A method for filtering out axial impact signals of drilling tools based on jerk filtering

Abstract

The application discloses a drilling tool axial impact signal filtering method based on jerk filtering. The method does not need the intervention of prior knowledge, and first uses the Z-axis of the three-axis accelerometer coupled with the drilling tool to obtain the drilling tool axial acceleration signal, and obtains the jerk value of the sampling signal, so as to greatly increase the amplitude of the sampling signal. Then, the characteristics of the short-time energy of the impact acceleration are used to define the jerk peak value, set the jerk peak value condition, and determine the number of the jerk peak value. The jerk impact threshold value is set, the threshold value sampling method (TVS-threshold-value sampling) is used to screen out the jerk impact value meeting the condition, and then the energy interpolation principle (EIP-energy interpolation principle) is used to restore the sampling point number. Finally, the measurement signal of the drilling tool after the axial impact acceleration is filtered is obtained through the operation of adjacent time.

Description

Technical Field

[0001] This invention belongs to the field of measurement while drilling, and specifically relates to a method for filtering axial impact signals of drill bits based on rapid filtering. Background Technology

[0002] In modern measurement-while-drilling (MWD) technology, it is necessary to acquire continuous and dynamic attitude angles in a timely and accurate manner to control the drilling trajectory. In inertial MWD systems, the accuracy of calculating the wellbore attitude angles using accelerometer data depends on the gravity signal in the drill string coordinate system. However, due to strong noise interference, the gravity signal in the drill string coordinate system can no longer be approximated as the accelerometer measurement value during attitude measurement. Therefore, it is necessary to effectively extract the drill string gravity acceleration signal. However, during drilling, the drill string often encounters different geological conditions with alternating soft and hard surfaces, causing the drill string axial gravity acceleration signal to be mainly affected by harmful impact acceleration. Therefore, it is very important to filter out impact signals from the Z-axis measurement signal of the strapdown triaxial accelerometer in the drill string axis.

[0003] Current filtering algorithms for measurement signal processing include amplitude limiting filtering, median filtering, and arithmetic average filtering. However, these algorithms lack specific application targets and have limitations in the engineering field. Furthermore, these algorithms are based on data processing and do not consider the physical characteristics of the measured object in actual engineering, resulting in the loss of useful signal energy in the filtered signal and leading to errors in the processed measurement signal. In the measurement while drilling process, a smoothing filtering method is used to process the axial acceleration measurement signal of the drill string. The effectiveness of this filtering method depends on the number of input sample data N, and the value of N needs to be determined by prior knowledge. Therefore, when studying the extraction of the axial gravity acceleration signal of the drill string, the processing of the measured acceleration signal should break free from the constraints of prior knowledge and combine the physical characteristics of the measured object to filter the measured signal. Summary of the Invention

[0004] The purpose of this invention is to provide a method for filtering axial impact signals from drill bits based on rapid motion filtering.

[0005] The technical solution to achieve the purpose of this invention is as follows: a method for filtering axial impact signals from drill bits based on rapid-motion filtering, comprising the following steps:

[0006] Step 1: Let t be the sampling time of the acceleration signal measured along the drill string axis, i.e., the Z-axis of the triaxial accelerometer. (n) (n=0,1,…N), calculate the abrupt change value j at each time step. k and all j k The value is denoted as an N-dimensional column vector J. k ;

[0007] Step 2: Define the abrupt peak value as j pSet the number of abrupt spikes to b and initialize it to 0;

[0008] Step 3: Set the condition for the existence of abrupt peak values. If the condition is met, return to Step 2 and execute the instruction b = b + 1. Simultaneously calculate the abrupt peak values ​​and collect all abrupt peak values ​​j. p denoted as b-dimensional column vector J p ;

[0009] Step 4: Calculate the sum of all abrupt changes from Step 1. Calculate the sum of all abrupt peak values ​​in step 3. Set the JIT (Just-In-Time) threshold;

[0010] Step 5: Using the sudden impact threshold obtained in Step 4, first use the threshold sampling method (TVS-threshold-value sampling) to delete j k For samples with JIT values ​​greater than or equal to the JIT value, the energy interpolation method (EIP) is used to recover the number of sampling points.

[0011] Step 6: Record all abrupt change values ​​obtained in steps 1 to 5 as JF. k(N×1) The inverse integration operation is performed to obtain the measured values ​​of the drill string axis at each moment without impact acceleration, which are used as the filtered drill string axis measurement signal.

[0012] The beneficial effects of this invention are:

[0013] (1) During the filtering process, it is free from the constraints of prior knowledge and does not require setting filter parameters and sliding step size;

[0014] (2) It takes into account the actual engineering physical characteristics of the drill bit axis during drilling, which is highly targeted and solves the problem of excessive signal energy loss. Attached Figure Description

[0015] Figure 1 This is a schematic flowchart of a method for filtering axial impact signals from a drill bit based on rapid-motion filtering, according to the present invention.

[0016] Figure 2 This is a simulation diagram of the acceleration abrupt value obtained after abrupt filtering of the axial acceleration measurement signal of the drill bit in an embodiment of the present invention. Detailed Implementation

[0017] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] like Figure 1 As shown, this invention provides a method for filtering axial impact signals from drill bits based on abrupt changes in direction. The detailed steps are as follows:

[0019] Step 1: Let t be the sampling time of the acceleration signal measured along the drill string axis, i.e., the Z-axis of the triaxial accelerometer. (n) (n=0,1,…N), calculate the abrupt change value J at each time point. k :

[0020]

[0021] In the formula: J k Let Δa be the abrupt change in the axial acceleration of the N-dimensional drill bit. z,k (k=0,1,…N-1) represents the change in the axial acceleration value measured by the drill bit, and Δt is the time interval corresponding to the change in acceleration, [j1 j2 … j N ] T This represents the abrupt change value at the corresponding sampling time.

[0022] Step 2: Define the abrupt peak value as j p Set the number of abrupt spikes to b and initialize it to 0;

[0023] Step 3: Determine the conditions for the existence of the acute peak value and calculate the acute peak value:

[0024] ① Set the conditions for the existence of abrupt peak values:

[0025]

[0026] In the formula: j k and j k+1 For the axial acceleration jump values ​​of the drill bit at adjacent moments, min{j k ,j k+1} represents the minimum value of the acceleration abrupt change at adjacent moments in the axial measurement of the drill bit;

[0027] ② If the condition of equation (2) is met, it means that there is an abrupt peak value between adjacent abrupt values. Return to step 2 and execute the instruction b = b + 1 to collect all abrupt peak values ​​j. p denoted as b-dimensional column vector J p ,but:

[0028]

[0029] In the formula: max{j k ,j k+1} represents the maximum value of the acceleration abrupt change at adjacent moments in the axial measurement of the drill bit; [j p1 j p2 … j pb ] T It is a b-dimensional column vector composed of all the abrupt peaks, where the dimension b of the abrupt peaks is the number of abrupt peaks during the sampling time;

[0030] Step 4: Based on steps 1 to 3, and according to the axial measurement data during drilling, set the JIT (Just-In-Time) threshold.

[0031]

[0032] In the formula: For all rapid values ​​j k The sum of, For all rapid peak values ​​j p The sum, N is the rapid value j k The number of peak values, b is the peak value of the rapid pulse j. p The number of;

[0033] Step 5: First, use threshold sampling (TVS-threshold-value sampling), which compares the sudden impact value at each time step with the threshold value for sudden impact, and deletes j. k For samples with values ​​≥ JIT, the energy interpolation principle (EIP) is used to reconstruct the number of sampling points.

[0034] ①J k The sampled value of ≥JIT is defined as region A, j k The JIT sampling value is defined as region B, and the abrupt change value for region A is set to... The agitation value for area B is set to...

[0035] ② Calculate the axial drilling kinetic energy of the drill string. According to the kinetic energy formula, then:

[0036]

[0037] V z (k)=V z,v (k)+V z,s (k)+V z,g (k) (6)

[0038] Where: ΔE z (k) represents the axial kinetic energy increment of the drill string; m represents the mass of the drill string during axial drilling; V z (k) represents the axial velocity of the drill bit, V z,v V is the axial vibration velocity of the drill bit. z,s V is the axial impact velocity of the drill string. z,g The useful axial velocity of the drill bit;

[0039] ③ According to the above formulas (2), (3), and (4), it can be seen that at each sampling point, the kinetic energy generated by axial impact acceleration is dominant in region A, while the kinetic energy generated by axial gravitational acceleration is dominant in region B. Based on this characteristic:

[0040] Extract each The two consecutive values ​​to the left value, and will The time corresponding to the value is defined as k. L ;like The two consecutive values ​​to the left A mutation occurred in the value, namely two One of the values ​​is Value or two The values ​​are all If the value is not found, continue iterating to the left until two values ​​are found. The value is obtained by traversing until no mutation occurs. The dimension is greater than The dimension of , taking into account the axial motion of the drill string during drilling, is as follows:

[0041] W z (k L ) = F z ·ΔX(k L -1,k L (7)

[0042] F z =ma z (k) (8)

[0043] In the formula: W z (k L (k) represents the axial work done by the drill string during drilling. L -1,k L )for The two consecutive values ​​to the left The value corresponds to the time interval, ΔX is the change in displacement of the drill string during axial drilling, and F z Let m be the axial traction force on the drill string, m be the mass of the drill string during axial drilling, and a be the axial traction force. z (k) represents the measured value of the drill bit's axial acceleration;

[0044] ④ Since the axial gravitational acceleration signal of the drill bit is a slowly changing signal, according to the law of conservation of energy, then:

[0045] W z (k L )=ΔE z,g (k-1,k) (9)

[0046]

[0047] Right now:

[0048]

[0049] so:

[0050]

[0051] therefore:

[0052]

[0053] Where: ΔV z,g (k-1,k) represents the useful axial velocity increment of the drill string; ΔE z,g (k-1,k) represents the kinetic energy increment generated by the useful axial velocity of the drill bit; Δt g a is the time interval corresponding to the useful velocity increment; z,g (k) represents the acceleration after filtering out the impact energy; This represents the abrupt change value at various moments in the drill string's axial direction, excluding impact acceleration. The alternative value;

[0054] ⑤ All the calculated results Values, respectively inserted into region A Restore the original number of sampling points to the positions corresponding to the values;

[0055] Step 6: Record all the abrupt changes obtained from the filtering as JF. k(N×1) By performing inverse integration to obtain the measured values ​​of the drill string axial direction at each moment without impact acceleration, then:

[0056]

[0057] Right now:

[0058]

[0059] In the formula: The measured acceleration t is the acceleration measured at each moment along the Z-axis of the accelerometer, excluding the impact component. (n) and t (n-1) For JF k(N×1) The corresponding adjacent sampling times.

Claims

1. A method for filtering axial impact signals from drill bits based on abrupt changes in direction, characterized in that, Includes the following steps: Step 1: Let the sampling time of the acceleration signal measured by the triaxial accelerometer along the drill string axis be . Calculate the abrupt change value at each time point. and all Values, denoted as N-dimensional column vectors ; Step 2: Define the peak value of the abrupt change as... Set the number of abrupt spikes to b and initialize it to 0; Step 3: Set the condition for the existence of abrupt peak values. If the condition is met, return to Step 2 and execute the instruction b=b+1, while simultaneously calculating the abrupt peak values ​​and collecting all abrupt peak values. denoted as a b-dimensional column vector ; Step 4: Calculate the sum of all abrupt changes from Step 1. Step 3: Sum of all rapid peak values And set a sudden impact threshold. In the formula For all rapid values The sum of, For all rapid peak values The sum of, The number of abrupt changes. The number of rapid spikes; Step 5: Using the sudden impact threshold obtained in Step 4, first use the threshold sampling method (TVS-threshold-valuesampling) to delete... The sampled values ​​are then used to recover the number of sampling points using the energy interpolation method (EIP). Step 6: Record all the abrupt change values ​​obtained in steps 1-5 as follows: The inverse integration operation is performed to obtain the measured values ​​of the drill string axis at each moment without impact acceleration, which are used as the filtered drill string axis measurement signal.

2. The method for filtering axial impact signals from drill bits based on abrupt filtering according to claim 1, characterized in that, In step 2, taking advantage of the fact that the axial acceleration signal of the drill bit is easily affected by impact acceleration, a sudden peak value is defined. Set the number of rapid spikes to b and initialize it to 0.

3. The method for filtering axial impact signals from drill bits based on abrupt filtering according to claim 1, characterized in that, In step 3, the conditions for the existence of the acute peak value are determined and the acute peak value is calculated: Set the conditions for the existence of a sudden spike: (1) In the formula, and This represents the abrupt acceleration values ​​at adjacent moments in the axial measurement of the drill bit. This represents the minimum value of the acceleration abrupt change at adjacent moments in the axial measurement of the drill bit; If the condition of equation (1) is met, it indicates that there is an abrupt peak value between adjacent abrupt values. Return to step 2 to execute the b=b+1 instruction, and calculate the abrupt peak value. Define all abrupt peak values ​​as a b-dimensional column vector. ,but: (2) In the formula, This represents the maximum value of the abrupt acceleration values ​​at adjacent moments in the axial measurement of the drill bit. It is a b-dimensional column vector composed of all the rapid peak values.

4. The method for filtering axial impact signals from drill bits based on abrupt filtering according to claim 1, characterized in that, In step 5, the number of sampling points is restored using the energy interpolation principle (EIP), which specifically includes the following steps: Will The sampled values ​​are defined as region A. The sampled values ​​are defined as region B, and the abrupt change values ​​of region A are set to... The abrupt change value for region B is set to... ; Calculate the axial drilling kinetic energy of the drill string; In region A, kinetic energy generated by axial impact acceleration is dominant, while in region B, kinetic energy generated by axial gravitational acceleration is dominant; based on this characteristic: Extract each The two consecutive values ​​to the left value, and will The time corresponding to the value is defined as ;like The two consecutive values ​​to the left A mutation occurred in the value, namely two One of the values ​​is Value or two The values ​​are all If the value is not found, continue iterating to the left until two values ​​are found. The value is obtained by traversing until no mutation occurs. The dimension is greater than The dimension of , taking into account the axial motion of the drill string during drilling, is as follows: (3) (4) In the formula, This refers to the amount of axial work done by the drill bit during drilling. for The two consecutive values ​​to the left The value corresponds to the time interval. This represents the change in displacement of the drill bit during axial drilling. This refers to the axial traction force on the drill bit. This refers to the mass of the drilling tool during axial drilling. This is the measured value of the drill bit's axial acceleration; Since the axial gravitational acceleration signal of the drill bit is a slowly changing signal, according to the law of conservation of energy, the following calculations are performed. Value, the aforementioned This represents the abrupt change value at various moments in the drill string's axial direction, excluding impact acceleration. The alternative value; All the calculated results Values, respectively inserted into region A The original number of sampling points is restored to the position corresponding to the value.

Images