Hole opening orientation method and hole opening orientation device

Abstract

The invention discloses a trepanning orientation method and a trepanning orientation device, and relates to the technical field of coal mine underground drilling construction. The method comprises the following steps: acquiring reference attitude data measured by a direction finder at a reference position of a drilling machine; acquiring installation attitude data measured at the actual installation position of the orientation device on the drilling machine; in response to the received reverse installation instruction, performing reverse data conversion processing on the reference attitude data; and according to the reference attitude data or the device angle difference between the data subjected to reverse conversion processing and the installation attitude data, performing compensation calculation on the original attitude data measured by the orientation device in real time at the actual installation position to obtain the actual attitude data of the drilling machine and outputting the actual attitude data. The orientator can be flexibly installed at any position on the drilling machine, the orientator can adapt to different drilling machines without special tools, meanwhile, correct data can be visually read when the orientator is reversely installed, and operation convenience and orienting precision are remarkably improved.

Description

Technical Field

[0001] This application relates to the field of underground drilling construction technology in coal mines, and in particular to a hole-opening orientation method and a hole-opening orientation instrument. Background Technology

[0002] The widespread adoption and application of mechanized coal mining technology has placed higher demands on the drilling accuracy of underground gas drainage holes and water exploration holes. To ensure underground operation safety and borehole success rate, drilling must be carried out strictly according to the designed azimuth and inclination angles, and the drilling trajectory should be kept as consistent as possible with the designed trajectory.

[0003] In existing hole-opening and directional drilling operations, there are two main technical solutions:

[0004] 1. Mechanical calibration method based on physical reference.

[0005] Traditional hole orientation methods often employ mechanical alignment based on physical reference points, such as determining and calibrating the baseline using rulers, angle gauges, or simple guiding tools. These methods involve multiple steps of on-site measurement and drill rig position adjustment, resulting in lengthy stabilization times. Furthermore, the calibration accuracy is highly dependent on the stability of the measurement reference and the standardization of the operation process, making it difficult to meet the demands of high-precision orientation.

[0006] 2. A hole orientation instrument based on gyroscope north-finding.

[0007] To overcome the accuracy limitations of mechanical calibration, borehole orientation instruments based on gyroscope north-finding technology have gradually become the mainstream choice. The basic process is as follows: a north-finding operation is performed using a gyroscope to obtain initial azimuth and inclination data; the orientation instrument is fixed at a specific position on the drilling rig; and the drilling rig is adjusted according to the designed angle.

[0008] However, in practical applications, this type of orientation instrument still has the following technical problems:

[0009] (1) Limited installation location: The orientation instrument usually needs to be fixed to the drill rod or a specific position coaxial with the drill rod by special tooling to ensure that the measurement reference is parallel to the drill axis. Due to the variety of drill types and drill rod specifications, the tooling design to adapt to different drills is complicated, and it is difficult to guarantee absolute parallelism under various installation conditions, which affects the measurement accuracy and ease of operation.

[0010] (2) Limited viewing angle: In complex downhole conditions, the installation position of the directional instrument may be limited by space, causing its display screen to face away from the operator. In this case, the operator cannot directly read the directional instrument data while adjusting the drilling rig, and must walk around to the front of the instrument or repeatedly adjust the installation position, which increases the complexity of operation and affects the efficiency of operation.

[0011] (3) Sensitive to environmental interference: The strong ferromagnetic environment downhole significantly interferes with the measurement accuracy of traditional magnetic sensors. Although gyroscopes can partially avoid this problem, their long-term stability and dynamic response still need further optimization. Summary of the Invention

[0012] This application provides a hole-opening orientation method and a hole-opening orientation instrument, aiming to solve the problems in the prior art where the installation position of the orientation instrument is limited, requiring special tooling to adapt to different drilling rigs, making observation in special working conditions inconvenient, and the measurement accuracy is easily affected by environmental interference, thus affecting the work efficiency and orientation accuracy.

[0013] In a first aspect, a hole orientation method is provided, applied to an orientation instrument, the method being used to determine the actual attitude of the drilling rig at a reference position when the orientation instrument is installed at any non-reference position on the drilling rig, the method comprising:

[0014] Acquire reference attitude data measured by the orientation instrument at the reference position of the drilling rig, wherein the reference position is a position coaxial with the drilling rig;

[0015] Based on whether a reverse installation instruction for the orientation instrument is received, it is determined whether the orientation instrument is installed in reverse. If it is determined to be installed in reverse, the reference attitude data is subjected to reverse conversion processing to obtain the converted reference attitude data.

[0016] Obtain the installation attitude data of the orientation instrument at its actual installation position on the drilling rig;

[0017] Based on the reference attitude data or the converted reference attitude data and the installation attitude data, determine the device angle difference between the actual installation position and the reference position;

[0018] During drilling operations, the original attitude data of the orientation instrument measured in real time at the actual installation location is acquired;

[0019] The original attitude data is compensated based on the device angle difference to obtain the actual attitude data of the drilling rig at the reference position and output.

[0020] Optionally, in the above scheme, the attitude data includes tilt angle and azimuth angle;

[0021] The device angle difference value includes the tilt device angle difference value. Angular difference between azimuth and azimuth device ;in, , Indicates the reference tilt angle. Indicates the tilt angle of the installation position;

[0022] Azimuth device angle difference value Based on the azimuth angle of the actual installation location relative to the reference azimuth The range and magnitude of the difference are determined as follows:

[0023] like , ;

[0024] like Then when hour, ;when hour, .

[0025] Optionally, in the above scheme, the actual attitude data includes the target tilt angle. and target azimuth ;

[0026] in, , This indicates the original tilt angle measured in real time at the installation location by the orientation instrument;

[0027] , This indicates the original azimuth angle measured in real time at the installation location by the orientation instrument. If it is less than 0, add 360. If the value is greater than or equal to 360, then subtract 360.

[0028] Optionally, in the above scheme, based on whether a reverse installation instruction for the orientation instrument is received, it is determined whether the orientation instrument is installed in reverse. If it is determined to be installed in reverse, then a reverse conversion process is performed on the reference attitude data, specifically including:

[0029] In response to the received reverse installation instruction, acquire the display attitude data after the turn;

[0030] The displayed attitude data after steering includes the steering lean angle. , rear azimuth angle and the high side after turning ;

[0031] in, , Indicates the lean angle before steering;

[0032] , Indicates the azimuth angle before turning. Then subtract 360°;

[0033] , Indicates the high edge before the turn.

[0034] Optionally, in the above scheme, the steering operation is triggered by pressing the direction switching button on the orientation instrument.

[0035] Optionally, the above scheme may also include:

[0036] Anomaly detection is performed on the actual pose data using a trained autoencoder model; the autoencoder model is an unsupervised learning model, trained on normal pose data, and is used to reconstruct the input data and calculate the reconstruction error.

[0037] When the reconstruction error exceeds a preset threshold, it is identified as abnormal data, and the weight of the gyroscope in the extended Kalman filter is adjusted to correct the attitude data of subsequent measurements.

[0038] In the above scheme, optionally, the preset threshold includes a first threshold and a second threshold, and the anomaly detection and adjustment of the gyroscope's weights in the extended Kalman filter include:

[0039] When the reconstruction error is less than or equal to the first threshold, the current data is determined to be normal data, and the gyroscope weight is maintained as the baseline weight.

[0040] When the reconstruction error is greater than the first threshold and less than or equal to the second threshold, the current data is determined to be slightly abnormal data. Then, the severity factor of the abnormality is calculated based on the ratio of the reconstruction error to the second threshold, and the weight of the gyroscope is reduced according to the preset rules.

[0041] When the reconstruction error exceeds the second threshold, the current data is determined to be severely abnormal, and the gyroscope weight is reduced to 100 to 500 times the base weight.

[0042] Secondly, a hole orientation instrument is also provided for performing the above-described method, including:

[0043] The acquisition module is used to acquire the reference attitude data of the drilling rig reference position, the installation attitude data of the directional instrument at the actual installation position on the drilling rig, and the original attitude data of the directional instrument measured in real time at the actual installation position.

[0044] The processing module is used to perform reverse data conversion processing on the reference attitude data in response to the received reverse installation instruction, and to perform compensation calculation on the original attitude data according to the device angle difference between the reference attitude data or the converted reference attitude data and the actual installation attitude data to obtain the actual attitude data corresponding to the drilling rig reference position.

[0045] The output module is used to output the actual attitude data.

[0046] Optionally, in the above scheme, the processing module is further configured to:

[0047] In response to a received reverse mounting instruction, the displayed attitude data after steering is acquired; the displayed attitude data after steering includes the steering roll angle, steering azimuth angle, and steering height, and is calculated according to the following rules:

[0048] Caster angle after steering = -inclination angle before steering;

[0049] The azimuth angle after turning = 180 + the azimuth angle before turning. If the result is ≥ 360°, then subtract 360°.

[0050] The height after turning = -the height before turning.

[0051] Optionally, the above scheme also includes an anomaly detection module, which is used to perform anomaly detection on the actual attitude data through a trained autoencoder model. If the detection result is anomaly, the weight of the gyroscope in the extended Kalman filter is adjusted.

[0052] Compared with the prior art, this application has at least the following beneficial effects:

[0053] Based on further analysis and research of existing technical problems, this application recognizes that existing technologies suffer from limitations in the installation location of the orientation instrument, requiring specialized tooling to adapt to different drilling rigs, inconvenient observation under special working conditions, and measurement accuracy being easily affected by environmental interference, thus impacting work efficiency and orientation accuracy. This application addresses these issues by establishing a device angular difference relationship between a reference position and any installation position, and by converting the reference attitude data in response to a reverse installation instruction. Furthermore, it compensates for real-time measurement data based on this device angular difference, enabling the orientation instrument to be flexibly installed at any position on the drilling rig and accurately outputting the actual attitude data of the drilling rig. Simultaneously, it ensures that correct data can still be read intuitively even when the instrument is installed in reverse. Therefore, it can adapt to different drilling rigs without the need for specialized tooling, improving operational convenience and observation ease, thereby increasing work efficiency and orientation accuracy. Attached Figure Description

[0054] Figure 1 This is a schematic flowchart of an opening orientation method provided in one embodiment of this application.

[0055] Figure 2 This is a schematic diagram of an autoencoder neural network structure used for anomaly detection in one embodiment of this application. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0057] In the description of this application, unless otherwise stated, the terms "including", "comprising", "having", etc., also mean "not limited to" (certain units, components, materials, steps, etc.).

[0058] In view of the numerous shortcomings of the existing technology described in the background section, the designers of this application, through in-depth research and meticulous design, combined with their rich experience and cutting-edge achievements in related fields, are committed to developing an innovative solution. The core objective of this application is to provide an advanced algorithm that first determines a reference angle through precise calculation, and then accurately displays the angle relative to that reference position regardless of where the instrument is fixedly installed on the drilling rig. This innovative breakthrough not only breaks the traditional limitation of instrument installation location being confined to the drill rod, but also greatly expands the instrument's applicability and flexibility.

[0059] Furthermore, this application introduces a unique algorithm specifically designed for complex situations where the orientation indicator's displayed direction is opposite to the observer's viewing direction. Even in such cases, the algorithm ensures that the orientation indicator always faces the observer and accurately acquires the actual tilt and azimuth data. This technological innovation significantly improves the accuracy and convenience of measurements, providing operators in related industries with a more efficient and reliable measurement tool, greatly enhancing work efficiency and user experience.

[0060] In one embodiment, a hole orientation method is provided, applied to an orientation instrument, the method being used to determine the actual attitude of the drilling rig at a reference position when the orientation instrument is installed at any non-reference position on the drilling rig, the method comprising:

[0061] Acquire reference attitude data measured by the orientation instrument at the reference position of the drilling rig, wherein the reference position is a position coaxial with the drilling rig;

[0062] Based on whether a reverse installation instruction for the orientation instrument is received, it is determined whether the orientation instrument is installed in reverse. If it is determined to be installed in reverse, the reference attitude data is subjected to reverse conversion processing to obtain the converted reference attitude data.

[0063] Obtain the installation attitude data of the orientation instrument at its actual installation position on the drilling rig;

[0064] Based on the reference attitude data or the converted reference attitude data and the installation attitude data, determine the device angle difference between the actual installation position and the reference position;

[0065] During drilling operations, the original attitude data of the orientation instrument measured in real time at the actual installation location is acquired;

[0066] The original attitude data is compensated based on the device angle difference to obtain the actual attitude data of the drilling rig at the reference position and output.

[0067] This application aims to address the limitations of existing orientation instruments in fixed positions by proposing an innovative data conversion method. This method, through advanced algorithms and technologies, effectively overcomes the limitations of traditional fixed orientation instrument methods, enabling accurate measurement and data conversion at different positions and angles. This significantly improves the flexibility and practicality of the orientation instrument, providing a more efficient and accurate solution for orientation measurement in coal mine drilling and other related fields. To achieve the above objectives, the technical solution adopted in this application is as follows:

[0068] A data conversion method for the reference installation position conversion algorithm described in this application is provided. The algorithm consists of three key steps:

[0069] Step 1: Confirm the reference position.

[0070] After completing the gyro north-finding operation, the first step is to confirm the reference position. The operator precisely places the equipment coaxially with the drilling rig, using this as the reference point. At this time, the equipment records the inclination and azimuth data of this reference position, which will be used as the reference data for subsequent calculations.

[0071] Step 2: Confirm the installation location.

[0072] Next, the orientation instrument is placed in another controllable and easily accessible location on the drilling rig. After the operator confirms that the orientation instrument is installed in this location, the system will record relevant information about the installation location, providing basic data for subsequent calculations.

[0073] Step 3: Data conversion and compensation calculation.

[0074] Through complex calculations using internal algorithms, the system compensates for the inclination and azimuth angles of the current installation position based on baseline data. Ultimately, the calculation results reflect the actual inclination and azimuth angles of the drilling rig's baseline. During this process, if the orientation instrument is fixed in the opposite direction to the observer's viewpoint, the operator can rotate the instrument to the observer's viewpoint and press the direction switch button. The equipment will then display the compensated inclination and azimuth angle data, ensuring the operator can accurately read the measurement results. Simultaneously, machine learning is used to determine the accuracy of the algorithm's output data; if the algorithm's output is inaccurate or significantly different, the calculation is recalculated.

[0075] Figure 1The diagram shows the relevant steps. This data conversion method not only improves the installation flexibility of the orientation instrument, but also ensures the accuracy and reliability of the measurement data through precise algorithm compensation, greatly improving the convenience of operation and work efficiency.

[0076] In one embodiment, the attitude data includes tilt angle and azimuth angle;

[0077] The device angle difference value includes the tilt device angle difference value. Angular difference between azimuth and azimuth device ;in, , Indicates the reference tilt angle. Indicates the tilt angle of the installation position;

[0078] Azimuth device angle difference value Based on the azimuth angle of the actual installation location relative to the reference azimuth The range and magnitude of the difference are determined as follows:

[0079] like , ;

[0080] like Then when hour, ;when hour, .

[0081] In one embodiment, the actual attitude data includes the target tilt angle. and target azimuth ;

[0082] in, , This indicates the original tilt angle measured in real time at the installation location by the orientation instrument;

[0083] , This indicates the original azimuth angle measured in real time at the installation location by the orientation instrument. If it is less than 0, add 360. If the value is greater than or equal to 360, then subtract 360.

[0084] The algorithm for converting installation location data at the reference location can be roughly summarized as follows:

[0085] The calculation method for the orientation instrument to be oriented at any position of the drilling rig is as follows:

[0086] Assuming the drilling rig's reference inclination angle is azimuth angle is ;

[0087] The actual installation location tilt angle is azimuth angle is ;

[0088] The original dip angle value read during drilling rig orientation is The original value of the azimuth angle is ;

[0089] A. The difference in device angle between the actual installation position and the reference position;

[0090] a. Inclination device angular difference value (Values ​​can be positive or negative, and the range is within...) (The upward tilt of the installation position is considered positive).

[0091] b. Azimuth device angle difference value :( represents positive or negative values, ranging from...) (The installation position is slightly to the right; this is considered positive).

[0092] like , ;

[0093] like Then when hour, ;when hour, .

[0094] B. The reference attitude value of the drilling rig displayed at the installation position during drilling rig orientation;

[0095] a. Drilling rig reference inclination angle value ;

[0096] b. Drilling rig reference azimuth angle value;

[0097] If the value is less than 0, add 360; if the value is greater than or equal to 360, subtract 360.

[0098] Based on the above algorithm, the measured value after benchmark compensation is obtained.

[0099] In one embodiment, based on whether a reverse installation instruction for the orientation instrument is received, it is determined whether the orientation instrument is installed in reverse. If it is determined to be installed in reverse, a reverse conversion process is performed on the reference attitude data, specifically including:

[0100] In response to the received reverse installation instruction, acquire the display attitude data after the turn;

[0101] The displayed attitude data after steering includes the steering lean angle. , rear azimuth angle and the high side after turning ;

[0102] in, , Indicates the lean angle before steering;

[0103] , Indicates the azimuth angle before turning. Then subtract 360°;

[0104] , Indicates the high edge before the turn.

[0105] In this embodiment, "high side" refers to the rotation angle of the orientation instrument's own coordinate system relative to the gravity high side direction when it is tilted. Specifically, when the drilling rig has a certain tilt angle, the highest point on the drilling rig's cross-section relative to the gravity direction is defined as the high side direction, and the high side angle measured by the orientation instrument is the angle between its reference axis and this high side direction. The high side angle, tilt angle, and azimuth angle together constitute a complete description of the drilling rig's attitude. When the orientation instrument undergoes a physical turn, its high side angle measurement value changes accordingly, requiring corresponding compensation through algorithms to ensure the accuracy of the output data.

[0106] The orientation instrument's reverse data algorithm, namely:

[0107] After the orientation instrument is turned (i.e., pointing in the opposite direction of the laser window), the displayed tilt angle, azimuth angle, and elevation should change. The calculation method is as follows:

[0108] Assuming the lean angle before steering is azimuth angle is The height is ;

[0109] The angle of inclination after turning is azimuth angle is The height is (Scope) ),but:

[0110] A. Inclination angle ;

[0111] B. Azimuth ,like Subtract 360° from the time;

[0112] C. High side .

[0113] In one specific embodiment, when the operator physically rotates the orientation device 180 degrees so that its screen faces them, the instrument's internal processing unit responds to the reverse installation instruction and performs reverse data conversion processing. The conversion rule for high-angle features is as follows: In other words, the converted high angle is equal to the negative of the original high angle. This conversion ensures that the displayed high angle can still accurately reflect the actual high tool face angle of the drilling rig after the instrument has been physically rotated, avoiding misreading caused by instrument rotation.

[0114] In one embodiment, the steering operation is triggered by pressing the direction switching button on the orienteering device.

[0115] In one embodiment, it also includes:

[0116] Anomaly detection is performed on the actual pose data using a trained autoencoder model; the autoencoder model is an unsupervised learning model, trained on normal pose data, and is used to reconstruct the input data and calculate the reconstruction error.

[0117] When the reconstruction error exceeds a preset threshold, it is identified as abnormal data, and the weight of the gyroscope in the extended Kalman filter is adjusted to correct the attitude data of subsequent measurements.

[0118] In one embodiment, the preset threshold includes a first threshold and a second threshold, and the anomaly detection and adjustment of the gyroscope's weights in the extended Kalman filter include:

[0119] When the reconstruction error is less than or equal to the first threshold, the current data is determined to be normal data, and the gyroscope weight is maintained as the baseline weight.

[0120] When the reconstruction error is greater than the first threshold and less than or equal to the second threshold, the current data is determined to be slightly abnormal data. Then, the severity factor of the abnormality is calculated based on the ratio of the reconstruction error to the second threshold, and the weight of the gyroscope is reduced according to the preset rules.

[0121] When the reconstruction error exceeds the second threshold, the current data is determined to be severely abnormal, and the gyroscope weight is reduced to 100 to 500 times the baseline weight.

[0122] In one embodiment, reducing the gyroscope weight according to a preset rule includes: according to the formula Adjust the process noise variance matrix of the gyroscope in the extended Kalman filter, where Indicates the baseline variance. This represents the severity factor of the anomaly, where K represents a preset coefficient, and the preset coefficient K ranges from 5 to 10; Error represents the reconstruction error. This represents the second threshold.

[0123] In one embodiment, the autoencoder model employs an unsupervised learning approach. The training set is used to learn the normal variation patterns of the data, obtained after compensation calculations from the orientation instrument at different installation positions during normal attitude adjustment of a coal mine drilling rig. In this embodiment, mean squared error (MSE) is used as the metric for reconstruction error. To distinguish different degrees of anomaly, a first threshold is preset. Second threshold .

[0124] In the actual testing process:

[0125] When reconstruction error At this point, the current data is determined to be normal. The data characteristics at this time are smooth / stable, synchronized with manual adjustments.

[0126] When reconstruction error At this point, the current data is determined to be slightly anomalous. The data exhibits minor, causeless jumps due to minor disturbances, but these jumps generally align with the adjustment actions.

[0127] When reconstruction error At this point, the current data is determined to be severely abnormal. This occurs when the data experiences significant fluctuations due to strong interference, completely disconnected from the adjustment actions.

[0128] Gyroscope EKF Weight Adjustment Strategy: The gyroscope weights are reflected in the process noise variance matrix of the Extended Kalman Filter (EKF). In the middle, the gyroscope weight and Inversely proportional. Set a baseline variance. .

[0129] The specific adjustment strategy is as follows:

[0130] When data is determined to be normal, maintain. Maintain a high weight for the gyroscope to ensure real-time attitude adjustment.

[0131] When data is identified as slightly anomalous, the anomalous severity factor is calculated. , and according to Dynamic weighting is applied, where K is a preset coefficient ranging from 5 to 10. This strategy strikes a balance between conformity and data smoothness.

[0132] When the data is determined to be severely abnormal, Set as The weight of the gyroscope is reduced by 100 to 500 times to eliminate distorted data. At this time, only the stable data of the accelerometer is fused to ensure that the displayed value is the true angle of the drilling rig.

[0133] The main processing unit performs neural network data statistical analysis on the azimuth and tilt data after the gyroscope has found north. An automatic encoder identifies abnormal data, thereby detecting the instrument's tilt and azimuth status in real time and verifying the accuracy of the relevant judgment algorithms. The principle is as follows:

[0134] An autoencoder, also known as an autoencoder, is an unsupervised learning model. Essentially, it uses a neural network to generate a low-dimensional representation of a high-dimensional input. Autoencoders are similar to Principal Component Analysis (PCA), but they overcome the linearity limitation of PCA when using non-linear activation functions.

[0135] The autoencoder consists of two main parts: the encoder and the decoder. The encoder's role is to discover a compressed representation of the given data, while the decoder is used to reconstruct the original input. During training, the decoder forces the autoencoder to select the most informative features, which are then stored in the compressed representation. The final compressed representation resides in the intermediate encoder layer.

[0136] by Figure 2 For example, the original data has a dimension of 10. The encoder and decoder each have two layers, and the coder in the middle has three nodes. This means that the original data is reduced to only 3 dimensions. The decoder then reconstructs the original data based on the dimensionality-reduced data, obtaining a 10-dimensional output again. In this process from input to output, the autoencoder also plays a role in noise reduction.

[0137] Anomaly detection is generally divided into supervised and unsupervised approaches. In unsupervised cases, we lack anomalous samples for learning, and the algorithm's basic assumption is that outliers follow a different distribution. An autoencoder trained on normal data can reconstruct normal samples, but it cannot reconstruct data points that deviate from the normal distribution well, resulting in a large reconstruction error.

[0138] If the features of the sample are all numerical variables, we can use MSE or MAE as the reduction error. For example Figure 2 If the input sample is:

[0139] ;in, Indicates the input sample. Let i represent the i-th input sample, i = 1, 2, 3, ..., 10.

[0140] The result after reconstruction by Autoencoder is:

[0141] ; or This represents the reconstructed output sample.

[0142] The restoration error MSE is:

[0143] ;

[0144] The restoration error MAE is:

[0145] ;

[0146] When the restoration error exceeds a certain threshold, we mark it as an outlier.

[0147] Based on the above algorithm, in a static state, the self-north-finding accuracy is trained by inputting the reference azimuth value, and an overall reconstruction model (3-layer neural network, judged once every 10 data points) is trained. When the reference azimuth value is abnormal, the overall autoencoder cannot reconstruct the normal data, indicating that the data is abnormal. The weight of the overall gyroscope inside the EKF will be adjusted accordingly to ensure the accuracy of the azimuth.

[0148] To address the two core issues mentioned in the background technology—"limited installation location of the orientation instrument, requirement for special tooling, and difficulty in ensuring parallelism with the drill pipe" and "inability to read data when the orientation instrument faces away from the operator under special working conditions"—the technical solution of this application systematically solves these problems through the following methods:

[0149] Firstly, addressing the limitations of installation location and the complexity of tooling, the core breakthrough of this solution lies in abandoning the traditional notion that "the orientation instrument must be fixed coaxially with the drill pipe." Instead, it utilizes a "reference installation position conversion algorithm" to achieve installation at any location. Specifically, the operator only needs to calibrate the equipment once at a reference position coaxial with the drilling rig and record the reference attitude data. Subsequently, the orientation instrument can be flexibly installed on the drilling rig at any easily fixed location. During subsequent measurements, the system automatically acquires real-time data of the installation position and uses a built-in algorithm to calculate the device angular difference between the current installation position and the reference position. This accurately compensates for the measured values, ultimately displaying the operator the converted data representing the actual attitude of the drilling rig. This method completely eliminates the reliance on specialized tooling; even if the orientation instrument is not parallel to the drill pipe, the algorithm ensures the accuracy of the measurement results.

[0150] Secondly, to address the issue of limited viewing angles, the solution introduces a "direction instrument reverse data algorithm." When the direction instrument must be placed with its back to the operator due to space constraints, the operator can simply rotate the direction instrument 180 degrees so that its screen faces them, and then press the direction switch button. At this time, the internal algorithm of the device will immediately convert the displayed data: invert the tilt angle, rotate the azimuth angle 180 degrees and zero it, and also invert the elevation data. In this way, although the physical orientation of the instrument has changed, the value displayed on the screen is still the actual correct orientation of the drilling rig, ensuring that the operator can intuitively and in real time read accurate data while adjusting the drilling rig, without having to go behind the instrument or repeatedly adjust the installation position.

[0151] Furthermore, to ensure that the two algorithms can output highly reliable data under various complex operating conditions, the solution also constructs an anomaly detection and correction mechanism based on an autoencoder neural network. When the instrument is stationary, the system trains a reconstruction model using a large amount of normal data. In actual operation, this model continuously monitors the compensated attitude data. Once it detects that the data features differ significantly from the normal model (i.e., the reconstruction error exceeds the standard), it determines that the current data is abnormal and adjusts the weights of the gyroscope in the data fusion algorithm (such as extended Kalman filtering) in real time, correcting the data through iterative calculations. This mechanism provides a guarantee of accuracy for the first two algorithms, ensuring that the final output azimuth and tilt data remain reliable even in dynamic environments or under interference.

[0152] In summary, this application addresses the issues of installation flexibility through "reference position conversion," the issues of observation convenience through "reverse data algorithm," and the issues of data accuracy through "neural network anomaly detection." The combination of these three approaches systematically addresses all the shortcomings mentioned in the background technology.

[0153] In one embodiment, an aperture orientation instrument is provided for performing the method provided in the above embodiments, including:

[0154] The acquisition module is used to acquire the reference attitude data of the drilling rig reference position, the installation attitude data of the directional instrument at the actual installation position on the drilling rig, and the original attitude data of the directional instrument measured in real time at the actual installation position.

[0155] The processing module is used to perform reverse data conversion processing on the reference attitude data in response to the received reverse installation instruction, and to perform compensation calculation on the original attitude data according to the device angle difference between the reference attitude data or the converted reference attitude data and the actual installation attitude data to obtain the actual attitude data corresponding to the drilling rig reference position.

[0156] The output module is used to output the actual attitude data.

[0157] The specific implementation details of each module can be found in the above description of the hole orientation method, and will not be repeated here.

[0158] In one embodiment, the processing module is further configured to:

[0159] In response to a received reverse mounting instruction, the displayed attitude data after steering is acquired; the displayed attitude data after steering includes the steering roll angle, steering azimuth angle, and steering height, and is calculated according to the following rules:

[0160] Caster angle after steering = -inclination angle before steering;

[0161] The azimuth angle after turning = 180 + the azimuth angle before turning. If the result is ≥ 360°, then subtract 360°.

[0162] The height after turning = -the height before turning.

[0163] In one embodiment, an anomaly detection module is further included, which is used to perform anomaly detection on the actual attitude data using a trained autoencoder model, and if the detection result is anomaly, adjust the weight of the gyroscope in the extended Kalman filter.

[0164] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A hole-opening orientation method, applied to an orientation instrument, characterized in that, The method is used to determine the actual attitude of the drilling rig at a reference position when the orientation instrument is installed at any non-reference position on the drilling rig, the method comprising: Acquire reference attitude data measured by the orientation instrument at the reference position of the drilling rig, wherein the reference position is a position coaxial with the drilling rig; Based on whether a reverse installation instruction for the orientation instrument is received, it is determined whether the orientation instrument is installed in reverse. If it is determined to be installed in reverse, the reference attitude data is subjected to reverse conversion processing to obtain the converted reference attitude data. Obtain the installation attitude data of the orientation instrument at its actual installation position on the drilling rig; Based on the reference attitude data or the converted reference attitude data and the installation attitude data, determine the device angle difference between the actual installation position and the reference position; During drilling operations, the original attitude data of the orientation instrument measured in real time at the actual installation location is acquired; The original attitude data is compensated based on the device angle difference to obtain the actual attitude data of the drilling rig at the reference position and output.

2. The method according to claim 1, characterized in that, The attitude data includes tilt angle and azimuth angle; The device angle difference value includes the tilt device angle difference value. Angular difference between azimuth and azimuth device ;in, , Indicates the reference tilt angle. Indicates the tilt angle of the installation position; Azimuth device angle difference value Based on the azimuth angle of the actual installation location relative to the reference azimuth The range and magnitude of the difference are determined as follows: like , ; like Then when hour, ;when hour, .

3. The method according to claim 2, characterized in that, The actual attitude data includes the target tilt angle. and target azimuth ; in, , This indicates the original tilt angle measured in real time at the installation location by the orientation instrument; , This indicates the original azimuth angle measured in real time at the installation location by the orientation instrument. If it is less than 0, add 360. If the value is greater than or equal to 360, then subtract 360.

4. The method according to claim 1, characterized in that, Based on whether a reverse installation instruction for the orientation instrument is received, it is determined whether the orientation instrument is installed in reverse. If it is determined to be installed in reverse, a reverse conversion process is performed on the reference attitude data, specifically including: In response to the received reverse installation instruction, acquire the display attitude data after the turn; The displayed attitude data after steering includes the steering lean angle. , rear azimuth angle and the high side after turning ; in, , Indicates the lean angle before steering; , Indicates the azimuth angle before turning. Then subtract 360°; , Indicates the high edge before the turn.

5. The method according to claim 4, characterized in that, The steering operation is triggered by pressing the direction switching button on the orienteering device.

6. The method according to claim 1, characterized in that, Also includes: Anomaly detection is performed on the actual pose data using a trained autoencoder model; the autoencoder model is an unsupervised learning model, trained on normal pose data, and is used to reconstruct the input data and calculate the reconstruction error. When the reconstruction error exceeds a preset threshold, it is identified as abnormal data, and the weight of the gyroscope in the extended Kalman filter is adjusted to correct the attitude data of subsequent measurements.

7. The method according to claim 6, characterized in that, The preset threshold includes a first threshold and a second threshold, and the anomaly detection and adjustment of the gyroscope's weights in the extended Kalman filter include: When the reconstruction error is less than or equal to the first threshold, the current data is determined to be normal data, and the gyroscope weight is maintained as the baseline weight. When the reconstruction error is greater than the first threshold and less than or equal to the second threshold, the current data is determined to be slightly abnormal data. Then, the severity factor of the abnormality is calculated based on the ratio of the reconstruction error to the second threshold, and the weight of the gyroscope is reduced according to the preset rules. When the reconstruction error exceeds the second threshold, the current data is determined to be severely abnormal, and the gyroscope weight is reduced to 100 to 500 times the base weight.

8. A hole orientation instrument for performing the method according to claims 1-6, characterized in that, include: The acquisition module is used to acquire the reference attitude data of the drilling rig reference position, the installation attitude data of the directional instrument at the actual installation position on the drilling rig, and the original attitude data of the directional instrument measured in real time at the actual installation position. The processing module is used to perform reverse data conversion processing on the reference attitude data in response to the received reverse installation instruction, and to perform compensation calculation on the original attitude data according to the device angle difference between the reference attitude data or the converted reference attitude data and the actual installation attitude data to obtain the actual attitude data corresponding to the drilling rig reference position. The output module is used to output the actual attitude data.

9. The orientation device according to claim 8, characterized in that, The processing module is also used for: In response to a received reverse mounting instruction, the displayed attitude data after steering is acquired; the displayed attitude data after steering includes the steering roll angle, steering azimuth angle, and steering height, and is calculated according to the following rules: Caster angle after steering = -inclination angle before steering; The azimuth angle after turning = 180 + the azimuth angle before turning. If the result is ≥ 360°, then subtract 360°. The height after turning = -the height before turning.

10. The orientation device according to claim 8, characterized in that, It also includes an anomaly detection module, which is used to detect anomalies in the actual attitude data using a trained autoencoder model. If the detection result is an anomaly, the weight of the gyroscope in the extended Kalman filter is adjusted.

Images