A fluvial facies reservoir along-river prestack gather data optimization processing method and device

Abstract

The embodiment of the present application discloses a river facies reservoir along-river prestack gather data optimization processing method and device, provides prestack gather data; depicts the spatial distribution form of the river channel, and applies it to the prestack gather data to obtain along-river prestack gather data; the along-river prestack gather data is denoised to obtain denoised along-river prestack gather data; the amplitude of the denoised along-river prestack gather data is corrected to obtain corrected gather data; the angle gather data is obtained by changing the angle of the corrected gather data; the angle gather data is preprocessed to obtain the flattening correction gather data. The present application can improve the accuracy and efficiency of the prestack seismic inversion of the river facies reservoir by depicting the spatial distribution form of the river channel, applying it to the prestack gather data to obtain along-river prestack gather data, and further performing denoising, flattening correction and other steps.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of oil and gas geophysical exploration, and more particularly, relates to a method and device for optimizing river facies reservoir along-river prestack gather data. BACKGROUND

[0002] According to statistical data, the oil reserves in river facies reservoirs account for more than 50% of the currently developed oil reserves in China, among which, meandering river point bar facies reservoirs account for a large proportion; and braided river alluvium reservoirs constitute some world-class large oil and gas fields, which shows that most large reservoirs are river facies or river-delta facies. Therefore, the research on river and river-delta, especially the research on meandering river and braided river, is a basic and key work related to the rational development and utilization of oil and gas resources of each oilfield.

[0003] Prestack seismic inversion is a main means for predicting river facies reservoirs and detecting fluids. At present, the development of prestack inversion technology still mainly focuses on correcting and perfecting various inversion algorithms, and often ignores the quality of inversion input data. If poor seismic data is used, even if advanced and complex prestack inversion algorithms are used, the final inversion effect will be affected. Therefore, the prestack gather needs to be processed more finely, and the common reflection point gather (CRP) needs to be more flat, the dynamic correction stretching effect needs to be smaller, and the signal-to-noise ratio needs to be higher while ensuring the relative strength relationship of reflection amplitudes.

[0004] In recent years, many optimization processing techniques and methods have been proposed by domestic and foreign scholars, such as the prestack seismic data regularization processing technique and the prestack random noise attenuation technique proposed by Wu Changyu et al., which solve the problem of prestack migration imaging in different work areas; high-density velocity analysis technology is used by Cheng Yunkun et al. to realize CRP gather without moveout stacking; Liu Suxu uses the step-by-step approximation method to process prestack gathers for depth migration imaging for complex structures; Zhang Zheng et al. propose anisotropic dynamic correction to process the residual moveout in the gather; Xu Zilong proposes a waveform-dependent residual moveout correction and a time-space variable wavelet threshold fidelity denoising method based on horizontal sliding, which improves the accuracy of residual moveout correction and better ensures the fidelity in the denoising process; Xu Chi et al. propose a CRP gather optimization processing technique based on structure median filtering; Zhou Peng et al. propose an absolute value cross-correlation gather flattening method independent of residual moveout, which can better flatten the prestack seismic gather and remove far-gather waveform distortion.

[0005] However, these methods can only improve gather quality to a certain extent for specific characteristic problems, and most of them ignore the different combinations of various gather optimization methods. A reasonable combination can better optimize the gather. In addition, fluvial reservoirs have spatial heterogeneity. Oilfield developers only focus on the lithology, physical properties and hydrocarbon content within the fluvial facies zone, and do not pay much attention to the geological conditions of the sedimentary background outside the facies zone. Therefore, the above methods cannot be well used to improve the accuracy and efficiency of pre-stack seismic inversion of fluvial reservoirs. Summary of the Invention

[0006] In view of this, embodiments of the present invention provide a method and apparatus for optimizing the processing of pre-stack gather data of fluvial reservoirs, which at least solves the technical problems of low accuracy and poor efficiency in pre-stack seismic inversion of fluvial reservoirs in the prior art.

[0007] In a first aspect, embodiments of the present invention provide a method for optimizing and processing pre-stack gather data of fluvial facies reservoirs, including:

[0008] Provide pre-stack gather data;

[0009] The spatial distribution morphology of the river channel is characterized and applied to the pre-stack gather data to obtain pre-stack gather data along the river channel.

[0010] The pre-stack gather data along the river channel is denoised to obtain denoised pre-stack gather data along the river channel.

[0011] The amplitude of the denoised pre-stack gather data along the river channel is corrected to obtain corrected gather data.

[0012] Angle gather data is obtained by changing the angle of the corrected gather data;

[0013] The angle gather data is preprocessed to obtain flattened and corrected gather data.

[0014] Optionally, before obtaining the pre-stack gather data along the river channel, the following steps are included:

[0015] Seismic data volume extraction includes data volume scanning and data volume indexing.

[0016] Optionally, the step of correcting the amplitude of the denoised pre-stack gather data along the river channel to obtain corrected gather data includes:

[0017] All the denoised pre-stack gather data along the river channel are stacked into gather data according to different offsets. Then, each pre-stack gather data is fitted with the gather data by least squares to obtain the corrected gather data.

[0018] Optionally, before obtaining the corrected gather data, the process includes:

[0019] least square fitting each of the pre-stack gather data with the gather data to obtain a correction coefficient.

[0020] Optionally, the expression of the correction coefficient is:

[0021] wherein t represents time, h represents offset distance, seis_o(x, y, h, t) represents the denoised channel-aligned pre-stack gather data, seis(h, t) represents the gather data, and q(h, t) represents the correction coefficient.

[0022] Optionally, the expression of the corrected gather data seis_c(x, y, h, t) is:

[0023] wherein seis_o(x, y, h, t) represents the denoised channel-aligned pre-stack gather data, q(h, t) represents the correction coefficient, and seis_c(x, y, h, t) represents the corrected gather data.

[0024] Optionally, in the step of pre-processing the angle gather data, the pre-processing refers to performing flattening processing by using a gather flattening technique.

[0025] Optionally, after obtaining the angle gather data, the method further comprises:

[0026] performing super-gather processing on the angle gather data to improve lateral continuity of the angle gather data.

[0027] In a second aspect, the embodiments of the present application further provide a device for optimizing channel-aligned pre-stack gather data of fluvial facies reservoirs, comprising:

[0028] a pre-stack gather data module configured to provide pre-stack gather data;

[0029] a channel-aligned pre-stack gather data module configured to depict spatial distribution patterns of channels and apply the spatial distribution patterns to the pre-stack gather data to obtain channel-aligned pre-stack gather data;

[0030] a denoising module configured to perform denoising processing on the channel-aligned pre-stack gather data to obtain denoised channel-aligned pre-stack gather data;

[0031] a correction module configured to correct amplitudes of the denoised channel-aligned pre-stack gather data to obtain corrected gather data;

[0032] an angle gather data module configured to obtain angle gather data by changing angles of the corrected gather data;

[0033] a pre-processing module configured to pre-process the angle gather data to obtain flattened corrected gather data.

[0034] Optionally, before the river channel space distribution pattern is depicted, the method comprises:

[0035] The river channel prestack gather data module is also used to select river channel sensitive attributes, perform planar attribute extraction, and determine well site calibration and threshold values.

[0036] The present application can improve the precision and efficiency of river facies reservoir prestack seismic inversion by depicting the river channel space distribution pattern, applying it to the prestack gather data, obtaining the river channel prestack gather data, and further performing steps such as denoising and flattening correction.

[0037] Other features and advantages of the present application will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0038] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which like reference characters refer to like parts throughout the several views, and in which:

[0039] Figure 1 A flowchart of a river facies reservoir river prestack gather data optimization processing method of an embodiment of the present application is shown;

[0040] Figure 2 A river channel prestack gather data technical flowchart of an embodiment of the present application is shown;

[0041] Figures 3a-3b A comparison diagram of planar attributes before and after extracting river channel gather data of an embodiment of the present application is shown;

[0042] Figures 4a-4b A gather display diagram in the vertical river channel direction after extracting river channel prestack gather data in an embodiment of the present application is shown;

[0043] Figure 5 An efficiency diagram of subsequent steps after extracting river channel prestack gather data in an embodiment of the present application is shown;

[0044] Figure 6 A comparison diagram of river channel top interface interpretation effects before and after extracting river channel prestack gather data in an embodiment of the present application is shown;

[0045] Figure 7 A comparison diagram before and after denoising of river channel prestack gather data in an embodiment of the present application is shown;

[0046] Figures 8a-8b A comparison diagram before and after correction of river channel prestack gather data after denoising in an embodiment of the present application is shown;

[0047] Figure 9 Fig. 1 shows a schematic diagram of converting the denoised riverway prestack gather data into angle gather data according to an embodiment of the present application;

[0048] Figure 10 Fig. 2 shows a schematic diagram of the effect of super-gather processing according to an embodiment of the present application;

[0049] Figure 11 Fig. 3 shows a schematic diagram of the flattened corrected gather data according to an embodiment of the present application;

[0050] Figure 12 Fig. 4 shows a schematic diagram of the partial angle stack gather data according to an embodiment of the present application. DETAILED DESCRIPTION

[0051] The preferred embodiments of the present application will be described in more detail below. Although the preferred embodiments of the present application are described below, it is understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein.

[0052] A method for optimizing river facies reservoir riverway prestack gather data, comprising:

[0053] providing prestack gather data;

[0054] Specifically, the prestack gather data is a result data in the process of seismic data processing, and is generally CRP gather data, which is a professional term in the field and will not be described in detail here.

[0055] characterizing the spatial distribution of the riverway and applying it to the prestack gather data to obtain riverway prestack gather data;

[0056] Specifically, the characterization of the riverway is to determine the spatial position of the riverway, i.e., the XY coordinates and depth (t) of the riverway boundary, by using seismic attributes that can reflect the spatial distribution characteristics of the riverway, and then extracting the data located at the coordinates from the massive seismic data according to the determined riverway boundary coordinates XY.

[0057] performing denoising processing on the riverway prestack gather data to obtain denoised riverway prestack gather data;

[0058] correcting the amplitude of the denoised riverway prestack gather data to obtain corrected gather data;

[0059] obtaining angle gather data by changing the angle of the corrected gather data;

[0060] performing preprocessing on the angle gather data to obtain flattened corrected gather data.

[0061] Optionally, before obtaining the riverway prestack gather data, the method comprises the following steps:

[0062] Extracting the seismic data volume, wherein the seismic data volume comprises a data volume scan and a data volume index.

[0063] Optionally, the step of correcting the amplitude of the denoised riverway prestack gather data to obtain corrected gather data comprises the following steps:

[0064] Stacking all the denoised riverway prestack gather data according to different offsets to obtain gather data, and then performing least square fitting on each of the prestack gather data and the gather data to obtain corrected gather data.

[0065] Optionally, before obtaining the corrected gather data, the method comprises the following steps:

[0066] Performing least square fitting on each of the prestack gather data and the gather data to obtain a correction coefficient.

[0067] Optionally, the expression of the correction coefficient is as follows:

[0068] wherein t represents time, h represents offset, seis_o(x, y, h, t) represents the denoised riverway prestack gather data, seis(h, t) represents the gather data, and q(h, t) represents the correction coefficient.

[0069] Optionally, the expression of the corrected gather data seis_c(x, y, h, t) is as follows:

[0070] wherein seis_o(x, y, h, t) represents the denoised riverway prestack gather data, q(h, t) represents the correction coefficient, and seis_c(x, y, h, t) represents the corrected gather data.

[0071] Optionally, in the step of pre-processing the angle gather data, the pre-processing refers to performing flattening processing by using gather flattening technology.

[0072] Optionally, after obtaining the angle gather data, the method further comprises the following steps:

[0073] Performing super-gather processing on the angle gather data to improve the lateral continuity of the angle gather data.

[0074] Embodiment one:

[0075] This embodiment is described in detail by taking the middle-shallow layer Jurassic riverway sandstone reservoir in Sichuan Basin as an example.

[0076] Figure 1 A flow chart of the river facies reservoir riverway prestack gather data optimization processing method of this embodiment is shown.

[0077] S1: Obtain prestack gather data along the channel

[0078] The technical procedure for obtaining prestack gather data along the channel is as follows: starting from the optimization of channel sensitive attributes, then extracting plane attributes, determining well locations and threshold values, depicting the spatial distribution pattern of the channel, applying it to the prestack gather data, and obtaining the prestack gather data along the channel that can be used for subsequent processing and interpretation, as shown in Figure 2 .

[0079] Specifically, the channel is depicted by using seismic attributes that can reflect the spatial distribution characteristics of the channel, the spatial position of the channel, i.e., the XY coordinates and depth (t) of the channel boundary, is determined, and then the data located at the coordinates is extracted from the massive seismic data according to the determined channel boundary coordinates XY.

[0080] The seismic data volume belongs to massive data, and there are many algorithms for extracting the seismic data volume. In order to obtain a very high extraction speed, we need to improve the data retrieval speed and quickly locate the CDP point, which specifically includes data volume scanning and data volume indexing.

[0081] Data volume scanning:

[0082] Data volume scanning is to scan the entire data volume before loading the data into the database, scan the data volume related attribute information, and store it in the database in the form of structured data for easy data volume basic information query.

[0083] Data volume indexing:

[0084] In order to obtain a very high seismic data extraction speed, the index file can be established by scanning each trace of the data volume. The index file records the X, Y coordinates, inline, crossline number and absolute position of the trace data. In this way, if we obtain the X, Y coordinates of each segment of the fold line of the section to be extracted, we can find the index file corresponding to the data volume of the work area through the structured data information of the database, and through the X, Y coordinates, we can find the absolute position of the data volume of the trace corresponding to the coordinates. We can directly locate the position of the trace in the data volume to read the data, avoiding scanning the entire data volume. Since the index file is generally very small, it can be scanned quickly, so the data extraction speed can be doubled through data volume indexing.

[0085] The obtained prestack gather data along the channel has the following advantages in subsequent data processing and interpretation:

[0086] First, it can improve the efficiency of river facies reservoir prediction and fluid detection in gather processing, prestack inversion, attribute extraction, etc. Figure 5The efficiency comparison chart of several key processing explanation modules before and after the gather extraction in the embodiment is shown.

[0087] S2: denoising processing

[0088] The time migration of conventional river channel prestack gather data is usually targeted at post-stack seismic data, and the river channel prestack gather data usually has various noises, especially Gaussian noise, before stacking. Before angle conversion and partial stacking, the purpose of denoising processing is to improve the signal-to-noise ratio of the gather data and the signal-to-noise ratio of the final angle stacking data, and to remove random noise. There are many methods for removing random noise, such as median filtering, polynomial fitting, f-x domain prediction denoising, time-space variable wavelet threshold fidelity denoising based on horizontal sliding, CRP gather rolling stacking to form super-gather denoising, etc. In the embodiment, an improved median filtering method is used to remove random noise. The principle of this method is to pick up the sampling points of the same time of adjacent seismic traces in the river channel prestack gather data, and by cutting off these abnormal points, the noise influence can be reduced. The specific steps are as follows: first, the seismic data at a certain time in the river channel prestack gather data are numerically sorted, a part of the maximum value and the minimum value are removed, and then the arithmetic mean of the remaining sample points is calculated as the sampling point value of the current center trace at the current time. Figure 7 The comparison diagram before and after denoising of the river channel prestack gather data in the embodiment is shown.

[0089] The median filtering calculation formula is:

[0090]

[0091] Wherein, S represents seismic data, x, y, t are three dimensions of seismic data, representing X direction, Y direction and time direction t respectively. Mn is the size of the median filtering window function, m is the window size in X direction, and n is the window size in Y direction.

[0092] S3: obtaining corrected gather data

[0093] The denoised river channel prestack gather data usually has an amplitude relationship of "strong-weak-strong", and in theory, except for phase conversion, none of the four types of AVO has an amplitude relationship of "strong-weak-strong" with the increase of offset distance. The unfaithful AVO relationship is very unfavorable for subsequent AVO analysis and AVO inversion, and the reason for this phenomenon is often related to the acquisition system. The amplitude anomaly caused by the acquisition system is the first concern. In the technical solution, the three-dimensional observation system amplitude anomaly that is not completely eliminated in the migration process is processed on the common offset gather after migration.

[0094] The specific method is that all the de-noised riverway pre-stack gather data is stacked into a gather data according to different offset distances, and the gather data only contains time t and offset distance h information, and is recorded as seis(h, t).

[0095] The least square fitting is performed on each pre-stack gather data seis_o(x, y, h, t) and the gather data seis(h, t) to obtain a correction coefficient q(h, t).

[0096] The expression of the correction coefficient is:

[0097] f(q(h, t)) = min||seis(h, t)*q(h, t)-seis_o(x, y, h, t)||2

[0098] The corrected gather is:

[0099] seis_c(x, y, h, t) = seis_o(x, y, h, t)*q(h, t)

[0100] The general expression of the pre-stack gather data contains four dimensions, X direction, Y direction, time t and offset distance h, seis_o(x, y, h, t) represents the pre-stack gather data, seis(h, t) represents the gather data, q(h, t) represents the correction coefficient, and seis_c(x, y, h, t) represents the corrected gather data.

[0101] Fig. 8 is a comparison diagram before and after correction in the embodiment.

[0102] S4: Obtain angle gather data

[0103] In the preprocessing of seismic data, after correction, the corrected gather data in the obtained seismic record is still in the offset distance domain, but since the AVO analysis is based on the incident angle as the variable, in order to facilitate the AVO analysis, it is necessary to convert the fixed offset record into the fixed incident angle or the superimposed gather record within a certain angle range. In the angle processing, the corresponding part of the fixed offset record within a certain reflection angle range is usually combined to obtain an angle of reflection, and by repeating this process and changing different angles or angle ranges, different angle gathers can be obtained to form an angle gather, so that the angle gather of different angles can be obtained. Figure 9 Fig. 8 is a comparison diagram before and after correction in the embodiment.

[0104] However, in AVO analysis, we often do not need all the angle gather data, and in an angle gather data, we also do not need all the angle range, so we can stack a certain range of angle gather data according to the need to form the angle partial stack gather data, which is the most basic angle stack gather data in AVO analysis, however, in the angle partial stack gather data, each angle gather partial stack gather data has a central angle, in actual processing, in order to improve the signal-to-noise ratio of AVO gather record and improve certain lateral resolution, we can choose the record covered by three times or more than three times in each angle gather to do stack.

[0105] For the angle range, not all the angle range of the gather can be stacked, we know that the actual seismic data acquisition is through multiple coverage. The purpose is to effectively improve the signal-to-noise ratio and improve the lateral resolution, so we can determine the range of angle according to the definition of the width of the first Fresnel zone.

[0106] S5: Super-gather processing

[0107] The signal-to-noise ratio of the angle gather data is analyzed, if the signal-to-noise ratio is not enough, further super-gather processing can be carried out to improve the signal-to-noise ratio, the so-called super-gather processing is the bin stack in a certain range, which improves the lateral continuity of the gather, Figure 10 The effect diagram of the super-gather processing of the angle gather data in the embodiment is shown.

[0108] S6: Obtain flattened corrected gather data

[0109] The existing residual moveout correction technology is mostly realized by anisotropic velocity analysis, but because of the difficulty in obtaining anisotropic parameters, the flattening problem has not been well solved, in the technical solution, the angle gather data is preprocessed by using the gather flattening technology based on automatic tracking of the same phase axis to carry out flattening processing, the basic principle of the gather flattening technology is to extract significant horizon points at the similar waveform of the pre-stack gather, to carry out automatic tracking of the horizon, to obtain the gather time shift amount to carry out flattening correction, to realize the residual moveout correction without velocity, Figure 11 The effect diagram of the residual moveout correction of the angle gather in the embodiment is shown.

[0110] Embodiment two:

[0111] The embodiment is illustrated in detail by taking the clastic rock block in Sichuan Basin of China.

[0112] Figure 1 The flowchart of the optimization processing method of the along-river pre-stack gather data of the fluvial facies reservoir is shown, which specifically includes the following steps:

[0113] S1: Obtain along-river pre-stack gather data

[0114] River channel prestack gather extraction technical process, specifically as follows: from the river channel sensitive attribute optimization, then plane attribute extraction, through well site calibration and threshold determination, the spatial distribution pattern of river channel is described, and then it is applied to prestack gather data to obtain river channel prestack gather data which can be used for subsequent processing and interpretation. The river channel gather extraction data technical process is as shown in Figure 2 ; Figures 3a-3b The front and rear plane attribute comparison chart before and after the river channel gather extraction data; Figures 4a-4b The river channel prestack gather data before and after in the vertical river channel direction can be compared. It can be seen that the processed gather removes the redundant information irrelevant to river channel interpretation, which lays a foundation for the efficiency and accuracy improvement of subsequent processing and interpretation.

[0115] The extracted river facies reservoir along river channel prestack gather data has the following advantages in subsequent data processing and interpretation: ①It can improve the efficiency of river facies reservoir prediction and fluid detection in gather processing, prestack inversion, attribute extraction, etc. Figure 5 The efficiency comparison of several key processing and interpretation modules before and after gather extraction in this embodiment is listed. It can be seen that whether it is prestack gather targeted processing or seismic inversion and attribute extraction, the single operation efficiency is improved by more than 90%; ②It improves the accuracy of river facies reservoir prediction and fluid detection, which is beneficial to the interpretation of river channel top and bottom interfaces and the elimination of background wall rock influence; due to the influence of sedimentation, the lateral continuity of river facies reservoir is short, and the up-down time difference fluctuates. The existence of these factors is not conducive to the interpretation of river channel top and bottom interfaces. Using the extracted river channel prestack gather data, the influence of surrounding redundant seismic traces is removed, and the river channel top and bottom interfaces can be more accurately identified. Figure 6 The comparison of well tie profile using all gather and river channel prestack gather data shows that the river channel top and bottom interfaces using river channel prestack gather data are more accurate.

[0116] S2: denoising processing

[0117] The time migration of the river channel prestack gather data is usually aimed at target poststack seismic data, and various noises, especially Gaussian noise, usually exist in the prestack data before stacking. Before angle conversion and partial stacking, the purpose of denoising is to improve the signal-to-noise ratio of the gather data and the signal-to-noise ratio of the final angle stacking data. In this embodiment, an improved median filtering method is used to remove random noise. The principle of this method is to pick up the sampling points of the same time of adjacent seismic traces in the river channel prestack gather data, and by cutting these abnormal points, the noise influence can be reduced. The specific steps are as follows: first, the seismic data of a moment in the river channel prestack gather data is sorted, a part of the maximum value and the minimum value is removed, and then the arithmetic mean of the remaining sample points is calculated as the sampling point value of the current center trace at the moment, Figure 7 The contrast diagram before and after denoising of the river channel prestack gather data in this embodiment is shown.

[0118] The median filtering calculation formula is:

[0119]

[0120] Wherein, S represents seismic data, x, y, t are three dimensions of seismic data, which represent X direction, Y direction and time direction t respectively. Mn is the size of the median filtering window function, m is the window size in X direction, and n is the window size in Y direction.

[0121] S3: Obtain corrected gather data

[0122] The denoised river channel prestack gather data usually has an amplitude relationship of "strong-weak-strong". In theory, except for phase conversion, none of the four types of AVO has an amplitude relationship of "strong-weak-strong" with the increase of offset distance. The non-true AVO relationship is very unfavorable for subsequent AVO analysis and AVO inversion. The reason for this phenomenon is often related to the acquisition system. The amplitude anomaly caused by the acquisition system is the first concern. In this technical solution, the three-dimensional observation system amplitude anomaly that has not been completely eliminated in the migration process is processed on the common offset gather data after migration.

[0123] The specific method is to stack all the denoised river channel prestack gather data according to different offsets to obtain a gather data, which only contains time t and offset h information, recorded as seis(h, t).

[0124] The least square fitting is performed on each prestack gather data seis_o(x, y, h, t) and the gather data seis(h, t) to obtain a correction coefficient q(h, t).

[0125] The expression of the correction coefficient is:

[0126] f(q(h,t)) = min ||seis(h,t)*q(h,t) - seis_o(x,y,h,t)||2

[0127] Then the corrected gather is:

[0128] seis_c(x,y,h,t) = seis_o(x,y,h,t)*q(h,t)

[0129] Where the general representation of pre-stack gather data contains 4 dimensions, X direction, Y direction, time t and offset h, seis_o(x,y,h,t) represents pre-stack gather data, seis(h,t) represents gather data, q(h,t) represents correction coefficient, and seis_c(x,y,h,t) represents corrected gather data.

[0130] Figures 8a-8b This is a schematic diagram of the comparison before and after correction in this embodiment.

[0131] S4: Obtain angle gather data

[0132] In the preprocessing of seismic data, after correction, the corrected gather data in the obtained seismic record is still in offset domain, but since AVO analysis is based on incident angle as a variable, in order to facilitate AVO analysis, it is necessary to convert the fixed offset record into fixed incident angle or a certain angle range of superimposed gather record. In angle processing, we usually combine the corresponding part of the fixed offset record within a certain reflection angle range to obtain an angle of reflection, and by repeating this process and changing different angles or angle ranges, we can obtain different angle gathers, forming an angle gather, so that we can obtain angle gathers of different angles, Figure 9 This is the offset conversion of the denoised along-channel pre-stack gather data into angle gather data in this embodiment.

[0133] However, in AVO analysis, we often do not need all the angle gather data, and in an angle gather data, we also do not need all the angle range, so we can stack a certain range of angle gather data according to the need to form angle partial stack gather data, which is the most basic angle stack gather data in AVO analysis. However, in angle partial stack gather data, each angle partial stack gather data has a central angle, and in actual processing, in order to improve the signal-to-noise ratio of AVO gather record and improve certain lateral resolution, we can select the record reaching three times or more than three times coverage in each angle gather data for stacking.

[0134] For the angle trace range, not all angle range of the trace data can be superimposed. We know that the actual seismic data acquisition is through multiple coverages, and the purpose is to effectively improve the signal-to-noise ratio and improve the lateral resolution, so we can determine the range of angle trace according to the definition of the width of the first Fresnel zone. Figure 12 For the comparison between the middle angle superposition (15°) of the partial angle superposition trace data profile in the embodiment and the full superposition profile, the angle partial superposition data is the input data of subsequent prestack seismic inversion.

[0135] S5: super trace processing

[0136] The signal-to-noise ratio of the angle trace data is analyzed, and if the signal-to-noise ratio is not enough, further super trace processing can be performed to improve the signal-to-noise ratio, and the so-called super trace processing is bin stacking in a certain range, which improves the lateral continuity of the trace data, Figure 10 The effect diagram of the super trace processing of the angle trace data in the embodiment is shown.

[0137] S6: obtaining flattened corrected trace data

[0138] The existing residual moveout correction technology is mainly achieved by anisotropic velocity analysis, but due to the difficulty in obtaining anisotropic parameters, the flattening problem has not been well solved. In the technical process, the pre-processing is performed on the angle trace data by using the trace flattening technology based on automatic trace tracking to perform flattening processing. The basic principle is to extract significant horizon points at the similar waveforms of the trace, to perform automatic horizon tracking, to obtain the trace time shift amount for flattening correction, to realize the residual moveout correction without velocity, Figure 11 The effect diagram of the residual moveout correction of the angle trace data in the embodiment is shown.

[0139] Embodiment three

[0140] The embodiment of the present application provides a device for optimizing river facies reservoir along-river prestack trace data, which comprises:

[0141] A prestack trace data module is used to provide prestack trace data.

[0142] An along-river trace prestack trace data module is used to depict the spatial distribution form of the river channel and apply it to the prestack trace data to obtain along-river trace prestack trace data.

[0143] Specifically, starting from the optimization of river channel sensitive attributes, then extracting plane attributes, and through well site calibration and threshold determination, the spatial distribution form of the river channel is depicted, and then it is applied to the prestack trace data to obtain along-river trace prestack trace data which can be used for subsequent processing and interpretation.

[0144] Seismic data belongs to mass data, there are many algorithms for extracting seismic data, in order to obtain high extraction speed, we need to improve the speed of data retrieval, quickly locate the CDP point, including data volume scanning and data volume index two parts.

[0145] Data volume scanning:

[0146] Data volume scanning is to scan the data volume header and trace header data before loading the data into the database, scan the related attribute information of the data volume, and store it in the database in the form of structured data, which is convenient for data volume basic information query.

[0147] Data volume index:

[0148] In order to obtain high seismic data extraction speed, the index file can be established by scanning each trace of the data volume, the index file records the X, Y coordinates of each seismic trace, inline, crossline number and the absolute position of the trace data. In this way, if we get the X, Y coordinates of each segment of the profile to be extracted, we can find the index file corresponding to the data volume of the work area through the structured data information of the database, and through the X, Y coordinates, we can find the absolute position of the trace data corresponding to the coordinates in the data volume, so we can directly locate the position of the trace to read the data, avoiding scanning the entire data volume. Since the index file is generally very small, it can be scanned quickly, so the data extraction speed can be doubled through the data volume index.

[0149] The obtained along-river prestack trace data has the following advantages in subsequent data processing and interpretation:

[0150] First, it can improve the efficiency of river facies reservoir prediction and fluid detection in trace set processing, prestack inversion and attribute extraction; second, it can improve the accuracy of river facies reservoir prediction and fluid detection, which is beneficial to the interpretation of river channel top and bottom interfaces and the elimination of background wall rock influence.

[0151] The denoising module is used for denoising the along-river prestack trace data to obtain denoised along-river prestack trace data.

[0152] Specifically, the conventional time migration of the riverway prestack gather data is usually aimed at target poststack seismic data, and the riverway prestack gather data usually has various noises, especially Gaussian noise, before stacking. Before angle conversion and partial stacking, the denoising processing is performed to improve the signal-to-noise ratio of the gather data and the signal-to-noise ratio of the final angle stacking data, and remove random noise. There are many methods for removing random noise, such as median filtering, polynomial fitting, f-x domain prediction denoising, time-space variable wavelet threshold fidelity denoising based on horizontal sliding, CRP gather rolling stacking to form super-gather denoising, etc. In the embodiment, an improved median filtering method is used to remove random noise. The principle of this method is to pick up the sampling points of the same time of adjacent seismic traces in the riverway prestack gather data, and the abnormal points are cut off to reduce the influence of noise. The specific steps are as follows: first, the seismic data of a time in the riverway prestack gather data is sorted, and a part of the maximum value and the minimum value is removed, and then the arithmetic mean of the remaining sample points is calculated as the sampling point value of the current center trace at the time.

[0153]

[0154] Wherein, S represents seismic data, x, y, t are three dimensions of seismic data, representing X direction, Y direction and time direction t respectively. Mn is the size of the median filtering window function, m is the window size in X direction, and n is the window size in Y direction.

[0155] The correction module corrects the amplitude of the denoised riverway prestack gather data to obtain corrected gather data.

[0156] Specifically, all the denoised riverway prestack gather data is stacked into a gather data according to different offset distances. The gather data only contains time t and offset distance h information, and is recorded as seis(h,t).

[0157] The least square fitting is performed on each prestack gather data seis_o(x,y,h,t) and the gather data seis(h,t) to obtain a correction coefficient q(h,t).

[0158] The expression of the correction coefficient is:

[0159] f(q(h,t))=min||seis(h,t)*q(h,t)-seis_o(x,y,h,t)||2

[0160] Then the corrected gather is:

[0161] seis_c(x,y,h,t)=seis_o(x,y,h,t)*q(h,t)

[0162] wherein t represents time, h represents offset distance, seis_o(x, y, h, t) represents pre-stack gather data, seis(h, t) represents gather data, q(h, t) represents correction coefficient, and seis_c(x, y, h, t) represents corrected gather data.

[0163] an angle gather data module, which acquires angle gather data by changing the angle of the corrected gather data;

[0164] Specifically, in the preprocessing of seismic data, after correction, the corrected gather data in the obtained seismic record is still in the offset distance domain, but since AVO analysis is based on the incidence angle as a variable, in order to facilitate AVO analysis, it is necessary to convert the fixed offset record into fixed incidence angle or stacked gather record within a certain angle range, and in the angle processing, the corresponding part of the fixed offset record within a certain reflection angle range is usually combined to obtain an angle of reflection, and by repeating this process and changing different angles or angle ranges, different angle gathers can be obtained to form an angle gather, so that the angle gather of different angles can be obtained.

[0165] a preprocessing module, which pre-processes the angle gather data to obtain flattened corrected gather data.

[0166] Specifically, the preprocessing module performs flattening processing on the angle gather data by using the gather flattening technique based on automatic trace of the same phase axis, the basic principle of the gather flattening technique is to extract significant horizon points at the similar waveforms of the gather, to automatically track the horizon, to obtain gather time shift amount for flattening correction, and to realize residual moveout processing without speed.

[0167] The above has described the embodiments of the present application, the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A fluvial facies reservoir along-river prestack gather data optimization processing method, characterized in that, The method comprises the following steps: providing pre-stack gather data; characterizing the spatial distribution of river channels and applying the spatial distribution of river channels to the pre-stack gather data to obtain river channel pre-stack gather data; performing denoising processing on the river channel pre-stack gather data to obtain denoised river channel pre-stack gather data; correcting the amplitude of the denoised river channel pre-stack gather data to obtain corrected gather data; obtaining angle gather data by changing the angle of the corrected gather data; performing preprocessing on the angle gather data to obtain flattened corrected gather data; wherein the step of correcting the amplitude of the denoised river channel pre-stack gather data to obtain corrected gather data comprises: stacking all the denoised river channel pre-stack gather data according to different offsets to obtain gather data, and then performing least square fitting on each pre-stack gather data and the gather data to obtain corrected gather data; before the step of characterizing the spatial distribution of river channels, the method comprises the following steps: selecting a river channel sensitive attribute, performing plane attribute extraction, and determining the well location and threshold.

2. The method according to claim 1, wherein, before the step of obtaining river channel pre-stack gather data, the method comprises the following steps: extracting a seismic data volume, wherein the extraction of the seismic data volume comprises data volume scanning and data volume indexing.

3. The method according to claim 1, wherein, before the step of obtaining the corrected gather data, the method comprises the following steps: performing least square fitting on each pre-stack gather data and the gather data to obtain a correction coefficient.

4. The method according to claim 3, wherein, The expression of the correction coefficient is: wherein t represents time, h represents offset, seis_o(x, y, h, t) represents denoised river channel pre-stack gather data, seis(h, t) represents gather data, and q(h, t) represents a correction coefficient.

5. The method according to claim 1, wherein, The expression of the corrected gather data seis_c(x, y, h, t) is: wherein seis_o(x, y, h, t) represents denoised river channel pre-stack gather data, q(h, t) represents a correction coefficient, and seis_c(x, y, h, t) represents corrected gather data.

6. The method according to claim 1, wherein, In the step of performing preprocessing on the angle gather data, the preprocessing refers to flattened processing by using gather flattening technology.

7. The method according to claim 1, wherein, after the step of obtaining angle gather data, the method further comprises the following steps: performing super-gather processing on the angle gather data to improve the lateral continuity of the angle gather data.

8. A device for optimizing fluvial facies reservoir along-river prestack gather data, characterized in that, The method comprises the following steps: a pre-stack gather data module is configured to provide pre-stack gather data; a river channel pre-stack gather data module is configured to characterize the spatial distribution of river channels and apply the spatial distribution of river channels to the pre-stack gather data to obtain river channel pre-stack gather data; a denoising module is configured to perform denoising processing on the river channel pre-stack gather data to obtain denoised river channel pre-stack gather data; a correction module is configured to correct the amplitude of the denoised river channel pre-stack gather data to obtain corrected gather data; an angle gather data module is configured to obtain angle gather data by changing the angle of the corrected gather data; a preprocessing module is configured to perform preprocessing on the angle gather data to obtain flattened corrected gather data; wherein the step of correcting the amplitude of the denoised river channel pre-stack gather data to obtain corrected gather data comprises: The de-noised riverway pre-stack gather data is stacked into gather data according to different offset distances, and then each of the pre-stack gather data is least square fitted with the gather data to obtain corrected gather data; Before the spatial distribution pattern of the riverway is depicted, the method comprises the following steps: The riverway pre-stack gather data module is further used for selecting a riverway sensitive attribute, performing plane attribute extraction, and determining a well site calibration and a threshold.

Images