A method for correcting probe tube deflection in drilling trajectory measurement

Abstract

The present invention discloses a method for drilling trajectory measurement probe deflection correction, firstly, data curve fitting is performed on the measured raw data, then correlation analysis and processing are performed based on the inclination angle, azimuth angle data fitting curve and the tool face angle fitting curve, after correlation processing, a suitable correction coefficient is selected for curve correction, and the drilling trajectory data processed by the correction method is closer to the true value of the drilling trajectory. The present invention can perform continuous trajectory correction on the trajectory to avoid the situation of missing the target or greatly adjusting the actual drilling trajectory. At the same time, the present invention corrects the actual drilling trajectory in advance, which is conducive to safe drilling construction and improves the construction target rate.

Description

Technical Field

[0001] The invention belongs to the technical field of drilling trajectory measurement, and in particular relates to a method for correcting the deflection of a probe tube used for drilling trajectory measurement. Background Art

[0002] Under ideal conditions, the measured drilling trajectory coincides with the designed trajectory, which is the best state for drilling construction. In the actual drilling construction process, due to the influence of various factors, there is usually an error between the actual drilling parameters and the designed parameters, and the drilling trajectory is not an ideal straight line, but a curve, which causes the actual drilling trajectory to deviate from the designed trajectory. The data correction of the attitude measurement unit of the drilling trajectory measurement system is an important guarantee for the accurate drawing of the drilling trajectory. There are problems in the data calculation of the existing drilling trajectory system, such as bad attitude points, uncorrected measurement data, and low calculation accuracy. These problems need to be studied. Summary of the invention

[0003] In view of the defects and shortcomings in the prior art, the present invention provides a method for correcting the deflection of a probe tube for measuring a drilling trajectory, so as to solve the problem that in the prior art, during the drilling construction, due to the influence of factors such as drilling tool conditions, a certain angle gap will be generated at the connection between the drill rods and the drill rods, which will cause the trajectory deviation of the drilling during the drilling process. The information of the drilling inclination, azimuth and tool face angle measured by the drilling trajectory measurement equipment also brings such influencing factors into the original measurement data, resulting in the problem that the measurement data often has a small mutation at adjacent measurement points.

[0004] To achieve the above object, the present invention adopts the following technical scheme:

[0005] A method for correcting the deflection of a probe tube for drilling trajectory measurement, comprising correction of inclination data and correction of azimuth data;

[0006] The inclination data correction includes: drawing a depth-inclination curve and obtaining a depth-inclination data curve after interpolation fitting, drawing a depth-tool face angle curve and obtaining a depth-tool face angle data curve after interpolation fitting, analyzing and judging the correlation between the fitted depth-inclination data curve and the fitted depth-tool face angle data curve and aligning the curves, calculating a correction coefficient after the curves are aligned and obtaining a preliminary corrected depth-inclination curve, and interpolating a high-order fitting of the preliminary corrected depth-inclination curve to obtain a true depth-inclination value of the drilling trajectory;

[0007] The azimuth data correction includes: drawing a depth-azimuth curve and obtaining a depth-azimuth data curve after interpolation fitting, drawing a depth-tool face angle curve and obtaining a depth-tool face angle data curve after interpolation fitting, analyzing and judging the correlation between the fitted depth-azimuth data curve and the fitted depth-tool face angle data curve and aligning the curves, calculating the correction coefficient after the curves are aligned and obtaining a preliminary corrected depth-azimuth curve, and interpolating the preliminary corrected depth-azimuth curve for high-order fitting to obtain a true depth-azimuth value of the drilling trajectory.

[0008] The present invention also includes the following technical features:

[0009] Optionally, the inclination data correction specifically includes the following steps:

[0010] Step A1, drawing a depth-inclination data curve: arranging the inclination data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-inclination data, and drawing a depth-inclination data curve according to the depth-inclination data;

[0011] Step A2, data curve fitting: using the least square method to fit the depth-inclination data curve to obtain a fitted depth-inclination data curve, denoted as Q;

[0012] Step A3, drawing a depth-tool face angle data curve: arranging the tool face angle data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-tool face angle data, and drawing a depth-tool face angle data curve, which is recorded as G;

[0013] Step A4, data curve fitting: the least square method is used to fit the depth-tool face angle data curve to obtain a fitted depth-tool face angle data curve, and the cosine function value of the fitted depth-tool face angle data is taken to generate a fitted depth-tool face angle function curve, which is recorded as COS(G);

[0014] Step A5, correlation analysis: The correlation coefficient r between the fitted depth-inclination data curve and the fitted depth-tool face angle function curve is:

[0015]

[0016] In the above formula, E{Q} is the expected value of the fitted depth-inclination data curve Q, E{COS(G)} is the expected value of the fitted depth-tool roll angle function curve COS(G), E{(QE{Q})×(COS(G)-E{COS(G)})} is the covariance between the fitted depth-inclination data curve and the fitted depth-tool roll angle function curve data, is the standard deviation of the fitted depth-inclination data curve Q, is the standard deviation of the fitted depth-tool face angle function curve COS(G); the closer the |r| value is to 1, the greater the correlation is; the correlation coefficient is a dimensionless measure to describe the correlation between two series of data;

[0017] Step A6, curve alignment: if the |r| value is between 0.8 and 1, the data curve is aligned, and the time series data is converted into function data and then time correction is performed to align the depth-inclination data curve and the depth-tool face angle function curve;

[0018] Step A7, calculate the correction coefficient and obtain the preliminary correction depth-inclination value: calculate the product of each inclination value in the depth-inclination data after the curve is aligned and each tool face angle cosine value in the depth-tool face angle function data to obtain the depth-correction curve data. 1 ×COS(G 1 ),q 2 ×COS(G 2 ),…,q n ×COS(G n ), which is the correction coefficient f at different depths; multiply each inclination value in the depth-inclination curve data by the corresponding correction coefficient to obtain fq 1 ,fq 2 ,…,fq n , thereby obtaining a preliminary corrected depth-inclination curve;

[0019] Step A8, performing high-order fitting on the preliminary corrected depth-inclination curve: For the preliminary corrected depth-inclination curve obtained in step A7, use the least squares method to perform high-order interpolation fitting, and the depth-inclination curve after the high-order interpolation fitting is the actual depth-inclination value of the borehole.

[0020] Optionally, the step A6 of aligning the curves includes:

[0021] Step A61. Functionalize discrete sequence data: transform the fitted depth-inclination data curve Q and the fitted depth-tool face angle function curve COS(G) into function data x 1 (T) and x 2 (T);

[0022] Assume that the fitted depth-inclination data curve Q is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 1 (T) = [x 1 (t 1 ),x 1 (t 2 ),…,x1 (t n ), where x 1 (t 1 ), x 1 (t 2 ), …, x 1 (t n ) are the data samples of the depth-dip data after fitting;

[0023] Let the sample function of the curve data COS(G) of the depth-tool face angle function after fitting at the sampling time points T = (t 1 , t 2 , …, t n ) be: x 2 (T) = [x 2 (t 1 ), x 2 (t 2 ), …, x 2 (t n )], where x 2 (t 1 ), x 2 (t 2 ), …, x 2 (t n ) are the data samples of the depth-tool face angle function data after fitting;

[0024] Step A62. Calculate the time difference vector: Let Δ = (δ 1 , δ 2 , …, δ n ) be the offset of x 1 (T) relative to x 2 (T) at the sampling time points T = (t 1 , t 2 , …, t n );

[0025] Step A63. Iterate the time difference vector to obtain the curve alignment result: Denote the time difference vector at the k-th iteration as Δ k = (δ k1 , δ k2 ,..., δ kn ), perform n - 2 times of conditional maximization to obtain Δ k+1 = (δ k1+1 , δ k2+1 ,..., δ kn+1 ) until convergence, i.e., |Δ k+1 - Δ k | < eps, where eps is a minimum value, to obtain the curve alignment result.

[0026] Optionally, the azimuth data correction specifically includes the following steps:

[0027] Step B1, drawing a depth-azimuth data curve: arranging the azimuth data in the drilling trajectory measurement data in order of drilling depth to obtain depth-azimuth data, and drawing a depth-azimuth data curve according to the depth-azimuth data;

[0028] Step B2, data curve fitting: using the least square method to fit the depth-azimuth data curve to obtain a fitted depth-azimuth data curve, denoted as FW;

[0029] Step B3, drawing a depth-tool face angle data curve: arranging the tool face angle data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-tool face angle data, and drawing a depth-tool face angle data curve, which is recorded as G;

[0030] Step B4, data curve fitting: the least square method is used to fit the depth-tool face angle data curve to obtain the fitted depth-tool face angle data curve, and the function value of the fitted depth-tool face angle data is taken to generate the fitted depth-tool face angle function curve, which is recorded as SIN(G) / COS(FW);

[0031] Step B5, correlation analysis: The correlation coefficient r between the fitted depth-azimuth curve data and the depth-tool face angle function curve data is:

[0032]

[0033] In the above formula, E{FW} is the expected value of the depth-azimuth curve data FW, E{SIN(G) / COS(FW)} is the expected value of the depth-tool angle function curve data SIN(G) / COS(FW), E{(FW-E{FW})×(SIN(G) / COS(FW)-E{SIN(G) / COS(FW)})} is the covariance between the depth-azimuth curve data and the depth-tool angle function curve data, is the standard deviation of the depth-azimuth curve data FW, is the standard deviation of the depth-tool face angle function curve data SIN(G) / COS(FW); the closer the |r| value is to 1, the greater the correlation; the correlation coefficient is a dimensionless measure to describe the correlation between two series of data;

[0034] Step B6, curve alignment: if the |r| value is between 0.8 and 1, the data curve is aligned, and the time series data is converted into function data and then time correction is performed to align the depth-azimuth data curve and the depth-tool face angle function curve;

[0035] Step B7, calculate the correction coefficient and obtain the preliminary correction depth-azimuth value: multiply each azimuth in the aligned depth-azimuth curve data and the depth-tool face angle function curve SIN(G) / COS(FW) to obtain the depth-correction curve data. 1 ×SIN(G 1 ) / COS(FW 1 ),FW 2 ×SIN(G 2 ) / COS(FW 2 ),…,FW n ×SIN(G n ) / COS(FW n ) is the correction coefficient f' at different depths; each azimuth value in the depth-azimuth curve data is multiplied by the corresponding correction coefficient to obtain f'FW 1 ,f'FW 2 ,…,f'FW n , thereby obtaining a preliminary corrected depth-azimuth curve;

[0036] Step B8, performing high-order fitting on the preliminary corrected depth-azimuth curve: For the initial corrected depth-azimuth curve obtained in step B7, use the least squares method to perform high-order interpolation fitting, and the depth-azimuth curve after the interpolation high-order fitting is the actual depth-azimuth value of the borehole.

[0037] Optionally, the step B6 of aligning the curves includes:

[0038] Step B61. Functionalize the discrete sequence data: transform the fitted depth-azimuth data curve FW and the fitted depth-tool roll angle function SIN(G) / COS(FW) into function data x 1 (T) and x 2 (T);

[0039] Assume that the depth-azimuth curve data after fitting is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 1 (T) = [x 1 (t 1 ),x 1 (t 2 ),…,x 1 (t n )], where x 1 (t),x 1 (t 2 ),…,x 1 (t n) is a data sample of the depth-azimuth curve;

[0040] Let the data of the fitted depth-tool face angle function curve at the sampling time points T = (t 1 , t 2 , …, t n ) be the sample function: x 2 (T) = [x 2 (t 1 ), x 2 (t 2 ), …, x 2 (t n )], where x 2 (t), x 2 (t 2 ), …, x 2 (t n ) are the data samples of the depth-tool face angle curve;

[0041] Step B62. Calculate the time difference vector: Let Δ = (δ 1 , δ 2 , …, δ n ) be the offset of x 1 (T) relative to x 2 (T) at the sampling time points T = (t n ); 1 (T) relative to x 2 (T);

[0042] Step B63. Iterate the time difference vector to obtain the curve alignment result: Denote the time difference vector at the k-th iteration as Δ k = (δ k1 , δ k2 , …, δ kn ), perform n - 2 times of conditional maximization to obtain Δ k+1 = (δ k1+1 , δ k2+1 , …, δ kn+1 ) until convergence, i.e., |Δ k+1 - Δ k | < eps, where eps is a minimum value, to obtain the curve alignment result.

[0043] Compared with the prior art, the beneficial technical effects of the present invention are:

[0044] The present invention first performs data curve fitting on the measured raw data, and then performs correlation analysis and processing based on the inclination angle, azimuth angle data fitting curve and the tool face angle fitting curve, and selects a suitable correction coefficient after correlation processing to perform curve correction. The drilling trajectory data processed using this correction method is closer to the true value of the drilling trajectory. The present invention can perform continuous trajectory correction on the trajectory to avoid the situation of missing the target or greatly adjusting the actual drilling trajectory. At the same time, the present invention corrects the actual drilling trajectory in advance, which is conducive to safe drilling construction and improves the construction target rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A schematic diagram of the drilling trajectory measurement error principle of an embodiment of the present invention;

[0046] Figure 2 The drilling inclination angle data curve fitting result of the embodiment of the present invention;

[0047] Figure 3 The drilling azimuth angle data curve fitting result of the embodiment of the present invention;

[0048] Figure 4 A comparison diagram of the preliminary correction depth-inclination curve of an embodiment of the present invention;

[0049] Figure 5 A comparison diagram of the preliminary corrected depth-azimuth curve of an embodiment of the present invention;

[0050] Figure 6 The comparison results before and after the tilt correction of the embodiment of the present invention are shown;

[0051] Figure 7 1 is a comparison result before and after azimuth correction of an embodiment of the present invention.

[0052] The present invention is described in detail below with reference to the accompanying drawings and specific implementation methods. DETAILED DESCRIPTION

[0053] The drilling trajectory measurement data is inaccurate and the adjacent measuring points jump greatly during the measurement process. Through the collation, analysis, statistics of the measurement data and the calibration of the measurement equipment, it is found that such data mutations are usually not caused by the drilling measurement equipment, but mainly due to the regular changes caused by the influence of the on-site construction conditions during the drilling construction. Among them, the change law of the drilling measurement inclination and azimuth is correlated with the change of the tool face angle, such as Figure 1As shown, if there is a connection gap between the measuring probe and the drill pipe, or between the drill pipes near the measuring probe, due to the construction process, an angle of θ will be generated between different drilling tools. The existence of this angle causes the measuring probe to produce sudden changes in inclination and azimuth due to changes in the tool face angle during the drilling trajectory measurement. The magnitude of the sudden change is related to the angle of θ. Such regular changes can be optimized through the correction calculation of the measurement data. This scheme corrects the original drilling measurement data through data statistics and calculation methods, so that the drilling measurement data is closer to the actual drilling trajectory and the calculation accuracy of the drilling trajectory is improved; this correction method simplifies the data correction process, improves accuracy, and is more in line with the drilling hole formation law, so that the correction result is more realistic.

[0054] The drilling trajectory measurement data is corrected by curve fitting, correlation analysis and processing of the measurement data; due to the different characteristics of the drilling inclination and azimuth, the drilling inclination correction method and the azimuth correction method are also different. The present invention provides a method for correcting the deflection of a drilling trajectory measurement probe, which includes inclination data correction and azimuth data correction.

[0055] In accordance with the above technical solution, the following are specific embodiments of the present invention. It should be noted that the present invention is not limited to the following specific embodiments, and all equivalent changes made on the basis of the technical solution of this application fall within the protection scope of the present invention. The present invention is further described in detail below in conjunction with the embodiments.

[0056] Embodiment 1:

[0057] This embodiment provides a method for correcting the deflection of a probe tube for drilling trajectory measurement. The method corrects the drilling trajectory measurement data by curve fitting, correlation analysis and processing of the measurement data. Due to the different characteristics of the drilling inclination and azimuth, the drilling inclination correction method is different from the azimuth correction method. The following analyzes the drilling inclination and azimuth correction methods and steps respectively:

[0058] Step A, inclination data correction method: drawing a depth-inclination curve and obtaining a depth-inclination data curve after interpolation fitting, drawing a depth-tool face angle curve and obtaining a depth-tool face angle data curve after interpolation fitting, analyzing and judging the correlation between the fitted depth-inclination data curve and the fitted depth-tool face angle data curve and aligning the curves, calculating the correction coefficient after the curves are aligned and obtaining a preliminary corrected depth-inclination curve, interpolating the preliminary corrected depth-inclination curve to obtain a true depth-inclination value of the drilling trajectory by high-order fitting;

[0059] The specific steps include:

[0060] Step A1, drawing a depth-inclination data curve: arranging the inclination data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-inclination data, and drawing a depth-inclination data curve according to the depth-inclination data;

[0061] Step A2, data curve fitting: The least square method is used to fit the depth-inclination data curve to obtain the fitted depth-inclination data curve, which is recorded as Q; Figure 2 The figure shows the depth-inclination data curve fitting result, where curve 21 is the depth-inclination data curve, and curve 22 is the depth-inclination data curve after fitting.

[0062] Specifically, curve fitting is function approximation, which is a numerical calculation method for obtaining approximate functions. The least squares method is used here to perform fitting calculations on borehole measurement data. The least squares method is a commonly used method in curve fitting. The least squares method curve fitting first seeks the function y=f(x) so that the function f(x) is closest to all data points under certain criteria, that is, the sum of the squares of the distances from all discrete points to a certain curve is the smallest. When this condition is met, the curve fitting is considered to be the best. The least squares method can reflect the overall trend of the data points measured by the borehole trajectory and eliminate local fluctuations. Through the analysis of the least squares data curve fitting method, the method is used to perform curve fitting calculations on the borehole trajectory measurement data to obtain the borehole inclination data curve fitting results. The data after curve fitting can be used for data correlation analysis.

[0063] Step A3, drawing a depth-tool face angle data curve: arranging the tool face angle data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-tool face angle data, and drawing a depth-tool face angle data curve, which is recorded as G;

[0064] Step A4, data curve fitting: the least square method is used to fit the depth-tool face angle data curve to obtain a fitted depth-tool face angle data curve, and the cosine function value of the fitted depth-tool face angle data is taken to generate a fitted depth-tool face angle function curve, which is recorded as COS(G);

[0065] Step A5, correlation analysis: The purpose of the correlation analysis is to determine the correlation relationship between the depth-inclination curve and the depth-tool face angle curve; the correlation coefficient r between the depth-inclination curve and the depth-tool face angle curve is the quotient of the covariance of the depth-inclination data and the depth-tool face angle function data and the product of the standard deviation of the depth-inclination data and the depth-tool face angle function data;

[0066] The correlation coefficient r between the fitted depth-inclination data curve and the fitted depth-tool face angle function curve is:

[0067]

[0068] In the above formula, E{Q} is the expected value of the fitted depth-inclination data curve Q, E{COS(G)} is the expected value of the fitted depth-tool roll angle function curve COS(G), E{(QE{Q})×(COS(G)-E{COS(G)})} is the covariance between the fitted depth-inclination data curve and the fitted depth-tool roll angle function curve data, is the standard deviation of the fitted depth-inclination data curve Q, is the standard deviation of the fitted depth-tool face angle function curve COS(G); the closer the |r| value is to 1, the greater the correlation is; the correlation coefficient is a dimensionless measure to describe the correlation between two series of data;

[0069] Step A6, curve alignment: if the |r| value is between 0.8 and 1, the data curve is aligned, and the curve alignment criterion of the data composition can be constructed by the functional correlation coefficient;

[0070] Step A61. Functionalize the discrete sequence data: Functionalize the discrete sequence data, that is, convert the fitted depth-inclination data curve Q and the fitted depth-tool face angle function curve COS(G) into function data x 1 (T) and x 2 (T);

[0071] Assume that the fitted depth-inclination data curve Q is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 1 (T) = [x 1 (t 1 ),x 1 (t 2 ),…,x 1 (t n )], where x 1 (t 1 ),x 1 (t 2 ),…,x 1 (t n ) is the data sample of the depth-inclination data after fitting;

[0072] Assume that the fitted depth-tool face angle function curve data COS(G) is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 2 (T) = [x 2(t 1 ), x 2 (t 2 ), …, x 2 (t n ), where x 2 (t 1 ), x 2 (t 2 ), …, x 2 (t n ) are data samples of the depth - tool face angle function data after fitting;

[0073] Step A62. Calculate the time difference vector: Let Δ = (δ 1 , δ 2 , …, δ n ) be the offset of x 1 (T) relative to x 2 (T) at the sampling time points T = (t n , t 1 , …, t 2 );

[0074] In the said step A62, smoothing processing of the time difference vector is also performed: Use P - spline to fit the function d k+1 (t) about the sequence Δ k+1 , and replace the original value with the fitted value, that is, Δ k+1 = d k+1 (q i ), i = 1, 2, …, n.

[0075] Step A63. Iterate the time difference vector to obtain the curve alignment result: Denote the k - th iteration time difference vector as Δ k = (δ k1 , δ k2 ,..., δ kn ), perform n - 2 times of conditional maximization to obtain Δ k+1 = (δ k1+1 , δ k2+1 ,..., δ kn+1 ) until convergence, that is, |Δ k+1 - Δ k | < eps, where eps is a minimum value, to obtain the curve alignment result;

[0076] Step A7. Calculate the correction coefficient and obtain the preliminary corrected depth - dip value: Multiply each dip value in the depth - dip data after curve alignment and each cosine value of the tool face angle in the depth - tool face angle function data to obtain the depth - correction curve data. The data q 1 ×COS(G 1 ), q 2 ×COS(G2 ),…,q n ×COS(G n ), which is the correction coefficient f at different depths; multiply each inclination value in the depth-inclination curve data by the corresponding correction coefficient to obtain fq 1 ,fq 2 ,…,fq n , thus obtaining a preliminary corrected depth-inclination curve; Figure 4 The figure shows a comparison diagram of the preliminary corrected depth-inclination curves, where curve 41 is the original inclination measurement data curve, curve 42 is the depth-inclination fitting curve, and curve 43 is the preliminary corrected depth-inclination curve.

[0077] Step A8, high-order fitting of the preliminary correction depth-inclination curve: the preliminary correction depth-inclination curve obtained in step A7 is interpolated and fitted using the least square method. The depth-inclination curve after the interpolation and high-order fitting is the actual depth-inclination value of the borehole. Figure 6 The figure shows the comparison results before and after the inclination correction. The curve 61 in the figure is the original inclination measurement data curve, and the curve 62 in the figure is the depth-inclination curve after high-order fitting.

[0078] Step B, azimuth data correction method: draw a depth-azimuth curve and obtain a depth-azimuth data curve after interpolation fitting, draw a depth-tool face angle curve and obtain a depth-tool face angle data curve after interpolation fitting, analyze and determine the correlation between the fitted depth-azimuth data curve and the fitted depth-tool face angle data curve and align the curves, calculate the correction coefficient after the curves are aligned and obtain a preliminary corrected depth-azimuth curve, interpolate the preliminary corrected depth-azimuth curve to obtain a true depth-azimuth value of the drilling trajectory by high-order fitting;

[0079] The specific steps include:

[0080] Step B1, drawing a depth-azimuth data curve: arranging the azimuth data in the drilling trajectory measurement data in order of drilling depth to obtain depth-azimuth data, and drawing a depth-azimuth data curve according to the depth-azimuth data;

[0081] Step B2, data curve fitting: The least square method is used to fit the depth-azimuth data curve to obtain a fitted depth-azimuth data curve, which is recorded as FW; Figure 3 The figure shows the depth-azimuth data curve fitting result, where curve 31 is the depth-azimuth data curve, and curve 32 is the depth-azimuth data curve after fitting.

[0082] Specifically, curve fitting is function approximation, which is a numerical calculation method for obtaining approximate functions. The least squares method is used here to fit the borehole measurement data. The least squares method is a commonly used method in curve fitting. The least squares curve fitting first finds the function y=f(x) so that the function f(x) is closest to all data points under certain criteria, that is, the sum of the squares of the distances from all discrete points to a certain curve is the smallest. When this condition is met, the curve fitting is considered to be the best. The least squares method can reflect the overall trend of the borehole trajectory measurement data points and eliminate local fluctuations. Through the analysis of the least squares data curve fitting method, the method is used to perform curve fitting calculations on the borehole trajectory measurement data to obtain the borehole azimuth data curve fitting results. The data after curve fitting can be used for data correlation analysis.

[0083] Step B3, drawing a depth-tool face angle data curve: arranging the tool face angle data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-tool face angle data, and drawing a depth-tool face angle data curve, which is recorded as G;

[0084] Step B4, data curve fitting: the least square method is used to fit the depth-tool angle data curve to obtain the fitted depth-tool angle data curve, and the function value of the fitted depth-tool angle data is taken to generate the fitted depth-tool angle function curve, which is recorded as SIN(G) / COS(FW).

[0085] Step B5, correlation analysis: The purpose of the correlation analysis is to determine the correlation relationship between the depth-azimuth curve and the depth tool facing angle curve; two arrays such as (FW, SIN(G) / COS(FW)) are used to represent the two curves, the data FW is the depth-azimuth data curve fitted in step B2, the array SIN(G) / COS(FW) is the depth-tool facing angle function curve after fitting, and the correlation coefficient r of the two data curves is the quotient of the covariance of the depth-azimuth data and the depth-tool facing angle function data and the product of the standard deviation of the depth-azimuth data and the depth-tool facing angle function data;

[0086] The correlation coefficient r between the depth-azimuth curve data and the depth-tool roll angle function curve data is:

[0087]

[0088] In the above formula, E{FW} is the expected value of the depth-azimuth curve data FW, E{SIN(G) / COS(FW)} is the expected value of the depth-tool angle function curve data SIN(G) / COS(FW), E{(FW-E{FW})×(SIN(G) / COS(FW)-E{SIN(G) / COS(FW)})} is the covariance between the depth-azimuth curve data and the depth-tool angle function curve data, is the standard deviation of the depth-azimuth curve data FW, is the standard deviation of the depth-tool face angle function curve data SIN(G) / COS(FW); the closer the |r| value is to 1, the greater the correlation; the correlation coefficient is a dimensionless measure to describe the correlation between two series of data;

[0089] Step B6, curve alignment: if the |r| value is between 0.8 and 1, the data curve is aligned, and the curve alignment criterion of the data composition can be constructed by the functional correlation coefficient;

[0090] Step B61. Functionalize the discrete sequence data: Functionalize the discrete sequence data, that is, convert the fitted depth-azimuth data curve FW and the fitted depth-tool face angle function SIN(G) / COS(FW) into function data x 1 (T) and x 2 (T);

[0091] Assume that the depth-azimuth curve data after fitting is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 1 (T) = [x 1 (t 1 ),x 1 (t 2 ),…,x 1 (t n )], where x 1 (t),x 1 (t 2 ),…,x 1 (t n ) is the data sample of the depth-azimuth curve;

[0092] Assume that the fitted depth-tool face angle function curve data is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 2 (T) = [x 2 (t 1 ),x2 (t 2 ),…,x 2 (t n )], where x 2 (t), x 2 (t 2 ),…, x 2 (t n ) are data samples of the depth-tool face angle curve;

[0093] Step B62. Calculate the time difference vector: Let Δ=(δ 1 , δ 2 ,…, δ n ) be the offset of x 1 , t 2 ,…, t n ) at the sampling time point T=(t 1 (T) relative to x 2 (T);

[0094] In the said step B62, smoothing processing of the time difference vector is also performed: Use P-spline to fit the function d k+1 (t) about the sequence Δ k+1 , and replace the original value with the fitted value, that is, Δ k+1 = d k+1 (q i ), i = 1, 2,…, n.

[0095] Step B63. Iterate the time difference vector to obtain the curve alignment result: Denote the time difference vector of the k-th iteration as Δ k =(δ k1 , δ k2 ,..., δ kn ), perform n - 2 times of conditional maximization to obtain Δ k+1 =(δ k1+1 , δ k2+1 ,..., δ kn+1 ) until convergence, that is, |Δ k+1 - Δ k | < eps, where eps is a minimum value, to obtain the curve alignment result;

[0096] Step B7. Calculate the correction coefficient and obtain the preliminary corrected depth-azimuth value: Multiply each azimuth angle in the depth-azimuth curve data after curve alignment and the depth-tool face angle function curve SIN(G) / COS(FW) to obtain the depth-corrected curve data. The data FW 1 ×SIN(G 1 ) / COS(FW 1 ), FW 2 ×SIN(G 2) / COS(FW 2 ),…,FW n ×SIN(G n ) / COS(FW n ) is the correction coefficient f' at different depths; each azimuth value in the depth-azimuth curve data is multiplied by the corresponding correction coefficient to obtain f'FW 1 ,f'FW 2 ,…,f'FW n , thereby obtaining a preliminary corrected depth-azimuth curve; Figure 5 The figure shows a comparison diagram of the initial correction depth-azimuth curve, in which curve 51 is the original azimuth measurement data curve, curve 52 is the depth-azimuth fitting curve, and curve 53 is the initial correction depth-azimuth curve.

[0097] Step B8, high-order fitting of the preliminary correction depth-azimuth curve: the initial correction depth-azimuth curve obtained in step B7 is interpolated and high-order fitted using the least square method. The depth-azimuth curve after the interpolation and high-order fitting is the actual depth-azimuth value of the borehole. Figure 7 The figure shows the comparison results before and after the azimuth correction. Curve 71 in the figure is the original azimuth measurement data curve, and curve 72 in the figure is the corrected depth-azimuth curve.

[0098] In the calculation of borehole trajectory data correction, it should be noted that: when the borehole is nearly horizontal or has a small inclination, the azimuth is not affected by the calculation process during the data calculation; but when the borehole inclination is large or close to 90°, the slightest change in the borehole inclination has a greater impact on the borehole azimuth; this is because when the borehole inclination is close to 90°, the borehole azimuth will produce random changes within 360°.

[0099] Through the above technical scheme, data correction first performs data curve fitting on the measured raw data, and then performs correlation analysis and processing based on the inclination, azimuth data fitting curve and the tool face angle fitting curve. After correlation processing, a suitable correction coefficient is selected for curve correction. The drilling trajectory data processed using this correction method is closer to the true value of the drilling trajectory. Continuous trajectory correction is performed on the trajectory to avoid the situation of missing the target or significantly adjusting the actual drilling trajectory. At the same time, the present invention corrects the actual drilling trajectory in advance, which is conducive to safe drilling construction and improves the construction target rate.

Claims

1. A method for correcting probe tube deflection in drilling trajectory measurement, It is characterized in that Including inclination data correction and azimuth data correction; The inclination data correction includes: drawing a depth-inclination curve and obtaining a depth-inclination data curve after interpolation fitting, drawing a depth-tool face angle curve and obtaining a depth-tool face angle data curve after interpolation fitting, analyzing and judging the correlation between the fitted depth-inclination data curve and the fitted depth-tool face angle data curve and aligning the curves, calculating a correction coefficient after the curves are aligned and obtaining a preliminary corrected depth-inclination curve, and interpolating a high-order fitting of the preliminary corrected depth-inclination curve to obtain a true depth-inclination value of the drilling trajectory; The azimuth data correction includes: drawing a depth-azimuth curve and obtaining a depth-azimuth data curve after interpolation fitting, drawing a depth-tool face angle curve and obtaining a depth-tool face angle data curve after interpolation fitting, analyzing and judging the correlation between the fitted depth-azimuth data curve and the fitted depth-tool face angle data curve and aligning the curves, calculating the correction coefficient after the curves are aligned and obtaining a preliminary corrected depth-azimuth curve, and interpolating the preliminary corrected depth-azimuth curve for high-order fitting to obtain a true depth-azimuth value of the drilling trajectory.

2. The method for correcting probe deflection in drilling trajectory measurement according to claim 1, It is characterized in that The inclination data correction specifically comprises the following steps: Step A1, drawing a depth-inclination data curve: arranging the inclination data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-inclination data, and drawing a depth-inclination data curve according to the depth-inclination data; Step A2, data curve fitting: using the least square method to fit the depth-inclination data curve to obtain a fitted depth-inclination data curve, denoted as Q; Step A3, drawing a depth-tool face angle data curve: arranging the tool face angle data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-tool face angle data, and drawing a depth-tool face angle data curve, which is recorded as G; Step A4, data curve fitting: the least square method is used to fit the depth-tool face angle data curve to obtain a fitted depth-tool face angle data curve, and the cosine function value of the fitted depth-tool face angle data is taken to generate a fitted depth-tool face angle function curve, which is recorded as COS(G); Step A5, correlation analysis: The correlation coefficient r between the fitted depth-inclination data curve and the fitted depth-tool face angle function curve is: In the above formula, E{Q} is the expected value of the fitted depth-inclination data curve Q, E{COS(G)} is the expected value of the fitted depth-tool roll angle function curve COS(G), E{(QE{Q})×(COS(G)-E{COS(G)})} is the covariance between the fitted depth-inclination data curve and the fitted depth-tool roll angle function curve data, is the standard deviation of the fitted depth-inclination data curve Q, is the standard deviation of the fitted depth-tool face angle function curve COS(G); the closer the |r| value is to 1, the greater the correlation is; the correlation coefficient is a dimensionless measure to describe the correlation between two series of data; Step A6, curve alignment: if the |r| value is between 0.8 and 1, the data curve is aligned, and the time series data is converted into function data and then time correction is performed to align the depth-inclination data curve and the depth-tool face angle function curve; Step A7, calculate the correction coefficient and obtain the preliminary correction depth-inclination value: calculate the product of each inclination value in the depth-inclination data after the curve is aligned and each tool face angle cosine value in the depth-tool face angle function data to obtain the depth-correction curve data. 1 ×COS(G 1 ),q 2 ×COS(G 2 ),…,q n ×COS(G n ), which is the correction coefficient f at different depths; multiply each inclination value in the depth-inclination curve data by the corresponding correction coefficient to obtain fq 1 ,fq 2 ,…,fq n , thereby obtaining a preliminary corrected depth-inclination curve; Step A8, performing high-order fitting on the preliminary corrected depth-inclination curve: For the preliminary corrected depth-inclination curve obtained in step A7, use the least squares method to perform high-order interpolation fitting, and the depth-inclination curve after the high-order interpolation fitting is the actual depth-inclination value of the borehole.

3. The method for correcting the deflection of a probe tube for drilling trajectory measurement according to claim 2, It is characterized in that The step A6 of aligning the curves comprises: Step A61. Functionalize discrete sequence data: transform the fitted depth-inclination data curve Q and the fitted depth-tool face angle function curve COS(G) into function data x 1 (T) and x 2 (T); Assume that the fitted depth-inclination data curve Q is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 1 (T) = [x 1 (t 1 ),x 1 (t 2 ),…,x 1 (t n )], where x 1 (t 1 ),x 1 (t 2 ),…,x 1 (t n ) is the data sample of the depth-inclination data after fitting; Assume that the fitted depth-tool face angle function curve data COS(G) is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 2 (T) = [x 2 (t 1 ),x 2 (t 2 ),…,x 2 (t n )], where x 2 (t 1 ),x 2 (t 2 ),…,x 2 (t n ) is a data sample of the depth-tool angle function data after fitting; Step A62. Calculate the time difference vector: Let Δ=(δ 1 ,δ 2 ,…,δ n ) is the sampling time point T = (t 1 ,t 2 ,…,t n ) x 1 (T) relative to x 2 (T) offset; Step A63. Iterate the time difference vector to obtain the curve alignment result: Denote the time difference vector at the k-th iteration as Δ k =(δ k1 , δ k2 ,..., δ kn ), perform n - 2 times of conditional maximization to obtain Δ k+1 =(δ k1+1 , δ k2+1 ,..., δ kn+1 ), until convergence, i.e., |Δ k+1 - Δ k | < eps, where eps is a minimum value, to obtain the curve alignment result.

4. The method for correcting probe deflection in drilling trajectory measurement according to claim 1, It is characterized in that The azimuth data correction specifically comprises the following steps: Step B1, drawing a depth-azimuth data curve: arranging the azimuth data in the drilling trajectory measurement data in order of drilling depth to obtain depth-azimuth data, and drawing a depth-azimuth data curve according to the depth-azimuth data; Step B2, data curve fitting: using the least square method to fit the depth-azimuth data curve to obtain a fitted depth-azimuth data curve, denoted as FW; Step B3, drawing a depth-tool face angle data curve: arranging the tool face angle data in the drilling trajectory measurement data in the order of drilling depth to obtain depth-tool face angle data, and drawing a depth-tool face angle data curve, which is recorded as G; Step B4, data curve fitting: the least square method is used to fit the depth-tool face angle data curve to obtain the fitted depth-tool face angle data curve, and the function value of the fitted depth-tool face angle data is taken to generate the fitted depth-tool face angle function curve, which is recorded as SIN(G) / COS(FW); Step B5, correlation analysis: The correlation coefficient r between the fitted depth-azimuth curve data and the depth-tool face angle function curve data is: In the above formula, E{FW} is the expected value of the depth-azimuth curve data FW, E{SIN(G) / COS(FW)} is the expected value of the depth-tool angle function curve data SIN(G) / COS(FW), E{(FW-E{FW})×(SIN(G) / COS(FW)-E{SIN(G) / COS(FW)})} is the covariance between the depth-azimuth curve data and the depth-tool angle function curve data, is the standard deviation of the depth-azimuth curve data FW, is the standard deviation of the depth-tool face angle function curve data SIN(G) / COS(FW); the closer the |r| value is to 1, the greater the correlation; the correlation coefficient is a dimensionless measure to describe the correlation between two series of data; Step B6, curve alignment: if the |r| value is between 0.8 and 1, the data curve is aligned, and the time series data is converted into function data and then time correction is performed to align the depth-azimuth data curve and the depth-tool face angle function curve; Step B7, calculate the correction coefficient and obtain the preliminary correction depth-azimuth value: multiply each azimuth in the aligned depth-azimuth curve data and the depth-tool face angle function curve SIN(G) / COS(FW) to obtain the depth-correction curve data. 1 ×SIN(G 1 ) / COS(FW 1 ),FW 2 ×SIN(G 2 ) / COS(FW 2 ),…,FW n ×SIN(G n ) / COS(FW n ) is the correction coefficient f' at different depths; each azimuth value in the depth-azimuth curve data is multiplied by the corresponding correction coefficient to obtain f'FW 1 ,f'FW 2 ,…,f'FW n , thereby obtaining a preliminary corrected depth-azimuth curve; Step B8, performing high-order fitting on the preliminary corrected depth-azimuth curve: For the initial corrected depth-azimuth curve obtained in step B7, use the least squares method to perform high-order interpolation fitting, and the depth-azimuth curve after the interpolation high-order fitting is the actual depth-azimuth value of the borehole.

5. The method for correcting probe deflection in drilling trajectory measurement according to claim 4, It is characterized in that The step B6 of aligning the curves comprises: Step B61. Functionalize the discrete sequence data: transform the fitted depth-azimuth data curve FW and the fitted depth-tool roll angle function SIN(G) / COS(FW) into function data x 1 (T) and x 2 (T); Assume that the depth-azimuth curve data after fitting is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 1 (T) = [x 1 (t 1 ),x 1 (t 2 ),…,x 1 (t n )], where x 1 (t),x 1 (t 2 ),…,x 1 (t n ) is the data sample of the depth-azimuth curve; Assume that the fitted depth-tool face angle function curve data is at the sampling time point T = (t 1 ,t 2 ,…,t n The sample function at ) is: 2 (T) = [x 2 (t 1 ),x 2 (t 2 ),…,x 2 (t n )], where x 2 (t),x 2 (t 2 ),…,x 2 (t n ) is a data sample of the depth-tool face angle curve; Step B62. Calculate the time difference vector: Let Δ=(δ 1 ,δ 2 ,…,δ n ) is the sampling time point T = (t 1 ,t 2 ,…,t n ) x 1 (T) relative to x 2 (T) offset; Step B63. Iterate the time difference vector to obtain the curve alignment result: Denote the time difference vector at the k-th iteration as Δ k =(δ k1 , δ k2 ,..., δ kn ), perform n - 2 times of conditional maximization to obtain Δ k+1 =(δ k1+1 , δ k2+1 ,..., δ kn+1 ), until convergence, i.e., |Δ k+1 - Δ k | < eps, where eps is a minimum value, to obtain the curve alignment result.

Images