A method for separating ultrasonic echoes from the cemented interface of a cased well

Abstract

The application discloses a casing well cement bonding two-interface ultrasonic echo separation method, which comprises the following steps: moving the waveforms in time, aligning casing inner wall echoes of all waveforms in a cycle; removing the casing inner wall echoes of all waveforms in a cycle; calculating resonance amplitudes of all waveforms in a cycle, and normalizing the waveforms with the amplitudes; selecting any waveform as a reference waveform, adjusting the remaining waveforms, so that the remaining waveforms have the same resonance wave period or center frequency as the reference waveform, and the time sampling interval remains unchanged; moving the adjusted waveforms in time, aligning the resonance waves of all waveforms with the resonance waveform of the reference; superimposing and averaging adjacent waveforms of each aligned waveform to obtain a smoothed waveform set; subtracting the smoothed waveform from the aligned waveform to obtain separated two-interface ultrasonic echoes; and restoring the separated two-interface ultrasonic echoes to the original amplitude and period. According to the application, the two-interface ultrasonic echoes can be effectively separated by collecting waveforms in one cycle, the processing speed is fast, and the demand for calculation memory is small.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of acoustic logging, in particular to a method for separating ultrasonic echo of two interfaces of cemented casing well. BACKGROUND

[0002] After the well is completed, the casing and the formation are filled with cement to prevent fluid from different layers from flowing together. In order to evaluate the cementing quality behind the casing, various measuring instruments and evaluation methods have been developed.

[0003] At present, the ultrasonic instrument that can rotate 360° downhole is the mainstream detection method. Such an instrument can realize full coverage behind the casing and obtain high azimuthal resolution. The ultrasonic probe installed on the instrument generates ultrasonic waves that are incident vertically into the casing, and reflection waves are generated on the inner wall of the casing. At the same time, resonance waves are formed in the casing and are transmitted back to be received by the ultrasonic probe. The analysis of the resonance waves can obtain the acoustic parameters of the casing and the medium behind the casing. For example, the thickness of the casing can be calculated from the frequency of the resonance waves to evaluate the casing corrosion, and the acoustic impedance of the medium outside the casing can be used to evaluate the casing to identify the string channel. There is also a group delay analysis of the entire waveform (casing inner wall echo and resonance wave) to calculate the thickness of the casing and the acoustic impedance of the medium outside the casing.

[0004] The ultrasonic waves incident into the casing have part of the energy penetrating the casing, reflecting when encountering the formation or the outer casing, and being transmitted back to the ultrasonic probe. This part of the echo is called two-interface echo. When the annular distance is short, this part of the echo may interfere with the resonance wave, affecting the calculation of the acoustic impedance and causing a large error, resulting in a nebula pattern on the acoustic impedance image. Since the casing is usually not completely centered in the wellbore, the annular distance is different, and the two-interface echo does not arrive at the same time and is incoherent in time. Using this property, the patent US5274604 proposes a spatial filtering method to separate the two-interface ultrasonic echo. However, in actual use, due to the manufacturing process of the casing or the corrosion of the casing, the thickness of the casing is usually uneven, so that the period of the resonance wave is different. Therefore, after applying the above-mentioned spatial filtering method, there are still many resonance wave components remaining, and the filtering effect is not good. To overcome this problem, US5859811 proposes a method of sorting and grouping all measured waveforms according to the casing thickness (resonance wave center frequency or period). In each group, the period of the resonance wave is very similar, and the arrival time of the two-interface ultrasonic echo is incoherent, so that the two-interface ultrasonic echo can be well separated. However, this method requires a large amount of data for processing, and requires high memory of the computer. In addition, when the number of waveforms is small, the effect is also poor. SUMMARY

[0005] In view of the deficiencies in the prior art, the purpose of the present application is to provide a casing well cemented two-interface ultrasonic echo separation method. In order to achieve the above-mentioned purposes and other advantages according to the present application, a casing well cemented two-interface ultrasonic echo separation method is provided, comprising the following steps:

[0006] S1, aligning the waveforms collected by rotating the ultrasonic instrument for one revolution after moving the main peak of the casing inner wall echo in time;

[0007] S2, removing the casing inner wall echo of the first waveform set;

[0008] S3, calculating the resonance wave root mean square amplitude of all waveforms, and normalizing the waveforms with their respective amplitudes;

[0009] S4, calculating the period of the resonance wave of all waveforms, selecting any waveform as a reference waveform, and adjusting the remaining waveforms so that they have the same resonance wave period or center frequency as the reference waveform, and the time sampling interval remains unchanged;

[0010] S5, moving the adjusted waveforms in time so that the resonance waves of all waveforms are aligned with the reference resonance waveform, obtaining a second waveform set;

[0011] S6, taking the adjacent waveforms of each aligned waveform and averaging them to obtain a smoothed waveform set;

[0012] S7, subtracting the smoothed waveforms from the aligned second waveform set one by one to obtain the separated two-interface ultrasonic echoes;

[0013] S8, restoring the separated two-interface ultrasonic echoes to their original amplitude and period to obtain the final two-interface ultrasonic echoes.

[0014] Preferably, in step S3, the time window of the resonance wave is selected to calculate the amplitude of the resonance wave, the starting position of the time window is at the third zero point of the main peak of the aforementioned inner wall echo, and the width of the time window is about 3-8 resonance wave periods, here 20μs is selected, the root mean square amplitude of each waveform is calculated in the waveform set W2(n,kΔt), denoted as A(n), then each waveform is normalized with A(n) to obtain the waveform set W3(n,kΔt).

[0015] Preferably, in step S4, the waveform w(kΔt) with a time sampling interval of is assumed to have a resonance wave period T, and the new sampling interval is calculated; the resonance wave period of the waveform w(kΔt') with a sampling interval of Δt' changes from T to T', and w(kΔt') is re-interpolated and sampled according to the Δt interval to obtain the waveform w1(kΔt), and the interpolation method uses linear interpolation or spline interpolation.

[0016] Preferably, the waveform adjustment in step S4 can be achieved by first Fourier transforming the waveform to the frequency domain, adjusting the frequency interval in the frequency domain, then re-interpolating, and then inverse Fourier transforming to the time domain.

[0017] Compared with the prior art, the method has the advantages that the two-interface cemented ultrasonic echo can be effectively separated by collecting the waveform for only one week, the processing speed is fast, and the demand for calculation memory is small. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 A schematic diagram of an ultrasonic instrument in a cased hole for measuring according to the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0019] Figure 2 A typical received waveform of an ultrasonic probe for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0020] Figure 3 A gray scale diagram of a one-week waveform arranged in order for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0021] Figure 4 A flowchart of separating the two-interface ultrasonic echo of a one-week waveform for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0022] Figure 5 A gray scale diagram of the echo of the inner wall of the casing after alignment for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0023] Figure 6 A gray scale diagram of removing the echo of the inner wall and retaining only the resonance wave for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0024] Figure 7 The center frequency of the resonance wave of different waveforms for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0025] Figure 8 A flowchart of adjusting the resonance wave period of the waveform for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0026] Figure 9 A gray scale diagram of the waveform after re-adjusting the sampling of the resonance wave in the time domain for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0027] Figure 10 A gray scale diagram of the waveform after spatial average filtering for the method for separating the two-interface cemented ultrasonic echo of the cased hole according to the application;

[0028] Figure 11The separated two-interface ultrasonic echo gray scale map for the casing well cemented two-interface ultrasonic echo separation method according to the present application;

[0029] Figure 12 The two-interface ultrasonic echo gray scale map with original amplitude and period recovered for the casing well cemented two-interface ultrasonic echo separation method according to the present application;

[0030] Figure 13 The two-interface ultrasonic echo gray scale map directly separated without adjusting period for the casing well cemented two-interface ultrasonic echo separation method according to the present application. DETAILED DESCRIPTION

[0031] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0032] Reference Figures 1-13 A casing well cemented two-interface ultrasonic echo separation method, comprising the following steps: S1, aligning the waveforms collected by rotating an ultrasonic instrument for one circle after moving the main peak of casing inner wall echo in time;

[0033] S2, removing the casing inner wall echo of the first waveform set;

[0034] S3, calculating the resonance wave root mean square amplitude of all waveforms, and normalizing the waveforms with their respective amplitudes;

[0035] S4, calculating the period of the resonance wave of all waveforms, selecting any waveform as a reference waveform, and adjusting the remaining waveforms so that they have the same resonance wave period or center frequency as the reference waveform, and the time sampling interval remains unchanged;

[0036] S5, moving the adjusted waveforms in time so that the resonance waves of all waveforms are aligned with the reference resonance waveform, and obtaining a second waveform set;

[0037] S6, superimposing and averaging the adjacent waveforms of each aligned waveform to obtain a smoothed waveform set;

[0038] S7, subtracting the smoothed waveform from the aligned second waveform set one by one to obtain the separated two-interface ultrasonic echo;

[0039] S8, recovering the separated two-interface ultrasonic echo to the original amplitude and period to obtain the final two-interface ultrasonic echo.

[0040] As Figure 1This is a schematic diagram of the instrument and wellbore of this application. The casing is made of steel, and cement is filled between the casing and the formation. The ultrasonic instrument is placed in the casing, which is filled with mud. As the ultrasonic probe rotates, it emits ultrasonic waves towards the inner wall of the casing. Some of these waves are reflected by the casing, while others penetrate the casing and continue to propagate. After being reflected by the formation, they are received again by the ultrasonic probe.

[0041] Figure 2 This is a typical waveform received by an ultrasonic probe. The horizontal axis represents time, and the vertical axis represents the normalized amplitude. The signal amplitude is relatively large in the range of 75-83μs, which is the surface reflection wave of the inner wall of the casing. After 83μs, it consists of the resonant wave of the casing and the formation reflection wave, which cannot be well distinguished.

[0042] The instrument acquires a waveform every 5°, for a total of 72 waveforms per revolution. Each waveform has 200 points, and the sampling interval is [missing information]. All waveforms are denoted as W(n,kΔt), where n ranges from 1 to 72, representing waveforms from different orientations, and k ranges from 1 to 200, representing time sampling points. The waveforms are arranged sequentially from bottom to top and displayed as a grayscale image, as shown below. Figure 3 As shown, the horizontal axis represents time, and the vertical axis represents the waveform number. The black rectangle in the figure represents the echo from the inner wall of the casing, which has the largest amplitude. Because the instrument is slightly off-center from the casing, the arrival times of the inner wall echoes from different locations are different. Following the inner wall echo is the casing's resonance wave, which has a weaker amplitude; the interface echo reflected from the formation is not obvious.

[0043] When searching for the inner wall echoes of 72 waveforms, waveform 1 is selected as the reference. Any other waveform can be chosen, and the inner wall echoes of waveforms 2-72 are aligned with the surface echo of waveform 1 to obtain the waveform set W1(n,kΔt). Figure 5 As shown, the inner wall echoes were perfectly aligned, but the phases of the subsequent resonant waves were not the same, which was caused by the uneven thickness of the sleeve in different orientations.

[0044] Searching backward from the main peak of the inner wall echo for the third zero-crossing point, setting all values ​​before this point to zero, and removing the inner wall echo, yields the waveform set W2(n,kΔt), see [link to waveform set]. Figure 6 The phase difference between the resonant waves appears more pronounced, and direct smoothing filtering of this difference results in poor separation of the interface echoes.

[0045] Next, the period of the resonance wave is adjusted in the time domain in this embodiment, so that all the waveforms have the same period of the resonance wave. First, the time window of the resonance wave is selected to calculate the amplitude of the resonance wave, the starting position of the time window is the third zero-crossing point of the aforementioned main peak of the inner wall echo, and the width of the time window is about 3-8 periods of the resonance wave, and 20 μs is selected here. The root mean square amplitude of each waveform is calculated in the waveform set W2(n, kΔt), denoted as A(n). Then each waveform is normalized by A(n) to obtain the waveform set W3(n, kΔt). The period T(n) of the resonance wave of all waveforms is calculated, n is from 1 to 72. The peak frequency of the amplitude spectrum can be obtained by performing FFT on the waveform set W3(n, kΔt) , and the period is The peak frequency of the group delay can also be obtained by performing FFT on the waveform set W(n, kΔt) , and the period is The frequencies obtained by the two methods are shown in Figure 7 . The two methods have differences, which have little effect on subsequent processing. The center frequency of the 72 resonance waves is between 0.219 MHz and 0.245 MHz. The periods of all the waveforms are adjusted to be the same. The flow chart of the adjustment is shown in Figure 8 , and the waveform w(kΔt) with a time sampling interval is assumed to have a resonance wave period T. First, the new sampling interval is calculated:

[0046] Δt’=Δt*(T‘ / T) (1)

[0047] It is known from FFT transformation that the frequency is inversely proportional to the time sampling interval. At this time, the waveform w(kΔt’) with the sampling interval Δt’ has a resonance wave period T’ instead of T. Finally, w(kΔt’) is resampled according to the interval Δt by linear interpolation or spline interpolation to obtain the waveform w1(kΔt).

[0048] Here, the waveform W3(1, kΔt) is selected as the reference waveform, and the rest of the waveforms in the waveform set are adjusted according to the flow chart of Figure 8 to obtain the waveform set W4(n, kΔt), and then the waveforms are moved in time to align all the resonance waves to obtain the waveform set W5(n, kΔt), the result is shown in Figure 9 . Compared with Figure 6 , because all the waveforms are adjusted to the same period, the phases of the resonance waves remain consistent. The circularly cyclic smoothing filter is performed on all the waveforms, that is, the average value of the superposition of adjacent M waveforms is selected, M=23 is selected here, that is, 23 waveforms, to obtain the waveform set W6(n, kΔt), the result is shown in Figure 10Finally, subtracting waveform set W5(n, kΔt) from W6(n, kΔt) to obtain waveform set W7(n, kΔt), the result is shown in Fig. 6. Figure 11 The stratum reflection wave in the lower right corner can be clearly observed, which is in a parabolic shape. From the position of the parabolic vertex, it can be preliminarily judged that the casing is close to the external stratum in the 10th waveform. No stratum reflection wave is observed in other directions because of the large deflection of the reflection wave caused by the casing eccentricity and the large ultrasonic attenuation. Restoring W7(n, kΔt) according to the original amplitude A(n) and resonance wave period T(n) to obtain waveform set W8(n, kΔt), which is shown in Fig. 7. Figure 12 .

[0049] As a comparison, if the method of patent US5274604 is used, without readjusting the period of the resonance wave, the result is shown in Fig. 8, there are many residues of the resonance wave in the image, and the stratum reflection wave cannot be clearly identified. Figure 13

[0050] The processing scale described herein is used to simplify the description of the present application, and the application, modification and change of the present application are obvious to those skilled in the art.

[0051] Although the embodiments of the present application have been disclosed as above, it is not limited to the application and embodiment listed in the description, it can be fully applied to various fields suitable for the present application, and other modifications can be easily realized by those skilled in the art, therefore, the present application is not limited to specific details and the figures shown and described herein, without departing from the general concept defined by the claims and the equivalent scope.​

Claims

1. A method for casing cement bond bi-interface ultrasonic echo separation, characterized in that, The method comprises the following steps: S1, aligning waveforms collected by rotating the ultrasonic instrument for one circle according to the time shift of the main peak of the casing inner wall echo; S2, removing the casing inner wall echo from the first waveform set; the first waveform set is the waveform set obtained by aligning the waveforms according to the time shift of the main peak of the casing inner wall echo; S3, calculating the root mean square amplitude of the resonant wave of the waveform set obtained by removing the casing inner wall echo in step S2, and normalizing the waveforms according to the respective amplitudes; S4, calculating the period of the resonant wave of the waveform set obtained in step S3, selecting any waveform as a reference waveform, and adjusting the remaining waveforms so that they have the same resonant wave period or center frequency as the reference waveform, and the time sampling interval remains unchanged; S5, moving the adjusted waveforms in time so that the resonant wave of the waveform set obtained in step S4 is aligned with the reference resonant waveform, and obtaining a second waveform set; S6, superimposing and averaging adjacent waveforms of the aligned waveforms to obtain a smoothed waveform set; S7, subtracting the smoothed waveforms from the aligned second waveform set one by one to obtain separated two-interface ultrasonic echoes; S8, restoring the separated two-interface ultrasonic echoes to the original amplitude and period to obtain the final two-interface ultrasonic echoes.

2. A method of casing cement interfacial ultrasonic echo separation as defined in claim 1, wherein, In step S3, the time window of the resonant wave is selected to calculate the amplitude of the resonant wave, the starting position of the time window is the third zero-crossing point of the main peak of the aforementioned inner wall echo, and the width of the time window is about 3-8 resonant wave periods, and 20 μs is selected here. The root mean square amplitude of each waveform is calculated in the waveform set W2(n, kΔt) after removing the casing inner wall echo, denoted as A(n), n takes 1-72, representing waveforms in different directions, and k takes 1-200, representing time sampling points; then each waveform is normalized by A(n) to obtain the waveform set W3(n, kΔt).

3. A method of casing cement interfacial ultrasonic echo separation as defined in claim 2 wherein, The time sampling interval in step S4 The waveform w(kΔt) is assumed to have a resonance wave period T, and a new sampling interval is calculated. The resonance wave period of the waveform w(kΔt') sampled at the sampling interval Δt' changes from T to T'. The waveform w(kΔt') is resampled at the interval Δt by interpolation, and the waveform w1(kΔt) is obtained. The interpolation method is linear interpolation or spline interpolation.

4. A method of casing cement interfacial ultrasonic echo separation as defined in claim 3, wherein, In step S4, the waveform adjustment is realized by first performing Fourier transform on the waveforms to the frequency domain, adjusting the frequency interval in the frequency domain, then re-interpolating, and then performing inverse Fourier transform to the time domain.

Images