularization and KL divergence regularization) can be added to prevent overfitting.

[0096] After training, the model has the following capabilities: given the current node coordinates, gamma value, drilling length, and historical posture sequence, it can output a set of candidate points with probability distribution for the next multiple steps.

[0097] In step S30, the step of generating several candidate paths by sequentially combining the coordinates and probabilities of all candidate points in the predicted candidate point set according to time steps refers to generating several candidate paths from each time step. Choose one of the candidate points, and sort them in chronological order 1, 2, ... A connection is formed, i.e., a candidate path is created; the total number of candidate paths is... Considering the actual computational cost, it is usually not necessary to enumerate all paths, but rather to generate several representative paths (such as the top few paths with the highest probability) based on dynamic programming.

[0098] From the generated candidate paths, a "preliminary predicted path" is selected, and different strategies can be chosen according to the needs. These include: the maximum probability product path, the path with the minimum cumulative arc length (among multiple possible paths formed by connecting multiple candidate nodes sequentially, the path with the smallest sum of the spatial straight-line distances (or curve lengths) between all adjacent nodes on each path), the path with the minimum bending energy (smoothest path), the weighted comprehensive score path (considering probability, cumulative arc length, and curvature for each path, assigning weight coefficients, and selecting the path with the highest comprehensive score), and the safety-first path (first selecting all paths that satisfy the equivalent stress). (Paths less than a preset threshold, etc.)

[0099] In drilling operations, the shortest arc length means the shortest drilling distance, typically saving time, fuel, and drill bit wear. In some embodiments, the preliminary predicted path is selected from a pool of candidate paths based on the minimum cumulative arc length. From the currently known node... Starting from the beginning, calculate the cumulative arc length (the sum of Euclidean distances between adjacent nodes) of each path, and select the path with the shortest total length, which is the path with the smallest cumulative arc length. However, the path with the smallest cumulative arc length is accompanied by a large curvature (sharp turn), which may lead to excessive stress on the drill pipe. Therefore, if this path is selected again, the equivalent stress of the drill pipe at the next predicted node on this path needs to be verified: if the equivalent stress exceeds the threshold, it is abandoned, and a smooth path with a larger radius of curvature is selected instead (in which case the cumulative arc length will increase) to prevent irreversible plastic bending or even fracture of the drill pipe.

[0100] In step S40, based on the drill pipe parameters, the current node attitude information of the drill pipe, and the preliminary predicted path, the equivalent stress of the next predicted node of the drill pipe is calculated, including the following steps:

[0101] Based on the preliminary predicted path, the attitude information and drilling progress of the next predicted node of the drill pipe are obtained;

[0102] The curvature of the predicted path is obtained by using the current node attitude information of the drill pipe, the drill progress, and the attitude information of the next predicted node of the drill pipe. ;

[0103] Then the equivalent stress at the next predicted node of the drill pipe :

[0104] ；

[0105] in, As the overall driving force.

[0106] In calculating equivalent stress At that time, the original cross-sectional area can be selected based on the health condition of the drill pipe. Or cross-sectional area after wear When wear is severe, use This makes the axial stress term This increases the stress safety threshold, thus enabling proactive protection of aging or damaged drill pipes.

[0107] The attitude information includes pitch angle. and heading angle ;

[0108] but Indicates the current node's pitch angle; Indicates the heading angle of the current node;

[0109] Based on the preliminary predicted path, the attitude information and drill progress of the next predicted node of the drill pipe are obtained:

[0110] The current node Next prediction node of the initial prediction path If we consider them as two points in space, then from point to The displacement vector is:

[0111] ;

[0112] The spatial length of this vector is the actual drilling progress. ;

[0113] Then the pitch angle of the next prediction node :

[0114] ;

[0115] Average pitch angle between two nodes :

[0116] ;

[0117] Next prediction node heading angle :

[0118]

[0119] The curvature of the predicted path :

[0120] .

[0121] In step S40, the step of generating the prediction path based on the equivalent stress of the next predicted node of the drill pipe includes the following steps:

[0122] Determine the equivalent stress at the next predicted node of the drill pipe. If the value exceeds a preset threshold, discard the initially predicted path and select a curvature path from several candidate paths. Curvature smaller than the initially predicted path The path that is predicted is used as the predicted path; otherwise, the preliminary predicted path is used as the predicted path.

[0123] The aforementioned preset threshold can be set based on the yield strength of the drill pipe material, preferably... This threshold ensures that the drill pipe always operates within the elastic range, avoiding entry into the plastic yield zone.

[0124] When the equivalent stress at the next predicted node of the drill pipe If the preset threshold is not exceeded, the preliminary predicted path will be used as the predicted path.

[0125] When the equivalent stress at the next predicted node of the drill pipe When the preset threshold is exceeded, the initially predicted path is discarded. From the pre-generated candidate paths (see the aforementioned "Candidate Paths Generated by Time Step Combination"), paths that meet the following conditions are selected:

[0126] Repeat the above calculation steps and select a curvature Curvature smaller than the initially predicted path The path and the equivalent stress of the next predicted node of that path. Paths with curvature less than a preset threshold are used as predicted paths. If multiple paths exist, the path with the smallest cumulative arc length is preferred. If this condition is not met, the path with the smallest maximum curvature is selected as the final predicted path. By forcibly changing the predicted path to reduce curvature, the "uncontrollable" risk is transformed into "transparent force," thus preventing irreversible plastic bending or even breakage of the drill pipe.

[0127] The above steps can be performed using an LLM decision layer to achieve the multi-objective optimization described above:

[0128] Equivalent stress at the next predicted node of the drill pipe When the threshold is exceeded, switch to safety priority mode and select a curvature from several candidate paths. Curvature smaller than the initially predicted path The path is used as the predicted path;

[0129] Equivalent stress at the next predicted node of the drill pipe When within the safe range, switch to correction priority mode and use the preliminary predicted path as the predicted path.

[0130] This application also includes calculating the deviation vector between the next predicted node coordinates of the drill pipe and the target path based on the next predicted node coordinates of the predicted path; and calculating the roll angle correction.

[0131] The roll angle correction is based on the guide vector generated by the fixed eccentricity angle of the drill bit; using the guide vector generated by the fixed eccentricity angle of the drill bit at 1.25 degrees, the roll angle correction is calculated according to the direction of the eccentricity. :

[0132] ；

[0133] in, This represents the deviation vector between the coordinates of the next predicted node of the drill pipe and the target path. This represents the formation coupling coefficient.

[0134] This application also provides a real-time drilling path prediction system, which applies the real-time drilling path prediction method and includes:

[0135] The drilling sensing end includes a triaxial IMU and a gamma detector; the triaxial IMU is used to acquire the pitch angle, roll angle and yaw angle of the current node in real time; the gamma detector is used to acquire the gamma value of the current node to reflect the formation lithology parameters.

[0136] The surface measurement end includes a hydraulic pressure sensor and an automatic section / meter counter. The hydraulic pressure sensor is used to collect propulsion pressure values ​​in real time, which are used to indirectly calculate the total propulsion force on the drill pipe. The automatic section / meter counter records the number of meters in real time. By recording the number of drill pipes lowered (sections) and the length of each section, it accurately calculates the current drilling depth and coordinate mapping point.

[0137] The data processing center includes a mechanical calculation unit, a path estimation engine, and a decision unit. The FEA mechanical calculation unit is used to calculate the equivalent stress of the next predicted node of the drill pipe. The path estimation engine is used to obtain the next predicted point set. The decision unit generates a predicted path based on the equivalent stress of the next predicted node of the drill pipe.

[0138] The human-computer interaction terminal provides a visual decision-making interface, including a path deviation dashboard, a stress warning screen, and an expected roll angle command interface. The path deviation dashboard dynamically displays a deviation graph between the designed path and the actual predicted path. The stress warning screen uses different colors to indicate the current fatigue stress level of the drill pipe. The expected roll angle command interface displays the UI block of the "expected roll angle" command to guide manual angle adjustment.

[0139] The system achieves high-precision attitude-lithology sensing through a drilling IMU and gamma detector, obtains the true drilling depth using an automatic section / meter counter combined with gamma correction, and then the FEA (Feature Analysis) mechanical calculation unit and path estimation engine work together to identify and avoid the risk of excessive bending stress in advance during the path planning stage. Finally, the complex decision results are simplified into intuitive stress color codes and roll angle commands through a human-machine interface terminal. This system achieves a dynamic balance between drilling accuracy and drill string safety, significantly reduces the probability of drill pipe breakage, and improves the intelligence level and operational efficiency of directional drilling in complex formations.

[0140] Finally, it should be noted that although the above embodiments have been described in the text and drawings of this application, this should not limit the scope of patent protection of this application. Any technical solutions that are based on the essential concept of this application and utilize the content described in the text and drawings of this application, resulting in equivalent structural or procedural substitutions or modifications, as well as the direct or indirect application of the technical solutions of the above embodiments to other related technical fields, are all included within the scope of patent protection of this application.

Claims

1. A real-time drilling path prediction method, characterized in that, Includes the following steps: The drill pipe parameters, current node coordinates, attitude information, gamma value while drilling, expected drilling progress, and current design plan operation instructions are obtained. The operation instructions include expected roll angle and thrust. The current node coordinates of the drill pipe, the gamma value while drilling, the expected drilling progress, and the current design plan operation commands are input into the drilling path prediction model to obtain a set of prediction candidate points; the set of prediction candidate points has There are 1 time steps, and each time step has 1 time step. There are 10 candidate points, each containing coordinates and its corresponding probability distribution; Based on the coordinates and probabilities of all candidate points in the predicted candidate point set, several candidate paths are generated sequentially according to time steps, and a preliminary predicted path is determined; including the following steps: From each time step Choose one from the candidate points, and sort them in chronological order 1, 2, ... Connect, form Candidate paths; from The candidate paths are selected from the candidate paths with the largest probability product as the candidate paths. Based on the requirements, different strategies are selected to determine the initial prediction path from several candidate paths; Based on the drill pipe parameters, the current node attitude information of the drill pipe, and the preliminary predicted path, the equivalent stress of the next predicted node of the drill pipe is calculated; the drill pipe parameters include the elastic modulus. Cross-sectional area Drill pipe outer diameter ; The calculation of the equivalent stress at the next predicted node of the drill pipe, based on drill pipe parameters, the current node attitude information of the drill pipe, the coordinates of the next predicted node of the drill pipe, and the preliminary predicted path, includes the following steps: Based on the coordinates of the next predicted node of the drill pipe and the preliminary predicted path, the attitude information and actual drilling progress of the next predicted node of the drill pipe are obtained. The curvature of the predicted path is obtained by using the current node attitude information of the drill pipe, the actual drill progress, and the attitude information of the next predicted node of the drill pipe. ; Then the equivalent stress at the next predicted node of the drill pipe : ； in, For overall propulsion; A prediction path is generated based on the equivalent stress of the next predicted node of the drill pipe; the step of generating the prediction path based on the equivalent stress of the next predicted node of the drill pipe includes the following steps: Determine the equivalent stress at the next predicted node of the drill pipe. If the value exceeds a preset threshold, discard the initially predicted path and select a curvature path from several candidate paths. Curvature smaller than the initially predicted path The path that is predicted is used as the predicted path; otherwise, the preliminary predicted path is used as the predicted path.

2. The real-time drilling path prediction method according to claim 1, characterized in that, The path prediction model training includes the following steps: A training sample set is constructed using thousands of sets of correlated data from historical drilling operations; each training sample includes: Input feature sequence: Operation command: Roll angle Propulsion ; Sensor feedback: Pitch angle Heading angle Gamma value during drilling ; Measured coordinates: ; Progress through exploration: ; Output label sequence: Measured path coordinates: ,…, ; Construct a time series prediction model based on the Transformer architecture; The feature sequence is input into a time-series prediction model based on the Transformer architecture; gamma data is used to correct the formation drag coefficient, and the time-series prediction model based on the Transformer architecture is trained to learn the turning efficiency of the drill bit under different formations; the difference between the output and the output label sequence is calculated using the loss function, and the model parameters are updated through backpropagation until the loss converges; the next prediction point set is obtained.

3. The real-time drilling path prediction method according to claim 1, characterized in that, The preliminary predicted path is the path with the smallest cumulative arc length selected from several candidate paths.

4. The real-time drilling path prediction method according to claim 1, characterized in that, The attitude information includes pitch angle. and heading angle ; The curvature of the predicted path : ； in, Indicates the current node's pitch angle; Indicates the heading angle of the current node; Indicates the pitch angle of the next prediction node; Indicates the heading angle of the next prediction node. It indicates that progress has been made in drilling. The average pitch angle between the two nodes.

5. The real-time drilling path prediction method according to claim 1, characterized in that, Using LLM decision layer to perform multi-objective optimization: Equivalent stress at the next predicted node of the drill pipe When the threshold is exceeded, switch to safety priority mode and select a curvature from several candidate paths. Curvature smaller than the initially predicted path The path is used as the predicted path; Equivalent stress at the next predicted node of the drill pipe When within the safe range, switch to correction priority mode and use the preliminary predicted path as the predicted path.

6. The real-time drilling path prediction method according to claim 1, characterized in that, It also includes calculating the deviation vector between the next predicted node coordinates of the drill pipe and the target path based on the next predicted node coordinates of the predicted path; and calculating the roll angle correction.

7. The real-time drilling path prediction method according to claim 6, characterized in that, The roll angle correction is based on the guide vector generated by the drill bit's fixed 1.25-degree eccentricity angle; Calculate the roll angle correction based on the eccentricity direction. : ； in, This represents the deviation vector between the coordinates of the next predicted node of the drill pipe and the target path. This represents the formation coupling coefficient.

8. A real-time drilling path prediction system, employing the real-time drilling path prediction method according to any one of claims 1-7, characterized in that, include: The drilling sensing end includes a three-axis IMU and a gamma detector; The three-axis IMU is used to collect the pitch angle, roll angle and yaw angle of the current node in real time; The gamma detector is used to collect the gamma value of the current node to reflect the lithological parameters of the formation. The surface measurement end includes a hydraulic pressure sensor and an automatic section / meter counter. The hydraulic pressure sensor is used to collect propulsion pressure values ​​in real time, which are used to indirectly calculate the total propulsion force on the drill pipe. The automatic section / meter counter records the number of meters in real time, and accurately calculates the current drilling depth and coordinate mapping point by recording the number of drill rods lowered and the length of a single section. The data processing center includes a mechanical calculation unit, a path estimation engine, and a decision-making unit; the mechanical calculation unit is used to calculate the equivalent stress of the drill pipe at the next predicted node; the path estimation engine is used to obtain the next predicted point set. The decision-making unit generates a prediction path based on the equivalent stress of the next prediction node of the drill pipe; Human-computer interaction terminals are used to provide a visual decision-making interface; Includes a path deviation dashboard, a stress warning screen, and an expected roll angle command interface; The path deviation dashboard dynamically displays the deviation between the designed path and the actual predicted path; the stress early warning screen uses different colors to indicate the current fatigue stress level of the drill pipe. The expected roll angle command interface displays the "Expected Roll Angle" command UI block, guiding manual adjustment of the angle.

Images