Dynamic tool surface measuring method and system based on multi-sensor fusion

Abstract

The invention provides a dynamic tool surface measurement method and system based on multi-sensor fusion, reliable tool surface measurement is realized through three-sensor data fusion and dynamic weight adaptive adjustment, and the method comprises the following steps: measuring the angular velocity of a drilling tool based on a single-axis gyroscope, and calculating GTF and delta GTF; measuring a gravity component based on a triaxial accelerometer, and calculating HTF; measuring a magnetic field component based on a three-axis magnetoresistive sensor, and calculating MTF and delta MTF; working condition modes are divided according to the rotating speed of the drilling tool, the weight of each sensor is dynamically adjusted, a fusion tool face TF = a * HTF + b * MTF + c * GTF is obtained through calculation, when the drilling tool is in a space static state, the drilling tool enters a static mode, and GTF and HTF data are fused; when the rotating speed of the drilling tool is smaller than a first threshold value, entering a low-rotating-speed mode, and fusing data which takes HTF and MTF as main data and GTF as auxiliary data; and when the rotating speed of the drilling tool is greater than or equal to a first threshold value, entering a high-rotating-speed mode, and fusing the fusion data of the low-rotating-speed mode and the incremental data delta GTF and delta MTF.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of drilling engineering, in particular, to a dynamic tool face measurement method and system based on multi-sensor fusion. BACKGROUND

[0002] In the field of oil drilling engineering, the dynamic measurement and control of downhole drilling tool posture (especially the tool face angle) directly determines the accuracy of directional drilling trajectory. In recent years, there are methods to improve measurement accuracy through data processing for rotary steering systems.

[0003] For example, prior art 1: patent application number CN202411545961.0, a multi-sensor fusion downhole drilling tool rotation speed control method and application, through a split drill collar structure, flexible switching of build-up / hold is achieved, and measurement accuracy is improved, but this technology derives a new measurement bottleneck: 1. Efficiency defect of tool face static measurement: tool face angle needs to be measured statically after stopping drilling, relying on three-axis output of accelerometer, causing process interruption. Although prior art 1 realizes drill collar rotation speed control through multi-sensor fusion (gyroscope + magnetoresistance sensor + vibration sensor), it focuses on rotation speed parameters and does not solve the real-time measurement problem of tool face in dynamic drilling; 2. Sensor failure risk in dynamic working conditions: accelerometer is affected by centrifugal force when drilling tool rotates, tool face calculation error > 5°; metal drill collar and downhole geomagnetic field disturbance cause magnetoresistance sensor failure, with magnetic shielding effect, and prior art 1 uses local magnetic field marker to measure rotation speed, but does not overcome magnetic interference in tool face calculation; gyroscope drifts in downhole high-frequency vibration, with vibration cumulative error, and prior art 1 uses vibration compensation, but only for rotation speed parameters; 3. Measurement and control disconnection: sensor is > 8 meters away from drill bit, with spatial delay, and instantaneous posture change at drill bit cannot be captured; control lag: conventional method indirectly calculates dynamic tool face through magnetic / gravity tool face difference angle, formula is complex and depends on calibration, with poor real-time performance (> 2 seconds), causing trajectory correction lag.

[0004] For example, comparative document 2: patent application number CN201610772580.5, a near-bit drilling tool posture while-drilling measurement device and method, its measurement sensors include three-axis accelerometer, three-axis gyroscope, three-axis magnetic sensor, and a temperature sensor, temperature sensor measurement data are used to correct three-axis accelerometer, three-axis gyroscope, and three-axis magnetic sensor measurement data, and corrected three-axis accelerometer and three-axis magnetic sensor measurement data are used to calculate attitude angle, thereby improving near-bit drilling tool posture while-drilling measurement accuracy. Although comparative document 1 realizes multi-sensor integration, its downhole adaptability still has significant defects: 1. Depend on the calibration parameters of the non-magnetic environment before entering the well, such as zero offset and orthogonal error, without considering the real-time compensation demand of dynamic magnetic interference in the downhole; 2. Fixed cutoff frequency filtering is adopted, which cannot adapt to the wideband vibration caused by sudden change of drilling tool speed, and the anti-vibration mechanism is insufficient; 3. Periodic quaternion reset is adopted, and the error accumulation is aggravated in high-speed rotation or strong vibration working conditions, and there is a limitation in the downhole; 4. The differentiated fusion strategy is not designed according to the speed partition (static / low speed / high speed), the weight coefficient is fixed, which cannot cope with complex working conditions such as stick-slip vibration, and cannot realize the working condition self-adaptation. SUMMARY

[0005] The purpose of the present application is to provide a dynamic tool face measurement method and system based on multi-sensor fusion, which realizes reliable tool face measurement through three-sensor data fusion and dynamic weight self-adaptive adjustment.

[0006] The embodiments of the present application are implemented as follows: In one aspect, the embodiments of the present application provide a dynamic tool face measurement method based on multi-sensor fusion, comprising the following steps: Measuring the angular velocity of the drilling tool based on the single-axis gyroscope, calculating the gyroscope tool face GTF and the tool face increment ΔGTF; Measuring the gravity component based on the three-axis accelerometer, calculating the gravity tool face HTF; Measuring the magnetic field component based on the three-axis magnetoresistance sensor, calculating the magnetic tool face MTF and the increment data ΔMTF; Based on the working condition identification of the drilling tool speed and the dynamic weight fusion tool face of the sensor, the fusion tool face TF=a*HTF+b*MTF+c*GTF is calculated, wherein: When the drilling tool is in a space static state, enter the static mode, fuse the gyroscope tool face GTF and the gravity tool face HTF data; When the drilling tool speed is less than the first threshold value, enter the low speed mode, fuse the data mainly with the gravity tool face HTF and the magnetic tool face MTF, and the gyroscope tool face GTF as auxiliary; When the drilling tool speed is greater than or equal to the first threshold value, enter the high speed mode, fuse the fusion data of the low speed mode and the increment data ΔGTF and ΔMTF of the gyroscope tool face and the magnetic tool face.

[0007] In the preferred embodiment of the present application, the above-mentioned first threshold value is 25-35 rpm, and the space static state is triggered by a drilling tool posture recognition algorithm.

[0008] In the preferred embodiment of the present application, the method for calculating the gravity tool face comprises: calculating an initial gravity tool face HTF0, generating HTF1 by combining the rotation speed data of the single-axis gyroscope to compensate for the delay error caused by hardware filtering, and dynamically filtering HTF1 through a sliding window adaptive to the rotation speed of the single-axis gyroscope.

[0009] In the preferred embodiment of the present application, the step for calculating the initial gravity tool face HTF0 is: If the absolute value of A x is less than 0.0001, A x =0.0001; If A x is less than 0, A y is greater than 0, HTF0=arctan ; A y is less than 0, HTF0=2π-arctan ; If A x is greater than 0, A y is greater than 0, HTF0=π-arctan ; A y is less than 0, HTF0=π+arctan .

[0010] In the preferred embodiment of the present application, the formula for calculating the compensated gravity tool face HTF1 is: HTF1=HTF0-(n×t×360°) / (1000×60), wherein n is the rotation speed of the single-axis gyroscope, and t is the delay time.

[0011] In the preferred embodiment of the present application, the dynamic filtering rule of the sliding window is: When n≤60 rpm, the number of window points is 1; When n>60 rpm, the upper and lower limits of the number of window points monotonically increase with n; When the rotation speed mutation rate is greater than 10 rpm / s, the number of window points is reset to 1.

[0012] In the preferred embodiment of the present application, the method for calculating the magnetic tool face MTF is: Initial parameter calibration: calibrate the magnetoresistance sensor in a non-magnetic environment before entering the well to correct the zero offset and orthogonal error; Dynamic adaptive calibration: update the calibration parameters in real time during the operation of the instrument to cope with dynamic environmental factors, including temperature drift, mechanical stress change, and external magnetic field interference.

[0013] In the preferred embodiment of the present application, the dynamic self-adaptive calibration comprises the following steps: A sliding time window buffer is set, and the data weight in the window decreases linearly; Random sampling fitting and least square method are used in sequence for online ellipse fitting, and the magnetoresistance data is updated; The fitted magnetoresistance data is taken as the original data, and the magnetic tool face MTF is calculated according to the formula MTF = arctan

[0014] In the preferred embodiment of the present application, the weight coefficient distribution rules of the gravity, magnetic force and gyro data are as follows: In the static mode, the gravity tool face weight is 0.95-0.99, and the gyro tool face GTF weight is 0.01-0.05; In the low speed mode, the gravity tool face weight is 0.4-0.6, the magnetic tool face MTF weight is 0.3-0.46, and the gyro tool face GTF weight is 0.1-0.14; In the high speed mode, the initial value is the fusion tool face value TF0 of the low speed mode, and the change value of the magnetic tool face MTF and the tool face increment AGTF is superimposed; When the single-axis gyroscope is dominant, the magnetic tool face increment AMTF weight is 0.1, and the tool face increment AGTF weight is 0.9; When the three-axis magnetoresistance sensor is dominant, the magnetic tool face increment AMTF weight is 0.4, and the tool face increment AGTF weight is 0.6.

[0015] In the preferred embodiment of the present application, the weight in the low speed mode is dynamically adjusted, and the dynamic adjustment method comprises: The weight in the low speed mode is defaulted as follows: the gravity tool face weight is 0.6, the magnetic tool face MTF weight is 0.3, and the gyro tool face GTF weight is 0.1; The radial vibration intensity is monitored in real time, and the gravity tool face weight is reduced according to the radial vibration increase amplitude, wherein the gravity tool face weight is greater than or equal to 0.4; meanwhile, the magnetic tool face MTF weight and the gyro tool face GTF weight are increased, and the increment weight ratio of the magnetic tool face MTF and the gyro tool face GTF is 4:1.

[0016] In the preferred embodiment of the present application, under the high speed mode, the magnetic tool face is used for calibration every 3 minutes, and the calibration standard is as follows: when the error of the magnetic tool face MTF and the gyro tool face GTF exceeds the set tolerance, the low speed mode is switched to for calibration.

[0017] In the preferred embodiment of the present application, the dynamic tool face measurement method further comprises a tool face closed loop control step: ​In high speed mode, deceleration is performed to the target speed of 20-30 rpm based on the gyroscope output; At the target speed, PID control is performed with the fused tool face as the feedback value; After reaching the designated tool face, the speed is linearly reduced; The tool face is measured while the speed is linearly reduced until the designated tool face is reached.

[0018] On the other hand, an embodiment of the present invention provides a dynamic tool face measurement system based on multi-sensor fusion, comprising: Single-axis gyroscope measurement and calculation module, which measures the angular velocity of the drill tool based on the single-axis gyroscope and calculates the gyro tool face GTF and tool face increment ΔGTF; The three-axis accelerometer measurement and calculation module calculates the gravity tool face HTF based on the gravity component measured by the three-axis accelerometer; The three-axis magnetoresistive sensor measurement and calculation module calculates the magnetic tool surface MTF and incremental data ΔMTF based on the magnetic field components measured by the three-axis magnetoresistive sensor; The tool face fusion module divides the working mode according to the drilling tool speed, dynamically adjusts the weight of each sensor, and adaptively allocates the weight coefficients of gravity, magnetic and gyroscope data according to preset rules for different working modes. The fused tool face TF is calculated as TF = a*HTF+b*MTF+c*GTF, where: When the drilling tool is in a static state in space, it enters the static mode and fuses the gyro tool face GTF and gravity tool face HTF data; When the drill speed is less than the first threshold, it enters the low speed mode, integrating the data based on the gravity tool surface HTF and magnetic tool surface MTF, supplemented by the gyro tool surface GTF; When the drill speed is greater than or equal to the first threshold, the high speed mode is entered, and the fusion data of the low speed mode and the incremental data ΔGTF and ΔMTF of the gyro tool face and the magnetic tool face are integrated.

[0019] The beneficial effects of the embodiments of the present invention are: 1. For the first time, the data from accelerometers, gyroscopes, and magnetoresistive sensors are integrated to construct a tool face measurement system. This system can address the technical issues of centrifugal force interference on single accelerometers, cumulative errors of gyroscopes under strong vibrations, and failure of magnetic sensors due to geomagnetic interference / magnetic shielding. 2. The magnetic tool face adopts dynamic adaptive calibration to correct the elliptical distortion caused by temperature drift / mechanical stress / magnetic field interference in real time, effectively improving the measurement stability of the magnetic tool face; 3. Adopt adaptive fusion control: intelligent decision-making of working condition partitioning, dynamic weight adjustment mechanism and high-precision closed-loop control process to adaptively suppress centrifugal force and vibration interference and improve data credibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as limiting the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0021] Figure 1 The flow chart of the gravity tool face processing in the dynamic tool face measurement method of the first embodiment of the present application; Figure 2 The flow chart of the tool face fusion processing in the dynamic tool face measurement method of the first embodiment of the present application; Figure 3 The tool face control flow in the dynamic tool face measurement method of the first embodiment of the present application; Figure 4 The functional block diagram of the dynamic tool face measurement system of the second embodiment of the present application; Legend: 110-single axis gyroscope measurement and calculation module; 120-three axis accelerometer measurement and calculation module; 130-three axis magnetoresistance sensor measurement and calculation module; 140-tool face fusion module. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the embodiments of the present application more clear, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings can be arranged and designed in various different configurations.

[0023] Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of the present application.

[0024] It should be noted that: similar reference numbers and letters represent similar items in the following drawings, and therefore, once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0025] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations can be implemented in any

[0026] Moreover, described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the

[0027] The block diagrams in the drawings show only the functionality of the embodiments and do not imply any particular physical or architectural arrangement of the devices, systems, or methods. No inference should be drawn regarding the

[0028] The flow diagrams depicted herein are examples of sequences of operations that can be performed, for example, by a computing device. The depicted examples are not meant to be limiting, as one of skill in the art could readily devise other methods that are equivalent in function to those illustrated. Moreover, the order in which the operations are described is not meant to be limiting, as some of the described operations can be performed in a different order or omitted entirely without departing from the scope of the example embodiments.

[0029] First Embodiment Referring to Figures 1-3 The embodiment provides a dynamic tool face measurement method based on multi-sensor fusion, comprising the following steps: Third, initial calculation of tool face of three sensors 1. Measure the angular velocity of the drilling tool based on the single-axis gyroscope, and calculate the gyroscope tool face GTF and the tool face increment ΔGTF Using the data measured by the single-axis gyroscope, the tool face of the drilling string at the current time is obtained, and the tool face GTF of the gyroscope is obtained.

[0030] The single-axis rate gyroscope directly outputs the angular velocity value around its sensitive axis, and uses the discrete time integral approximation formula: Wherein is the angle value (0-360°), n is the total number of sampling points, i is the nth sampling point, is the sampling time, is the angular velocity. The signal is input into a high-pass filter, and the extracted data is the tool face increment AGTF. Meanwhile, the angular velocity output by the gyroscope may be input data of other sensors.

[0031] 2. Measure the gravity component based on the three-axis accelerometer, and calculate the gravity tool face HTF See Figure 1 the gravity tool face processing flowchart.

[0032] 2.1 Calculate the initial gravity tool face HTF0 The calculation steps of the initial gravity tool face HTF0 are as follows: If the absolute value of A x is less than 0.0001, A x = 0.0001; If A x < 0, A y > 0, then HTF0 = arctan ; A y ≤ 0, then HTF0 = 2π - arctan ; If A x > 0, A y > 0, then HTF0 = π - arctan ; A y ≤ 0, then HTF0 = π + arctan .

[0033] 2.2 Hardware filtering delay compensation To reduce the influence of vibration on the accelerometer, the original data of the accelerometer is processed by hardware filtering. The signal itself is delayed before and after hardware filtering, the delay time is fixed, the delay angle is positively correlated with the rotation speed, and at this time, the tool face compensation needs to be combined with the rotation speed data provided by the gyroscope.

[0034] The calculation formula of the compensated gravity tool face HTF1 is as follows: HTF1 = HTF0 - (n x t x 360°) / (1000 x 60), where n is the single-axis gyroscope rotation speed (unit: rpm), and t is the delay time (unit: ms).

[0035] In this step, the original data of the accelerometer is preprocessed by hardware filtering, which effectively reduces the influence of vibration interference on the measurement accuracy and improves the data stability. At the same time, for the fixed time delay introduced by hardware filtering, dynamic compensation is combined with real-time rotation speed data to eliminate the rotation speed-related angle deviation and ensure the real-time accuracy of the tool face angle.

[0036] 2.3 Sliding window filtering The compensated gravity tool face HTF1 data is filtered by a sliding window, and the window size is determined by the real-time rotation speed provided by the gyroscope. The window size is adaptively adjusted according to the rotation speed. When the rotation speed is less than 60, the window point number is 1, which is used for fast response. When the rotation speed is greater than 60, the window point number increases, which is used for strong noise suppression. Through dynamic noise suppression and smoothing, rotation speed adaptive data smoothing is realized, and the real-time performance and noise suppression ability are balanced.

[0037] Specifically, the relationship between the window size and the rotation speed range data is shown in Table 1.

[0038] Table 1:

[0039] When the rotation speed is in the range of 60 to 100 rpm, the window size is recommended to be set to 6 to 10 data points. In this embodiment, due to the low rotation speed, the actual window size is 5 points. When the rotation speed exceeds 100 r / min, the window size can be selected according to 1 / 10 of the actual rotation speed value, or the average value can be used as the window point number.

[0040] When the rotation speed changes suddenly (change rate ≥ 10 rpm / s), the window size will be forcibly fixed to 1. This setting can accelerate the elimination of old data points and ensure that the window is immediately filled with data under the new rotation speed after the rotation speed stabilizes, avoiding the lag error caused by old data when the rotation speed changes suddenly. When the rotation speed tends to be stable, the window size will start from the minimum value of 1 and gradually accumulate to the target value (for example, 30 points) at a preset rate. Taking the case where the rotation speed jumps from 200 r / min to 300 r / min as an example: during the transient process, the window size is maintained at 1; after the rotation speed stabilizes, the window size increases linearly from 1 to 30, accelerating the adaptation to the new rotation speed state and minimizing the dynamic delay. This dynamic adjustment strategy can effectively alleviate the tool face angle calculation delay caused by sudden changes in rotation speed.

[0041] The gravity tool face calculation process introduces hardware filter delay compensation and sliding window filtering steps, which not only ensures data stability but also significantly improves the dynamic tracking ability of the system to rotation speed changes, especially in the high dynamic environment of downhole drilling tools.

[0042] 3. Measure the magnetic field components based on the three-axis magnetoresistive sensor, calculate the magnetic tool face MTF and the incremental data AMTF Since the downhole magnetic interference is not constant, the type and strength of the magnetic interference may be different at different positions and depths. Therefore, when calculating the tool face using a magnetic sensor, the magnetic sensor data needs to be processed first.

[0043] Ideally, the waveforms of the x-axis and y-axis outputs of the fluxgate are standard sinusoidal waves, and the relationship between x and y is a standard ellipse with the focus center at zero point. Once disturbed, the ellipse will deform and deviate from zero. Based on this, the calculation process of the magnetic tool face MTF is as follows: 3.1 Initial parameter calibration. Before entering the well, the magnetoresistance sensor is calibrated in a non-magnetic environment to correct the zero offset and orthogonal error. 3.2 Dynamic adaptive calibration. The calibration parameters are updated in real time during the instrument operation to cope with dynamic environmental factors, including temperature drift, mechanical stress change, and external magnetic field interference. The dynamic adaptive calibration includes the following steps: 3.21 Set a sliding time window buffer. The data weight in the window decreases linearly. In this step, the data buffer is set to 2s. New data is entered and old data is discarded. Each data point in the window is assigned a weight coefficient. The weight decreases linearly from the current point to the beginning of the window. If the magnetoresistance sensor output suddenly changes, for example, if an interference is detected, the window length is shortened to 1s to accelerate the forgetting of old data. Dynamic filtering of transient interference avoids the influence of historical contaminated data on the current calibration accuracy.

[0044] 3.22 Online ellipse fitting is performed using random sampling fitting and least squares method in turn, and the magnetoresistance data is updated. This step is as follows: Data collection: collect a set of data points in the initial sliding window; Outlier rejection: use the RANSAC algorithm to randomly sample candidate ellipse models, remove outliers through consistency test, and generate a clean data set. This step can significantly improve the robustness of the model, resist noise and outlier interference, and avoid pollution of the initial fitting; Initial ellipse fitting: apply the traditional least squares method to the clean data set for initial ellipse fitting to output the initial parameter vector and covariance matrix. Provide optimal unbiased estimation: obtain the statistically optimal initial parameters and uncertainty measurement on the clean data set.

[0045] Recursive parameter update: for each new data point, update the design matrix and target vector according to the recursive least squares formula to calculate efficiently avoid full re-fitting; apply the normalized parameter or adjust the covariance matrix constraint optimization to the updated parameter to calculate the new parameter. In constraint optimization, you can choose normalization or covariance matrix optimization according to different needs, for example, to prevent solution degeneration, you can force to meet the ellipse constraint by normalization or covariance adjustment; to suppress the matrix ill-conditioned problem, you can choose covariance matrix optimization to ensure numerical stability.

[0046] The above steps can be repeated until the parameters converge or the window data is completed. Adaptive iteration is used to dynamically adjust the parameters until they are stable to ensure the reliability of the results.

[0047] Since the traditional least squares method is sensitive to outlier data, this embodiment first uses RANSAC pre-screening to exclude outliers by random sampling, thereby ensuring the robustness of the initial ellipse fitting; then recursive least squares method is used for initial ellipse fitting, and the normalized parameter / covariance adjustment can ensure the physical reasonableness of the ellipse. It has the advantages of noise resistance and low power consumption, and is suitable for real-time operation of embedded systems.

[0048] At the same time, temperature drift / mechanical stress causes sensor zero drift and other problems. This method uses covariance matrix tracking to monitor fitting uncertainty and automatically assess confidence without interrupting manual recalibration to maintain full well depth calibration accuracy.

[0049] 3.23 The fitted magnetic resistance data is used as the original data, and the magnetic tool face MTF is calculated according to the formula MTF=arctan , and delay compensation and window filtering steps can be added according to the accuracy requirements.

[0050] The entire magnetic tool face MTF calculation process is upgraded from a one-time behavior to a continuous self-optimizing system, realizing a "measurement-fitting-compensation" closed loop and laying the foundation for high-reliability magnetic tool face calculation.

[0051] II. Tool face fusion processing According to the drilling tool speed, the working condition mode is divided, and the weight of each sensor is dynamically adjusted. According to different working condition modes, the weight coefficients of gravity, magnetic force and gyroscope data are adaptively distributed according to the preset rules, and the fused tool face TF=a*HTF+b*MTF+c*GTF is calculated. For the first time, the speed threshold segmentation is combined with vibration feedback to realize the dynamic switching of sensor advantages, and the full-cycle high-reliability tool face output. The processing flow is shown in Figure 2 .

[0052] The tool face fusion processing will be processed according to different downhole working conditions. The real-time output speed of the gyroscope is used as the initial judgment standard: When the input real-time speed of the gyroscope is less than 30 rpm, enter the low-speed mode channel / space static mode channel; When the input real-time speed of the gyroscope is greater than 30 rpm, enter the high-speed mode channel.

[0053] The space static working condition is an independent mode, which is triggered by the real-time detection of the drilling tool attitude recognition algorithm. The space static state is determined based on the variance detection of the drilling tool attitude angle change rate, which has a higher priority than the low-speed channel. Generally, the speed decision threshold is 25-35 rpm, and in this embodiment, the first threshold is selected as the nominal value of 30 rpm.

[0054] When the drilling tool is in a space static state, enter the static mode, only fuse the gyroscopic tool face GTF and the gravity tool face HTF data, use the gravity absolute reference, and completely eliminate the magnetic interference effect. In this mode, TF=a*HTF+c*GTF, the weight coefficient range a∈[0.95, 0.99], and c∈[0.01, 0.05]. The default value of this embodiment is a=0.98 and c=0.02.

[0055] The gyroscopic basic tool face is periodically corrected every 2s, GTF is obtained by calculating the incremental tool face through the gyroscope, and the reference value closed-loop correction is realized.

[0056] When the drilling tool rotation speed is less than the first threshold value, enter the low rotation speed mode, the mode is less affected by the radial centrifugal force, and the data of fusing the gravity tool face HTF and the magnetic tool face MTF as the main part and the gyroscopic tool face GTF as the auxiliary part is considered for accuracy. The centrifugal force interference and the magnetic positioning demand are balanced. In this mode, TF0=a*HTF+b*MTF+c*GTF, the weight coefficient range a∈[0.4, 0.6], b∈[0.3, 0.46], and c∈[0.1, 0.14]. The default values of this embodiment are a=0.6, b=0.3, and c=0.1.

[0057] The radial vibration intensity is monitored in real time, and the gravity tool face weight is reduced according to the radial vibration increase amplitude, wherein the gravity tool face weight is greater than or equal to 0.4, the low speed mode monitors the vibration in real time, the HTF weight is automatically adjusted to the minimum of 0.4 as the vibration increases; at the same time, the magnetic tool face MTF weight and the gyroscopic tool face GTF weight are increased, and the incremental weight ratio of the magnetic tool face MTF and the gyroscopic tool face GTF is 4:1.

[0058] When the drilling tool rotation speed is greater than or equal to the first threshold value, enter the high rotation speed mode, fuse the fusion data of the low rotation speed mode and the incremental data △GTF and △MTF of the gyroscopic tool face and the magnetic tool face, and suppress the HTF failure caused by high-speed centrifugal vibration. In this mode, TF=TF0+b*△MTF+c*△GTF, the initial value is the fusion tool face value TF0 of the low rotation speed mode, and the change value of the use of the magnetic tool face MTF and the tool face increment △GTF is superimposed. In the high speed mode, the incremental calculation is adopted, the original data is avoided to be repeatedly processed, and the computing power can be greatly improved.

[0059] When the single-axis gyroscope is dominant, the magnetic tool face increment △MTF weight is 0.1, and the tool face increment △GTF weight is 0.9. When the three-axis magnetoresistance sensor is dominant, the magnetic tool face increment △MTF weight is 0.4, and the tool face increment △GTF weight is 0.6.

[0060] High-speed mode every 3 minutes, use magnetic tool face calibration, calibration standards: when the magnetic tool face MTF and gyro tool face GTF error exceeds the set tolerance, switch to low-speed mode calibration, only detect amplitude change rate, without complex spectrum analysis. For example, when the original magnetic tool face and gyro tool face error exceeds 10°, select the appropriate time to enter low-speed mode as soon as possible, use the tool face in low-speed mode for calibration, and then enter high-speed mode after calibration.

[0061] III. Tool face adaptive control During drilling, the downhole tool can work in three states: spatially stationary, low-speed rotation, and high-speed rotation through servo control. Among them, the speed of low-speed rotation and high-speed rotation is not uniform, and vibration will always accompany this process. Tool face precise control is mainly used in the spatially stationary phase of the system, i.e. the spatially stationary state, which requires the system to be relatively stationary in space and stop at a specified location, which requires the system to combine speed control to achieve tool face control. Please refer to Figure 3 the working face closed-loop control flowchart, the specific steps of tool face closed-loop control are: In high-speed mode, based on the output of the gyroscope, the speed is reduced to the target speed of 20-30 rpm; At the target speed, PID control is performed with the fused tool face as the feedback value; After reaching the vicinity of the specified tool face, linearly reduce the speed; Linearly reduce the speed while measuring the tool face until the specified tool face is reached. The speed reduction slope is bound to the tool face error, the larger the error, the slower the speed reduction. Real-time fusion data continuously corrects the end point to achieve "zero speed zero deviation" parking, avoiding the 0.5°~3° residual error caused by early braking in traditional solutions.

[0062] For example: when performing tool face control, if in high-speed mode, first reduce the speed to low-speed mode 20 rpm through the gyroscope, then use the tool face fusion module for tool face closed-loop control after speed reduction, and measure the tool face while linearly reducing the speed until the specified tool face is reached.

[0063] Tool face closed-loop control reduces the speed in stages, which releases kinetic energy smoothly and controls the overshoot angle within ±1°, which can solve the problem of tool face overshoot caused by sudden stop when the drilling tool rotates at high speed.

[0064] Second embodiment This embodiment is based on the dynamic tool face measurement method based on multi-sensor fusion provided in the first embodiment, and further provides a dynamic tool face test system, please refer to Figure 4 The system comprises: A single-axis gyroscope measurement calculation module 110 calculates a gyroscope tool face GTF and a tool face increment AGTF based on a single-axis gyroscope measurement of a drill string angular velocity; A three-axis accelerometer measurement calculation module 120 calculates a gravity tool face HTF based on a three-axis accelerometer measurement of a gravity component; A three-axis magnetoresistance sensor measurement calculation module 130 calculates a magnetic tool face MTF and increment data AMTF based on a three-axis magnetoresistance sensor measurement of a magnetic field component; A tool face fusion module 140 dynamically adjusts weights of each sensor according to a drill string rotating speed to divide a working condition mode, and adaptively allocates weight coefficients of gravity, magnetic force and gyroscope data according to a preset rule for different working condition modes to calculate a fusion tool face TF=a*HTF+b*MTF+c*GTF, wherein: When the drill string is in a space static state, a static mode is entered to fuse gyroscope tool face GTF and gravity tool face HTF data; When the drill string rotating speed is less than a first threshold value, a low rotating speed mode is entered to fuse data mainly with the gravity tool face HTF and the magnetic tool face MTF and secondarily with the gyroscope tool face GTF; When the drill string rotating speed is greater than or equal to the first threshold value, a high rotating speed mode is entered to fuse the fusion data of the low rotating speed mode and increment data AGTF and AMTF of the gyroscope tool face and the magnetic tool face.

[0065] A tool face adaptive control module 150 completes accurate positioning control. If in the high rotating speed mode, the drill string needs to be first reduced in speed to the low rotating speed mode 20 rpm through the gyroscope, and after the speed reduction, the tool face fusion module 140 is used for tool face closed-loop control, linear speed reduction is performed after reaching the vicinity of the specified tool face, tool face measurement is performed at the same time, and the specified tool face is reached.

[0066] Third embodiment This embodiment is based on the dynamic tool face measurement method based on multi-sensor fusion provided in the first embodiment, and simulates the following working condition: in the process of drilling a horizontal well section using a rotary steerable drilling system, the drill string is in a high rotating speed working condition (110 rpm) and encounters local magnetic interference, and the interference source is determined as a casing coupling residual magnetic field.

[0067] The initial data of the three sensors are: The three-axis accelerometer (vibration interference): Ax=-0.103g, Ay=0.201g, Az=0.978g; The magnetoresistance sensor (magnetic interference anomaly): Mx=-35.2μT, My=18.6μT, note: the measured magnetic field strength deviates seriously from the theoretical value (expected≈48 μT), and it is determined as a casing residual magnetic field interference; Single-axis gyroscope (instantaneous angular velocity): ω = 11.52 rad / s ≈ 110 rpm.

[0068] Control target: correct the downhole tool face angle from the current 120° orientation to the target angle 185°.

[0069] The specific calculation steps are as follows: One, three-sensor tool face initial calculation 1. Based on the single-axis gyroscope to measure the angular velocity of the drilling tool, calculate the gyroscope tool face GTF and the tool face increment ΔGTF The discrete-time integral approximation formula is , the sampling period is 1 ms, and the cumulative gyroscope tool face GTF is 132.7° in 10 seconds.

[0070] The tool face increment ΔGTF = current GTF - previous period GTF = 132.7° - 120.5° = 12.2° is extracted using a high-pass filter, effectively suppressing cumulative errors.

[0071] Continuous measurement of the gyroscope tool face, by providing continuous angle updates at high speed (110 rpm), supports real-time control.

[0072] 2. Based on the three-axis accelerometer to measure the gravity component, calculate the gravity tool face HTF 2.1 The initial gravity tool face HTF0 is calculated to be 62.3°; 2.2 Hardware filter delay compensation The filter extension time t is 20 ms, and the single-axis gyroscope speed n is 110 rpm.

[0073] HTF1 = HTF0 - (n x t x 360°) / (1000 x 60) = 62.3° - (110 x 20 x 360°) / (1000 x 60) = 49.1°; 2.3 Sliding window filtering According to the correspondence between the gyroscope speed and the sampling points shown in Table 1, when the single-axis gyroscope actually measures the speed n = 110 RPM, the recommended sampling window count range is 10-20 points. This scheme is based on the steady-state working condition to select the middle value n = 15 points for sliding average filtering, and the processed output gravity tool face HTF = 50.2°.

[0074] The initial HTF0=62.3° is calculated under the vibration interference (accelerometer data anomaly). The noise is effectively suppressed by hardware delay compensation (compensation of 13.2°) and sliding window filtering (window point number=15), and the output stable HTF=50.2°. The error is reduced to ±0.9° under the vibration interference, providing a reliable reference for subsequent fusion.

[0075] 3. Measure the magnetic field components based on the three-axis magnetoresistance sensor, calculate the magnetic tool face MTF and the incremental data AMTF Input data: original magnetic survey data (including abnormal interference points) within a 2-second time window; Processing flow: Dynamic window adjustment: the magnetic field mutation My is detected from 22 μT to 18.6 μT, Δ=15.5%, triggering the window to adaptively shrink to 1 second; Robust outlier rejection: RANSAC algorithm (iteration number=500, tolerance=3σ) is used to filter outliers, and the effective sample retention rate is ≥83%; Online ellipse fitting: recursive least squares fitting ellipse parameters, fitting ellipse equation: (Mx+2.1) 2 / 1600+(My-0.8) 2 / 900=1; Magnetic field vector correction: Mx corr =-33.1 μT, My corr =19.8 μT; MTF calculation: MTF=arctan(19.8 / 33.1)=31.0°; The final tool face MTF=180°+31°=211° is determined in combination with the quadrant. Here, the calculation steps of the initial gravity tool face HTF0 are contrary. In this embodiment, Ax=-0.103g, Ay=0.201g.[a1] This step is aimed at magnetic interference (My from 22 μT to 18.6 μT), and dynamic ellipse fitting (RANSAC algorithm) is used: detecting data mutation and adaptively shortening the time window to 1 second; after removing outliers, the ellipse equation is fitted by recursive least squares: (Mx+2.1) 2 / 1600+(My-0.8) 2 / 900=1; the corrected data is output: Mxcorr=-33.1 μT, Mycorr=19.8 μT; finally, the final MTF=211.0° is calculated in combination with the quadrant correction. The data error is reduced from >50% to <10% under the magnetic interference, and the MTF reliability is significantly improved, laying a foundation for the fusion strategy.

[0076] II. Tool face fusion processing According to the drill tool rotation speed 110 rpm > 30 rpm, enter the high rotation speed mode channel, the adaptive weight adjustment mechanism automatically optimizes the reliable data source when the sensor conflicts, and the fusion error is controlled within ±1°.

[0077] The magnetic tool face MTF and gyro tool face GTF error = |211.0°- 132.7°|=78.3°>10°, trigger the magnetic dominant strategy, the magnetic tool face increment ΔMTF weight is 0.4, and the tool face increment ΔGTF weight is 0.6.

[0078] TF=TF0+b*△MTF+c*△GTF =120°+0. 4×(211.0°-205.0°)+0.6×12.2° =120°+2.4°+7.3° =129.7° Three, adaptive control of tool face Tool face closed-loop control, based on gyro feedback to perform rotation speed step switching: 110 rpm to 20 rpm, time-consuming 8s, from high rotation speed mode to low rotation speed mode; Low rotation speed mode fusion (weight adaptation): Vibration increases, a (HTF weight) decreases to 0.45, b (MTF) increases to 0.35, and c (GTF)=0.2; (The weight coefficients are: a=0.6, b=0.3, c=0.1 by default, with the increase of the degree of radial vibration, the input proportion of the magnetic resistance sensor is increased, the weight coefficient a is smaller, and the minimum is not less than 0.4, b and c weights are increased accordingly, and the increase ratio is b:c=2:8.)[a2] TF=a•HTF+b•MTF+c•GTF =0.45×52.1°+0.35×189.2°+0.2×130.5° =130.8°; Precise positioning control: Deviation calculation: |130.8° - 185°| = 54.2°; Linear speed reduction: 20RPM → 0RPM (tool face uniform speed approach target); Final positioning: tool face is stabilized at 184.3° (error <1°).

[0079] During the whole process, the tool face is adjusted from 120° to 184.3°, and the absolute error is only 0.7°, which is significantly better than the industrial standard (usually requires error <5°). The speed reduction strategy effectively eliminates the influence of vibration and ensures the positioning accuracy. This embodiment successfully verifies the effectiveness of the adaptive fusion and control algorithm in harsh working conditions, and provides a technical benchmark for practical industrial applications.

[0080] In combination with the high-precision performance of the calculation process and data results, the scheme exhibits the following core advantages: 1. Excellent robustness and precision: In extreme working conditions (high speed, strong magnetic interference, and severe vibration), the tool face adjustment error is less than 1°, fully demonstrating the high reliability and precision of the algorithm in complex environments. 2. Intelligent adaptive capability: The calculation process integrates dynamic ellipse fitting, RANSAC algorithm, weight dynamic allocation, and high-low speed mode automatic switching strategies. The system can perceive and autonomously respond to interference changes in real time, greatly reducing the need for manual intervention. 3. Optimized multi-sensor fusion: The data of the gravity sensor, magnetometer, and gyroscope have strong complementarity. Through the carefully designed fusion strategy, the advantages of each sensor are effectively coordinated to minimize the overall error. 4. The scheme provides a reusable anti-interference control framework for rotary steerable drilling, especially suitable for scenarios with frequent magnetic interference underground, improving drilling efficiency and safety.

[0081] Those skilled in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the above description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0082] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division, and actual implementation can have another division manner. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices, or units, and can also be electrical, mechanical, or other forms of connection.

[0083] The units described as separate components may or may not be physically separate, and as will be appreciated by one skilled in the art, the units of examples described in connection with the embodiments disclosed herein, and the algorithm steps of each example, can be implemented in electronic hardware, computer software, or any combination thereof. To clearly illustrate the interchangeability of hardware and software, various illustrative components, blocks, and steps have been described above generally in terms of their functionality, and could be implemented with any number of hardware or software components configured to perform the stated functionality, and such examples are well within the scope of the present application. Skilled persons can employ different methods to implement the described functions for each specific application, but such implementation should not be considered to go beyond the scope of the present application.

[0084] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0085] When the integrated unit is realized in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions to make a computer device (which can be a personal computer, a server, or a grid device, etc.) execute all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various program code storage media.

[0086] The above specific embodiments further illustrate the purpose, technical solutions, and beneficial effects of the present application. It should be understood that the above description is only a specific embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A dynamic tool face measurement method based on multi-sensor fusion, characterized in that: The steps include: Based on the single-axis gyroscope to measure the angular velocity of the drill tool, the gyro tool face GTF and tool face increment ΔGTF are calculated; Calculate the gravity tool face HTF based on the gravity component measured by the triaxial accelerometer; The magnetic field components are measured by a three-axis magnetoresistive sensor to calculate the magnetic tool face MTF and incremental data ΔMTF; Based on the working condition identification of the drill speed and the dynamic weight fusion of the sensor, the fused tool face TF is calculated as TF = a*HTF+b*MTF+c*GTF, where: When the drilling tool is in a spatially stationary state, it enters a stationary mode and fuses the gyro tool face GTF and gravity tool face HTF data; When the drill speed is less than the first threshold, the drill enters a low speed mode, integrating data based on the gravity tool face HTF and the magnetic tool face MTF, supplemented by the gyro tool face GTF; When the drill rotation speed is greater than or equal to the first threshold, the high rotation speed mode is entered, and the fusion data of the low rotation speed mode and the incremental data ΔGTF and ΔMTF of the gyro tool face and the magnetic tool face are fused.

2. The dynamic tool face measurement method based on multi-sensor fusion according to claim 1, characterized in that: The first threshold is 25-35 rpm, and the spatial static state is triggered by a drilling tool posture recognition algorithm.

3. The dynamic tool face measurement method based on multi-sensor fusion according to claim 1, characterized in that: The method for calculating the gravity tool face includes: calculating the initial gravity tool face HTF0, compensating for the delay error caused by hardware filtering by combining the rotation speed data of the single-axis gyroscope to generate HTF1, and dynamically filtering HTF1 through a sliding window with adaptive rotation speed of the single-axis gyroscope.

4. The dynamic tool face measurement method based on multi-sensor fusion according to claim 3, characterized in that: The calculation steps of the initial gravity tool face HTF0 are: If A x The absolute value of A is less than 0.0001. x =0.0001; If A x <0, A y > 0, then HTF0 = arctan ; A y ≤ 0, then HTF0 = 2π - arctan ; If A x >0, A y >0, then HTF0 = π - arctan ; A y ≤0, then HTF0 = π + arctan .

5. The dynamic tool face measurement method based on multi-sensor fusion according to claim 3, characterized in that: The calculation formula of the compensated gravity tool face HTF1 is: HTF1=HTF0-(n×t×360°) / (1000×60), where n is the single-axis gyroscope speed and t is the delay time.

6. The dynamic tool face measurement method based on multi-sensor fusion according to claim 4, characterized in that: The dynamic filtering rule of the sliding window is: When n≤60 rpm, the number of window points is 1; When n>60 rpm, the upper and lower limits of the window points increase monotonically with n; When the speed mutation rate is greater than 10 rpm / s, the window points are reset to 1.

7. The dynamic tool face measurement method based on multi-sensor fusion according to claim 1, characterized in that: The calculation method of the magnetic tool face MTF is: Initial parameter calibration: calibrate the magnetoresistive sensor in a non-magnetic environment before entering the well to correct zero bias and orthogonality errors; Dynamic adaptive calibration updates calibration parameters in real time during instrument operation to address dynamic environmental factors such as temperature drift, mechanical stress changes, and external magnetic field interference.

8. The dynamic tool face measurement method based on multi-sensor fusion according to claim 7, characterized in that: The dynamic adaptive calibration includes the following steps: Set a sliding time window buffer, and the weight of the data in the window decreases linearly; Random sampling fitting and least square method are used in turn to perform online ellipse fitting and update the magnetoresistance data; The fitted magnetic resistance data is used as the original data, according to the formula MTF=arctan , calculate the magnetic tool face MTF.

9. The dynamic tool face measurement method based on multi-sensor fusion according to claim 1, characterized in that: The weight coefficient allocation rule for gravity, magnetism and gyroscope data is: In stationary mode, the gravity tool face weight is 0.95-0.99, and the gyro tool face GTF weight is 0.01-0.05; In low-speed mode, the gravity tool face weight is 0.4-0.6, the magnetic tool face MTF weight is 0.3-0.46, and the gyro tool face GTF weight is 0.1-0.14; In high-speed mode, the initial value is the fused tool face value TF0 of the low-speed mode, combined with the usage change values ​​of the magnetic tool face MTF and the tool face increment ΔGTF; When the single-axis gyroscope is dominant, the weight of the magnetic tool face increment ΔMTF is 0.1, and the weight of the tool face increment ΔGTF is 0.9; When the three-axis magnetoresistive sensor is dominant, the weight of the magnetic tool face increment ΔMTF is 0.4, and the weight of the tool face increment ΔGTF is 0.

6.

10. The dynamic tool face measurement method based on multi-sensor fusion according to claim 9, characterized in that: The weight in the low speed mode is dynamically adjusted, and the dynamic adjustment method includes: The default weights for low speed mode are: gravity tool face weight 0.6, magnetic tool face MTF weight 0.3, gyro tool face GTF weight 0.1; Monitor the radial vibration intensity in real time. Reduce the gravity tool face weight according to the increase in radial vibration, where the gravity tool face weight is ≥0.

4. At the same time, increase the magnetic tool face MTF weight and the gyro tool face GTF weight, where the incremental weight ratio of the magnetic tool face MTF and the gyro tool face GTF is 4:

1.

11. The dynamic tool face measurement method based on multi-sensor fusion according to claim 10, characterized in that: In the high speed mode, the magnetic tool face is used for calibration every 3 minutes. The calibration standard is: when the error between the magnetic tool face MTF and the gyro tool face GTF exceeds the set tolerance, the calibration is switched to the low speed mode.

12. The dynamic tool face measurement method based on multi-sensor fusion according to claim 1, characterized in that: It also includes the tool face closed-loop control steps: In high speed mode, deceleration is performed to the target speed of 20-30 rpm based on the gyroscope output; At the target speed, PID control is performed with the fused tool face as the feedback value; After reaching the designated tool face, the speed is linearly reduced; The tool face is measured while the speed is linearly reduced until the designated tool face is reached.

13. A dynamic tool face measurement system based on multi-sensor fusion, characterized in that: include: Single-axis gyroscope measurement and calculation module, which measures the angular velocity of the drill tool based on the single-axis gyroscope and calculates the gyro tool face GTF and tool face increment ΔGTF; The three-axis accelerometer measurement and calculation module calculates the gravity tool face HTF based on the gravity component measured by the three-axis accelerometer; The three-axis magnetoresistive sensor measurement and calculation module calculates the magnetic tool surface MTF and incremental data ΔMTF based on the magnetic field components measured by the three-axis magnetoresistive sensor; The tool face fusion module divides the working mode according to the drilling tool speed, dynamically adjusts the weight of each sensor, and adaptively allocates the weight coefficients of gravity, magnetic and gyroscope data according to preset rules for different working modes. The fused tool face TF is calculated as TF = a*HTF+b*MTF+c*GTF, where: When the drilling tool is in a spatially stationary state, it enters a stationary mode and fuses the gyro tool face GTF and gravity tool face HTF data; When the drill speed is less than the first threshold, the drill enters a low speed mode, integrating data based on the gravity tool face HTF and the magnetic tool face MTF, supplemented by the gyro tool face GTF; When the drill rotation speed is greater than or equal to the first threshold, the high rotation speed mode is entered, and the fusion data of the low rotation speed mode and the incremental data ΔGTF and ΔMTF of the gyro tool face and the magnetic tool face are fused.

Images