Scattering point identification technology method based on wavenumber domain

Abstract

The invention discloses a scattering point identification technology method based on a wavenumber domain, and relates to the field of data scattering imaging identification, and the method comprises the steps: S1, obtaining a three-dimensional post-stack data volume, carrying out the analysis of the three-dimensional post-stack data volume, and converting the three-dimensional post-stack data volume into a wavenumber domain, and obtaining wavenumber domain data; s2, filtering the wave number domain data by using an attenuation filter to obtain wave number energy in different directions; s3, performing inverse transformation on the wave number energy in different directions to obtain corresponding structure images in different directions; and S4, applying a multiplication imaging condition to the time domain positive spectrum data and the time domain negative spectrum data to obtain an image of the scattering point.

Description

Technical Field

[0001] The invention relates to the field of wall sealing, and in particular to a scattering point identification technical method based on wave number domain. Background Art

[0002] As seismic exploration moves deeper into the strata, the requirements for imaging accuracy become increasingly higher, and the effective identification of scattering points has become one of the important topics in oil and gas exploration.

[0003] Post-stack plane wave decomposition method: The plane wave decomposition method is also called (PWD method). It achieves the purpose of eliminating flat layer reflection and retaining scattered energy by breaking continuous event axes. However, this method will inevitably destroy the scatterer when removing the flat layer, especially when extracting three-dimensional scatterers, which will change the original shape of the scatterer and cause the scatterer event axes to be broken and distorted, which is not conducive to the subsequent oil and gas interpretation work.

[0004] Compared with the post-stack scatterer extraction method, the calculation amount and calculation time of the pre-stack scatterer extraction method will increase geometrically, which is not enough to achieve the demand of simple, fast and accurate scatterer extraction.

[0005] Therefore, a scattering point identification technology method based on the wavenumber domain is urgently needed to improve imaging accuracy. Summary of the invention

[0006] To solve the above problems, the present invention provides a scattering point identification method based on wave number domain, comprising:

[0007] S1, obtaining a three-dimensional post-stack data volume, analyzing and transforming the three-dimensional post-stack data volume into a wavenumber domain to obtain wavenumber domain data;

[0008] S2, using an attenuation filter to filter the wave number domain data to obtain wave number energy in different directions;

[0009] S3, performing inverse transformation on the wave number energies in different directions to obtain corresponding structural images in different directions;

[0010] S4. Apply multiplication imaging conditions to the time domain positive spectrum data and the time domain negative spectrum data to obtain an image of the scattering point.

[0011] Preferably, the attenuation filter mentioned in S2 is an improved rectangular attenuation filter, which can effectively separate the wavenumber domain spectrum energy and suppress the artifacts caused by the truncation effect, preserve and separate the spectrum energy and avoid the dispersion phenomenon caused by the truncation effect.

[0012] Preferably, the expression for filtering the wavenumber domain data using an attenuation filter to obtain wavenumber energy in different directions is:

[0013] Hi '(k)=H i (k)×A x ×A y ×A z ;

[0014]

[0015]

[0016] Among them, A x is the attenuation operator in the X direction, A y is the Y direction attenuation operator, A z is the Z-direction attenuation operator, H i (k) is the wave number domain data, H' i (k) is the wave number energy after attenuation in different directions, f0<f<f1, f1<f<f2 and f2<f<f3 are different attenuation intervals selected in different directions, a is the formula corresponding to the attenuation interval as the attenuation operator, where the empirical value of a is 3.125, and the three-dimensional wave number domain data is divided into 8 parts, i=1, 2, 3, 4, i is the three-dimensional wave number domain data on different components, k is the wave number in different directions, where,

[0017]

[0018] Preferably, the inverse transformation of the wave number energy in different directions to obtain the corresponding expression of the structural image in different directions is:

[0019]

[0020] in, Structural images in different directions. ;

[0021] The expression of the image of the scattering point obtained by applying the multiplication imaging condition to the time domain positive spectrum data and the time domain negative spectrum data is:

[0022]

[0023] in, is the number of spectra in different directions, I(x,y,z) diff is the image of the scatter points.

[0024] In summary, the scattering point identification technology method based on wavenumber domain of the present invention has smaller calculation amount and calculation time than traditional technology, can be used in a larger working area, and achieves good results. The extracted scatterers have better integrity, are not damaged, and have higher accuracy in scatterer extraction.

[0025] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 Schematic diagrams of the present invention before and after the experiment, (a) is a three-layer velocity model, (b) is a schematic diagram of synthetic seismic records;

[0027] Figure 2 The structural image of the wave number energy of the present invention, (a) is the structural image obtained by forward transformation, and (b) is the structural image obtained by inverse transformation;

[0028] Figure 3 Schematic diagram of the dip angle portion of the spectrum of the present invention, wherein (a) is a schematic diagram of the positive dip angle portion of the spectrum, and (b) is a schematic diagram of the negative dip angle portion of the spectrum;

[0029] Figure 4 This is the scattered energy imaging diagram of the present invention;

[0030] Figure 5 The pictures before and after the profile experiment at a depth of 9200 meters of the embodiment of the present invention, (a) is the profile at 9200 meters, and (b) is the result of 3D wave number domain scattering separation processing;

[0031] Figure 6 This is a diagram showing the energy suppression effect of the attenuation filter of the present invention for small inclination angles in two dimensions, wherein (a) is a schematic diagram before suppression, and (b) is a schematic diagram after suppression. DETAILED DESCRIPTION

[0032] The technical solution of the present invention is further described below by means of the accompanying drawings and embodiments. It should be noted that unless otherwise specifically stated, the relative arrangement of components and steps, numerical expressions and numerical values ​​described in these embodiments do not limit the scope of the present invention.

[0033] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application, or uses.

[0034] Technologies, systems, and devices known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, systems, and devices should be considered part of the specification.

[0035] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0036] Unless otherwise defined, technical or scientific terms used in the present invention shall have the common meanings understood by one having ordinary skills in the field to which the present invention belongs.

[0037] The present invention uses a scattering point identification technology based on the wave number domain after stacking to identify the scattering points that exist independently or are hidden under the reflection.

[0038] The dip angle range of the scattering point is wider than that of the reflection. After changing to the wave number domain, the dip angle of the scattering point will appear in both the positive and negative parts, while the dip angle of the continuous reflection interface will only exist in the positive or negative part. The image of the scattering point can be obtained by inversely changing the wave number domain of different parts to the time domain and applying the multiplication imaging condition to the obtained positive and negative images. This method can be effectively used in three-dimensional data volumes.

[0039] The present invention provides a scattering point identification technology method based on wave number domain, comprising:

[0040] S1, obtaining a three-dimensional post-stack data volume, analyzing and transforming the three-dimensional post-stack data volume into a wavenumber domain to obtain wavenumber domain data;

[0041] S2, using an attenuation filter to filter the wave number domain data to obtain wave number energy in different directions;

[0042] S3, performing inverse transformation on the wave number energies in different directions to obtain corresponding structural images in different directions;

[0043] S4. Apply multiplication imaging conditions to the time domain positive spectrum data and the time domain negative spectrum data to obtain an image of the scattering point.

[0044] Preferably, the attenuation filter mentioned in S2 is an improved rectangular attenuation filter, which can effectively separate the wavenumber domain spectrum energy and suppress the artifacts caused by the truncation effect, preserve and separate the spectrum energy and avoid the dispersion phenomenon caused by the truncation effect.

[0045] The scattering points are globally distributed in the spectrum, while for images with specific inclination angles, their distribution in the spectrum shows a certain regularity. The two-dimensional spectrum interval is filtered to obtain structural images with different inclination angles that conform to the inclination law, so as to enhance the scattered energy and suppress the energy of the continuous medium. The inclination component reflected by an image point of a continuous medium exists in the positive inclination structural image or the negative inclination structural image, but not in both at the same time. The sample-by-sample multiplication of the two structural images weakens the reflection event and enhances the diffraction energy. In other words, under full illumination conditions, diffraction radiates in all directions, ranging from -90° to 90°, so their energy can exist in two structural images, while reflection can only exist in one of the structural images. In order to reasonably apply this law, the present invention designs a new attenuation rectangular filter, which can not only separate the spectrum energy but also avoid the dispersion phenomenon caused by the stage effect.

[0046] The expression for filtering the wave number domain data with an attenuation filter to obtain the wave number energy in different directions is:

[0047] H' i (k) = H i (k)×A x ×A y ×A z ;

[0048]

[0049] Among them, A x is the attenuation operator in the X direction, A y is the Y direction attenuation operator, A z is the Z-direction attenuation operator, H i (k) is the wave number domain data, H' i (k) is the wave number energy after attenuation in different directions, f0<f<f1, f1<f<f2 and f2<f<f3 are different attenuation intervals selected in different directions, a is the formula corresponding to the attenuation interval as the attenuation operator, where the empirical value of a is 3.125, and the three-dimensional wave number domain data is divided into 8 parts, i=1, 2, 3, 4, i is the three-dimensional wave number domain data on different components, k is the wave number in different directions, where,

[0050]

[0051]

[0052] Among them, f0 is 0, f1 is selected according to the attenuation range that the user wants to achieve, the range from f1 to f2 is not attenuated to keep the original data, f3 is the maximum point value in different directions, if the part of the X-direction spectrum greater than zero has 100 points, the X-direction attenuation operator f3 is 100, if the part of the Y-direction spectrum greater than zero has 200 points, the Y-direction attenuation operator f3 is 200.

[0053] Preferably, in the case of one-dimensional improved attenuated rectangular filter, the expression used by the attenuation operator of the attenuated rectangular filter is:

[0054]

[0055] Among them, f0<f<f1, f1<f<f2 and f2<f<f3 are attenuation intervals, and the formula corresponding to the attenuation interval is the attenuation operator, where the empirical value of a is 3.125, and a=3.125 makes the minimum value of the attenuation result close to zero.

[0056] like Figure 6 As shown, in the two-dimensional case, the slope from each point in the spectrum to the two-dimensional center point is calculated, and the two-dimensional attenuation filter expression is constructed with the help of the one-dimensional attenuation filter expression. The expression used by the improved rectangular attenuation filter in the two-dimensional case is:

[0057]

[0058] Among them, k = (z-iz) / (x-ix), x, z are the coordinates of wave numbers in different directions, ix, iz are the coordinates of the center point of the wave number domain, g0<k<g1, g1<k<g2 is the two-dimensional slope attenuation interval, and A is the attenuation operator.

[0059] Set the slope range to f0 to f2, k is the slope from each point in the wavenumber domain to the center point, f0 is the minimum slope 0 at the beginning, f1 is the manually controlled slope attenuation interval, if you want to attenuate the slope K = 0 to 0.2, then take 0.2, f2 is the maximum slope value, generally max_z-iz / max_x-ix.

[0060] Preferably, the inverse transformation of the wave number energy in different directions to obtain the corresponding expression of the structural image in different directions is:

[0061]

[0062] in, For structural images in different directions, the expression of the image of the scattering point obtained by applying the multiplication imaging condition to the time domain positive spectrum data and the time domain negative spectrum data is:

[0063]

[0064] in, is the number of spectra in different directions, I(x,y,z) diff is the image of the scatter points.

[0065] The inclination of a continuous medium is the inclination of a specularly reflecting medium, and exists in an image with only negative or only positive inclination. However, the inclination of a discontinuous medium exists in both positive and negative inclination images. The positive and negative inclination parts of the discontinuous and reflected image points are multiplied, and the reflected image points are shown as zero, while the discontinuous image points are imaged. The inclination can be expressed as:

[0066] α=(θ r -θ s ) / 2;

[0067] In the formula, θ s is the incident angle, θ r Represents the angle of incidence, and the inclination angle is defined as the angle from the normal direction of the reflection point to the vertical axis.

[0068] After the seismic waves propagate to the scattering point, they are recorded on different receivers. If the acquisition aperture is infinite, the full range of dip angles of the imaging point can be covered, that is, the dip angle coverage of the discontinuous medium is wider than the mirror reflection dip angle coverage of the continuous medium. This is the difference between diffraction and reflection in the dip angle set. By multiplying the two traces point by point, the image of the scattering point will appear, and the energy will be amplified.

[0069] After stacking, the data is directly transformed into the wavenumber domain, and an attenuation filter is used to filter out the two parts with positive and negative dip angles. The image of the discontinuous medium at the same position can be obtained in both the positive and negative dip images. However, the reflection image at the same position will only be displayed in one image with positive or negative dip.

[0070] For two-dimensional data, in the wave number domain, negative inclination angles include (kz>0, kx>0) and (kz<0, kx<0), and positive inclination angles include (kz>0, kx<0) and (kz<0, kx>0). After inverse transformation of the positive and negative inclination angles, two images with different structures are obtained, which are called positive structure image and negative structure image.

[0071] After applying the multiplication imaging condition to the two structure images, the continuous reflection energy can be separated from the discontinuous medium scattering energy. For three-dimensional data,

[0072] The present invention needs to obtain four components of a structural image with different inclinations and azimuths, rather than a structural image with only negative inclinations and positive inclinations.

[0073] Therefore, in the wave number domain, the first part includes (kz>0,ky>0,kx>0) and (kz<0,ky<0,kx<0), the second part includes (kz>0,ky>0,kx<0) and (kz<0,ky<0,kx>0), the third part includes (kz>0,ky<0,kx>0) and (kz<0,ky>0,kx<0), and the fourth part includes (kz>0,ky<0,kx<0) and (kz<0,ky>0,kx>0).

[0074] By inverting the different parts, four structural components can be obtained. Finally, the different types of energy can be separated by multiplying the four parts.

[0075] The present invention designs an experiment and proposes a numerical simulation to prove the effectiveness and convenience of this method. Figure 1 (a) shows a three-layer velocity model, with the first layer being 2000 m / s, the second layer being 2200 m / s, and the third layer being 2300 m / s. In the first layer, the present invention buries five discontinuities of different shapes and scales, with a velocity 1.1 times that of the background velocity. In the reflection layers of the second and third layers, scattering points of the same scale and shape as the first layer are also buried, with a velocity 1.13 times that of the background velocity.

[0076] During the experiment, the present invention uses 20 Hz Ricker wavelet to convolve with the reflection coefficient to obtain a synthetic seismic record such as Figure 1 The seismic record is transformed into the wave number domain to obtain the spectrum, and the attenuation filter is used to obtain the positive (kz>0, kx>0) and (kz<0, kx<0), negative (kz>0, kx<0) and (kz<0, kx>0) dip parts of the spectrum as shown in Figure 3 (a) and Figure 3 (b). The two parts are inversely transformed to obtain positive and negative structural images as shown in Figure 2 (a) and Figure 2 (b). Figure 4 As shown, by taking the two structure images as input and bringing into the multiplication imaging condition, the reflected and scattered energies can be separated.

[0077] The present invention applies this method to actual 3D data, and the data size nx*ny*nz is 3000*2500*750, wherein the intervals in the x and y directions are both 25 meters, and the interval in the z direction is 20 meters.

[0078] The present invention selects a section with a depth of 9200 meters for demonstration, such as Figure 5 The corresponding scattering energy separation result is shown in (a). Figure 5 As shown in (b).

[0079] The experimental method of the present invention is based on the wave number change of post-stack data, so that the non-continuous medium image can be obtained efficiently, which plays an important role in the interpretation and positioning of oil and gas reservoirs in actual exploration. The method has the advantages of simple algorithm and fast calculation speed, and has achieved good results in actual processing, which has promoted the development direction of precise seismic interpretation work.

[0080] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A scattering point identification method based on wave number domain, characterized in that: include: S1, obtaining a three-dimensional post-stack data volume, analyzing and transforming the three-dimensional post-stack data volume into a wavenumber domain to obtain wavenumber domain data; S2, using an attenuation filter to filter the wave number domain data to obtain wave number energy in different directions; S3, performing inverse transformation on the wave number energies in different directions to obtain corresponding structural images in different directions; S4. Apply multiplication imaging conditions to the time domain positive spectrum data and the time domain negative spectrum data to obtain an image of the scattering point.

2. The scattering point identification method based on wave number domain according to claim 1 is characterized in that: The attenuation filter mentioned in S2 is an improved rectangular attenuation filter, which can effectively separate the wave number domain spectrum energy and suppress the artifacts caused by the truncation effect, preserve and separate the spectrum energy and avoid the dispersion phenomenon caused by the truncation effect.

3. The scattering point identification method based on wave number domain according to claim 2 is characterized in that: The expression for filtering the wave number domain data with an attenuation filter to obtain the wave number energy in different directions is: H i '(k)=H i (k)×A x ×A y ×A z ; Among them, A x is the attenuation operator in the X direction, A y is the Y direction attenuation operator, A z is the Z-direction attenuation operator, H i (k) is the wave number domain data, H′ i (k) is the wave number energy after attenuation in different directions, f0<f<f1, f1<f<f2 and f2<f<f3 are different attenuation intervals selected in different directions, a is the formula corresponding to the attenuation interval as the attenuation operator, where the empirical value of a is 3.125, and the three-dimensional wave number domain data is divided into 8 parts, i=1, 2, 3, 4, i is the three-dimensional wave number domain data on different components, k is the wave number in different directions, where, 4. The scattering point identification method based on wave number domain according to claim 3 is characterized in that: The inverse transformation of the wave number energy in different directions to obtain the corresponding structural images in different directions is as follows: in, Structural images in different directions.

5. The scattering point identification method based on wave number domain according to claim 4 is characterized in that: The expression of the image of the scattering point obtained by applying the multiplication imaging condition to the time domain positive spectrum data and the time domain negative spectrum data is: in, is the number of spectra in different directions, I(x,y,z) diff is the image of the scatter points.

Images