A method for automatically extracting petroleum seismic surface wave dispersion curve

Abstract

The present application relates to the field of oil seismic exploration method, disclose a kind of oil seismic surface wave dispersion curve automatic extraction method, including prestack single shot data preprocessing, dispersion energy spectrum calculation, threshold transformation processing, mask image morphological dilation processing, mask small area object removal processing, mask morphological corrosion processing, mask and dispersion spectrum point multiplication (Hadamard product) operation, dispersion automatic picking.The present application uses image threshold transformation and morphological operation to process the dispersion energy spectrum of oil seismic surface wave, suppresses other types of wave field except surface wave in the energy spectrum of frequency-velocity domain, retains the dispersion band information of surface wave, and automatically identifies and extracts dispersion curve from the image containing only surface wave dispersion energy band, so as to realize the automatic measurement of dispersion curve.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of oil seismic exploration method, especially to the field of seismic surface wave data processing and analysis, in particular to an automatic extraction method of oil seismic surface wave dispersion curve. BACKGROUND

[0002] Surface wave is a kind of seismic wave field propagating along the free surface of the earth, and the wave propagation velocity of different frequency components is different in the propagation process, which has the characteristic of dispersion. Theory and practice have shown that the dispersion characteristics of surface wave are controlled by the velocity structure of the medium below the propagation path, and the internal structure of the medium can be obtained by using the dispersion characteristics of surface wave. Therefore, surface wave detection methods are widely used in different fields such as engineering investigation, resource exploration and imaging of the internal structure of the earth.

[0003] Surface wave in oil seismic data usually occupies the strongest energy, and is often regarded as interference and noise in reflection wave exploration. However, the characteristics of surface wave are closely related to the near-surface cover layer. In oil exploration, the cover layer is relatively loose, and there is no obvious wave impedance interface in the cover layer, so it is difficult to obtain information of the cover layer by using reflection wave. However, the cover layer has obvious absorption and attenuation effect, and it is of great significance to improve the accuracy and resolution of exploration by finding out the structure of the cover layer. Using surface wave information in oil seismic data for inversion is a powerful means to establish the structure of the cover layer.

[0004] In the process of using surface wave to establish the structure of the cover layer, the dispersion characteristics of surface wave are usually used to extract the dispersion curve, and then the dispersion curve is inverted to obtain the velocity structure of the underground. The extraction of dispersion curve is usually obtained by f-k transform, τ-p transform and other wave field transform methods to obtain the dispersion energy spectrum of single shot seismic record, and then the phase velocity corresponding to the point with the strongest energy is selected from the dispersion energy spectrum by artificial identification. When the exploration area is large, hundreds or even thousands of shot seismic records are usually collected. If the dispersion curve is still extracted from the common shot point gather of each shot by artificial identification method, it will be a task with extremely large workload. Although surface wave occupies a large energy in the dispersion energy spectrum and shows a banding feature, the phase velocity of surface wave can be automatically extracted by selecting the maximum point on each frequency section. However, there are other types of wave fields such as reflection wave, direct wave and refracted wave in seismic record, and other energy groups except surface wave will appear in the dispersion energy spectrum, which will cause problems in automatic extraction of phase velocity. Therefore, how to suppress or eliminate the energy groups formed by reflection wave, direct wave and other non-surface wave fields in the dispersion energy spectrum to form a simple and single surface wave dispersion energy band image is the key to realize the automatic extraction of dispersion curve. SUMMARY

[0005] In view of the above-mentioned deficiencies in the prior art, the present invention aims to provide a method for automatically extracting petroleum seismic surface wave dispersion curves.

[0006] To achieve the above objectives, the present invention adopts the following technical solutions:

[0007] The present invention provides a method for automatically extracting a petroleum seismic surface wave dispersion curve, comprising the following steps:

[0008] Step 1: Extract the gather from the single shot seismic record and extract the record of surface wave distribution. If the original single shot record is recorded as U(x, t), then the extracted gather record containing surface wave is recorded as U s (x, t), where x represents the distance between the geophone and the earthquake source, and t represents the recording time of each geophone;

[0009] Step 2: Record the extracted gather containing surface waves U s (x, t) is transformed into τ-p, and the transformation result of τ-p domain is mapped to fv domain to form the dispersion energy spectrum E s (f,v);

[0010] Step 3: Dispersion energy spectrum E s The (f, v) image is thresholded, with the threshold being p. Points in the spectrum with energy amplitudes higher than p are marked as 1, and points with energy amplitudes lower than p are marked as 0. The image formed after the thresholding transformation, which only contains 0 / 1 values, is called mask M.

[0011] Step 4: Perform morphological dilation on the mask M. The mask M after dilation becomes M p ;

[0012] Step 5: Dilate the mask M p The part with a value of 0 is regarded as the background, and the part with a value of 1 is regarded as the area with a shape. p The small area in the middle, only the area with the largest area is retained, forming a mask M that only retains one connected area area s , at this time M s The connected area in the middle is the corresponding surface wave dispersion energy band range in the dispersion spectrum;

[0013] Step 6: Mask M s Perform morphological corrosion processing, and the mask after corrosion processing is recorded as M d ;

[0014] Step 7: Let mask M d The original dispersion energy spectrum image E before threshold transformation s (f, v) performs matrix dot multiplication, and the result is recorded as E d (f,v), calculated E d(f,v) the value of the energy strip area of the medium frequency and the E s (f,v) the value of the energy strip area of the medium frequency and the E

[0015] Step eight: from E d (f,v) automatically determine each frequency f i corresponding to the maximum energy A(f i ), A(f i ) corresponding to the velocity v i is the phase velocity value of the surface wave of frequency f i , and the set composed of all f i and v i forms the dispersion curve.

[0016] Preferably, in step three, the threshold value p is taken as 0.9 times the amplitude value of the largest point in the dispersion energy spectrum.

[0017] Further, step three is to generate a mask of the same size and dimension as the dispersion energy spectrum image obtained in step two, and assign values to the pixels at different positions in the mask, and generate a 0 / 1 value mask M through threshold processing.

[0018] Further, the morphological dilation processing in step four uses a flat disc structure with a radius of 3 as the structure function of the dilation processing.

[0019] Further, in step five, the small area region is removed, and the region with a pixel value of 1 and mutual connectivity in the mask is regarded as a small whole, and the area of each connected region in the mask is counted, and the area size is the number of all pixels in the region.

[0020] Further, the morphological erosion processing in step six uses a flat disc structure with a radius of 3 as the structure function of the erosion processing.

[0021] Further, in step seven, the mask is used to "screen" the original dispersion energy spectrum, and the dispersion energy corresponding to the position with a pixel value of 1 in the mask is retained, and the dispersion energy at the remaining positions is set to zero, and the obtained result is a dispersion energy spectrum containing only the surface wave dispersion strip, which provides a basis for automatic extraction of the dispersion curve.

[0022] Compared with the prior art, the present application has the following beneficial effects

[0023] (1) The present application uses image threshold transformation and morphological operation to process the dispersion energy spectrum of the oil seismic data containing surface waves, suppresses other types of wave fields except surface waves in the energy spectrum in the frequency-velocity domain, retains the dispersion strip information of the surface waves, and automatically identifies and extracts the dispersion curve from the image containing only the surface wave dispersion energy strip, thereby realizing the automatic measurement of the dispersion curve.

[0024] (2) The dispersion curve automatic measurement method provided by the application greatly improves the automation degree of data processing compared with the conventional manual dispersion curve extraction method, and also improves the dispersion energy resolution and convergence of surface waves. BRIEF DESCRIPTION OF DRAWINGS

[0025] Other features, objects, and advantages of the application will become more apparent from the following detailed description of non-limiting embodiments, when read in conjunction with the accompanying drawings:

[0026] Figure 1 A flow chart of the method of the application;

[0027] Figure 2 A schematic diagram of the original single-shot seismic record in Example 2;

[0028] Figure 3 A schematic diagram of the seismic record containing surface waves obtained by channel set extraction in Example 2;

[0029] Figure 4 A schematic diagram of the original dispersion energy spectrum in Example 2;

[0030] Figure 5 A schematic diagram of the mask after threshold transformation in Example 2;

[0031] Figure 6 A schematic diagram of the result of morphological dilation operation on the mask in Example 2;

[0032] Figure 7 A schematic diagram of the result after retaining the largest area region of the mask in Example 2;

[0033] Figure 8 A schematic diagram of the result of morphological erosion operation on the mask in Example 2;

[0034] Figure 9 A schematic diagram of the dispersion energy spectrum containing only surface waves and the automatically picked dispersion curve in Example 2. DETAILED DESCRIPTION

[0035] The application will be described in detail below with specific embodiments. The following examples will help those skilled in the art to further understand the application, but do not limit the application in any form. It should be noted that for those skilled in the art, without departing from the concept of the application, a number of modifications and improvements can be made. These are within the scope of the application.

[0036] Example 1

[0037] As Figure 1As shown, a method for automatically extracting the dispersion curve of surface waves in seismic data, comprising the following specific steps:

[0038] Step one: extract the gather from single shot seismic record, extract the record of surface wave distribution, because the surface wave propagation speed is low, in the far channel has exceeded the record length, so the surface wave is usually located in the tens of channels of the adjacent shot point on the side of the shot point, if the original single shot record is U(x,t), the extracted record containing surface wave is U s (x,t), wherein subscript s represents surface wave, x represents the distance between the geophone and the source, t represents the recording time of each geophone, for example: U(x i ,t i ) represents the amplitude of the given geophone x i at t i time.

[0039] Note that U s (x,t) also contains other types of wave fields such as first arrival and reflection.

[0040] Step two: perform τ-p transform on the extracted record containing surface wave U s (x,t) and map the transform results in τ-p domain to f-v domain to form the dispersion energy spectrum E s (f,v).

[0041] Step three: threshold transform the image of dispersion energy spectrum E s (f,v), let the threshold value be p, the points with energy amplitude higher than p in the spectrum are recorded as 1, and the points with energy amplitude lower than p are recorded as 0, the threshold value p can be taken as 0.9 times the amplitude of the largest point in the dispersion energy spectrum. The dispersion energy spectrum is regarded as an image, and the amplitude value of each point is regarded as the pixel value of the image. Through the above threshold transform processing, the image pixel value corresponding to the points with amplitude greater than p in the dispersion energy spectrum becomes 1, and the image pixel value corresponding to the remaining points becomes 0, that is, an image containing only 0 / 1 values is obtained, which is denoted as mask M.

[0042] Step four: perform morphological dilation operation on the mask M, and the dilation structure kernel function during processing is taken as a flat disc structure with a radius of 3, and the mask M after dilation processing becomes M p , which aims to improve the connectivity of the corresponding dispersion energy area in the mask and fill the small cavities, which will also slightly expand the energy group of other types of wave fields.

[0043] Step five: dilate the mask M pThe part with value 0 is regarded as background, and the part with value 1 is regarded as a shaped area region. The region with pixel value 1 and mutual connection in the image can be regarded as a small whole. Then the area of each whole connected region in the mask is counted. The area size is the number of all pixels with value 1 in the region. Then the small area region is removed (i.e. the pixel value in the region is set to zero) according to the counting result, only the largest area region is reserved, and finally the mask M is formed which only reserves one largest connected region s At this time, the connected area region in M s is the corresponding surface wave dispersion energy band in the dispersion spectrum.

[0044] Step six: morphological erosion processing is performed on the mask M s . The erosion structure kernel function is a flat disc structure with a radius of 3 during processing. The mask after erosion processing is denoted as M d . The purpose is to make the corresponding dispersion energy region in the mask more convergent.

[0045] Step seven: the mask M d is multiplied with the original dispersion energy spectrum image E s (f,v) by matrix point multiplication (Hadamard product) operation. The original dispersion energy is "screened" by the mask. The dispersion energy corresponding to the position with value 1 in the mask is reserved, and the dispersion energy of the remaining positions is set to zero. The obtained result is an amplitude spectrum containing only the surface wave dispersion band. The obtained result is denoted as E d (f,v). The dispersion energy band region in E d (f,v) is consistent with the dispersion energy band in E s (f,v) and remains unchanged, and the values of the remaining regions are all changed to 0.

[0046] Step eight: the maximum energy A(f d ) of each frequency f i section in E i (f,v) is automatically judged. The velocity v i corresponding to A(f i ) is the phase velocity value of the surface wave with frequency f i . The set composed of all f i and v i constitutes the dispersion curve.

[0047] Example 2

[0048] As shown in Figure 1 , an automatic extraction method of the oil seismic surface wave dispersion curve includes the following specific steps:

[0049] Step one: Figure 2For a single shot seismic record in a real seismic exploration data of oil, the record channel number is 360, the shooting mode is middle shot, the shot point is between 180-181 channel, the distance between the shot point and the adjacent geophone is 40m, the distance between the adjacent geophones of the rest channels is 20m, the sampling rate of each channel is 500Hz, and the sampling length is 7s. From the record, the obvious surface wave wave field record can be seen. The adjacent 48 channels (181-228 channels) on the right side of the shot point containing the complete surface wave record are extracted from the record, and recorded as U s (x,t) is shown in Figure 1. Figure 3

[0050] Step two: τ-p transform operation is performed on U s (x,t), and the transform result in the τ-p domain is mapped to the f-v domain to form the frequency dispersion energy spectrum E s (f,v), as shown in Figure 2. Figure 4 From the figure, the surface wave dispersion band and the energy group formed by other wave fields can be seen.

[0051] Step three: threshold transform is performed on the image of the frequency dispersion energy spectrum E s (f,v). First, a mask with the same dimension as the image of E s (f,v) is generated, and each pixel point in the mask corresponds to a point in E s (f,v), and then the threshold is set to 0.9, that is, 0.9 times of the maximum point of the entire frequency dispersion energy spectrum. If the amplitude of the same position point on the frequency dispersion spectrum corresponding to the pixel in the mask is higher than 0.9 times of the maximum value, the pixel is assigned a value of 1, and if it is lower than 0.9 times of the maximum value, the pixel is assigned a value of 0. In this way, a mask M containing only 0 / 1 values is obtained, as shown in Figure 3. Figure 5

[0052] Step four: morphological dilation operation is performed on the mask M, and the dilation structure kernel function is taken as a flat disc structure with a radius of 3. After the dilation processing, the mask M becomes M p , as shown in Figure 4. Figure 6 From the figure, it can be seen that the connectivity of each energy group in the mask after the dilation operation becomes better.

[0053] Step five: the part with a value of 0 in the dilated mask M p is regarded as background, and the part with a value of 1 is regarded as a shaped area region. Small area regions in M p are deleted, and only the largest area region is retained, recorded as M s , as shown in Figure 5. Figure 7 The connected area region in M s is the corresponding surface wave dispersion energy band range in the frequency dispersion spectrum.

[0054] Step six: morphological opening operation is performed on the mask M s ​​Morphological corrosion is performed, and a flat disc structure with a radius of 3 is taken as a corrosion structure kernel function during the processing, and the mask after the corrosion processing is denoted as M d As shown in Figure 8 , it can be seen that the energy area of dispersion is more convergent.

[0055] Step seven: the mask M d is multiplied by the original energy spectrum image E s (f,v) before threshold transformation, and the result is denoted as E d (f,v), as shown in Figure 9 , it can be seen that the energy band of surface wave dispersion in the frequency-velocity domain.

[0056] Step eight: each frequency f d corresponding to the energy maximum value A(f i ) is automatically identified from E i (f,v) with a frequency interval of 0.5 Hz, and the velocity v i corresponding to A(f i ) is the phase velocity value of the surface wave with the frequency f i , and the position corresponding to (f i ,v i ) is plotted with a red star in Figure 9 , and the phase velocity points of each frequency are connected by a smooth curve, and the dispersion curve is obtained.

[0057] The specific embodiments of the present application are described above. It should be understood that the present application is not limited to the above specific embodiments, and those skilled in the art can make various modifications or changes within the scope of the claims, which does not affect the essential content of the present application.

Claims

1. A method for automatically extracting seismic surface wave dispersion curve of oil, characterized in that, It comprises the following steps: Step one: extract the gather from single shot seismic record, extract the record of surface wave distribution, if the original single shot record is U(x, t), the extracted gather record containing surface wave is U s (x, t), where x represents the distance between the geophone and the source, t represents the recording time of each geophone; Step two: record U of the extracted surface wave containing gather s (x, t) τ-p transform, and the τ-p domain transform results are mapped to f-v domain, forming the dispersion energy spectrum E s (f, v); Step three: threshold transform the image of the frequency dispersion energy spectrum E s (f,v) with a threshold value p, the points with energy amplitude higher than p are recorded as 1, and the points with energy amplitude lower than p are recorded as 0, the image formed after threshold transform containing only 0 / 1 values is recorded as a mask M; Step four: morphological dilation operation is performed on the mask M, and the mask M after the dilation operation becomes M p ; Step five: the mask M p The part with value 0 is regarded as background, and the part with value 1 is regarded as a shaped area region. Small area regions in M p are deleted, and only the largest area region is reserved to form a mask M s with only one connected area region. At this time, the connected area region in M s is the corresponding surface wave dispersion energy strip range in the dispersion spectrum. The small area region removal takes the region with pixel value 1 and mutual connection in the mask as a small whole, counts the area of each connected region in the mask, and the area size is the number of all pixels in the region; Step six: morphological erosion is performed on the mask M s , and the mask after the erosion is denoted as M d ; step seven: matrix point multiplication is performed between the mask M d and the original frequency dispersion energy spectrum image E s (f,v) before threshold transformation, and the result is denoted as E d (f,v); in the calculated E d (f,v), the values of the dispersion energy band region are consistent with E s (f,v), and the values of the remaining regions are all changed to 0. Step eight: From E d (f,v) automatically determine each frequency f i corresponding to the maximum energy A(f i ), A(f i ) corresponding to the velocity v i is the phase velocity value of the surface wave of frequency f i , and the set composed of all f i and v i constitutes the dispersion curve.

2. The method for automatically extracting the dispersion curve of the seismic surface wave of oil according to claim 1, characterized in that, In step three, the threshold p is taken as 0.9 times the amplitude of the point with the maximum amplitude in the frequency dispersion energy spectrum.

3. The method for automatically extracting the petroleum seismic surface wave dispersion curve according to claim 1, characterized in that, Step three is to generate a mask with the same size and dimension as the frequency dispersion energy spectrum image obtained in step two, and assign values to the pixels at different positions in the mask. Through threshold processing, a mask M with 0 / 1 values is generated.

4. The method for automatically extracting the petroleum seismic surface wave dispersion curve according to claim 1, characterized in that, In step four, the morphological dilation processing uses a flat disc structure with a radius of 3 as the structure function of the dilation processing.

5. The method for automatically extracting the petroleum seismic surface wave dispersion curve according to claim 1, characterized in that, In step six, the morphological erosion processing uses a flat disc structure with a radius of 3 as the structure function of the erosion processing.

6. The method for automatically extracting the petroleum seismic surface wave dispersion curve according to claim 1, characterized in that, In step seven, the mask is used to "screen" the original frequency dispersion energy spectrum. The frequency dispersion energy corresponding to the positions with pixel value 1 in the mask is retained, and the frequency dispersion energy at the remaining positions is set to zero. The result is a frequency dispersion energy spectrum containing only the surface wave dispersion strip, providing a basis for automatic extraction of the dispersion curve.

Images