A method for multiple wave prediction based on adaptive weighting

Abstract

The application provides a multiple wave prediction method based on adaptive weighting, and the prior art method only simply stacks multiple wave contribution gathers when predicting multiple waves, and the development of multiple waves at different offsets is obviously different, and generally, the development of multiple waves is strong at near offsets, and the development of multiple waves is weak at far offsets, so the application provides a three-dimensional wave equation multiple wave prediction technology of adaptive weighting stacking for near-mid-far different offsets, the technology can improve the accuracy of multiple wave prediction, can effectively improve the multiple wave prediction precision, and further provides a high-quality multiple wave model for a subsequent multiple wave subtraction stage.

Description

Technical Field

[0001] This invention belongs to the field of petroleum exploration, and specifically relates to multiple wave prediction and suppression in the field of seismic data processing. More specifically, it relates to a multiple wave prediction method based on adaptive weighting. Background Technology

[0002] Multiples are not only a crucial issue in marine seismic processing but also a prominent problem in onshore seismic exploration. Their presence severely degrades data quality, affects the accuracy and reliability of seismic imaging, and interferes with subsequent seismic data interpretation. In my country's eastern oilfields, multiples are a common problem in seismic exploration. Their presence not only affects structural exploration but also distorts the amplitude, frequency, and phase of effective reflections, impacting lithological exploration. Therefore, resolving multiples is of significant practical importance for onshore exploration. With in-depth research on multiples, it has become increasingly clear that free-surface multiples constitute the dominant energy source in seismic data.

[0003] Anstey and Newman pioneered research on free-surface correlated multiples, finding that multiples could be well described by convolving the seismic trace with itself. Riley and Claerbout proposed a forward modeling algorithm for free-surface correlated multiples in two-dimensional medium space, but they failed to propose a corresponding inversion method to eliminate multiples from the original data. Berkhout et al. proposed a theoretical method for eliminating surface-correlated multiples in multidimensional space, using the original data itself as a multiple operator to simulate free-surface correlated multiples. This method does not require prior knowledge of detailed subsurface medium information and can adapt to complex subsurface medium conditions. In the 1990s, Verschuur et al. proposed the Free-Surface Correlated Multiples Adaptive Suppression Method (SRME) and successfully applied it to the processing of real data. This method uses the minimum energy criterion to adaptively eliminate predicted multiples from the input data. Weglein et al. used the backscattering series method to eliminate free-surface multiples and interlayer multiples. The suppression multiple wave methods proposed by Verschuur and Weglein both fall under the category of elastic wave theory methods. These methods require almost no prior information, have good application effects, and have become the mainstream of suppression multiple wave methods and technological development. Summary of the Invention

[0004] This application provides an adaptive weighted multiple prediction method, which is a three-dimensional wave equation multiple prediction technology based on adaptive weighted superposition for different offset distances from near to middle to far. This technology can improve the accuracy of multiple prediction, thereby providing a high-quality multiple model for the subsequent multiple subtraction stage.

[0005] This application provides an adaptive weighted multiple prediction method, including:

[0006] Acquire 3D seismic data containing multiples;

[0007] Based on the differences in the development of multiples at different offsets, the contribution gathers of three-dimensional multiples are obtained.

[0008] Among them, based on the differences in the development of multiples at different offsets, the three-dimensional multiple contribution gather is obtained, including:

[0009] Given the detector point (x) r y r The common receiver point set P(x) related to r ,y r ,x k ,y k ,t) and with the source (x s ,y s The common shot point set P(x) related to ) k ,y k ,x s ,y s ,t), using their convolution to predict multiple channels m(x r ,y r ,x s ,y s ,t);

[0010] In the spatial frequency domain, the 3D surface multiple wave prediction formula based on the wave equation can be expressed as:

[0011]

[0012] In the formula, M0 is the predicted multiple wave, and P0(x) r ,y r ,x k ,y k ,ω) denotes the Fourier transform of the original data common receiving point gather, P(x k ,y k ,x s ,y s (x, ω) denotes the Fourier transform of the common shot point gather, (x k y k () indicates the downlink reflection point, and MCG indicates that the epicenter is located at (x) s ,y s ) and the detector point is located at (x r y r The multiple wave contribution gathers for all downlink reflection points (x) k y k Perform calculations;

[0013] The predicted multiple waves M0 are weighted and superimposed to obtain M. weight The specific algorithm is shown in the following formula:

[0014]

[0015] In the formula, W is the weighting factor.

[0016] Among them, the weighting factor W is related to the relative distance, which is the relative distance between the downlink reflection point and the shot-receiver point of the multiple wave to be predicted.

[0017] The weighting factor W is calculated as follows:

[0018]

[0019] D SR D represents the distance between the shot and receiver of the multiple waves to be predicted. SK D represents the distance from the shot point to the downlink reflection point. RK This represents the distance from the detector point to the downlink reflection point, where i is a natural number greater than 1.

[0020] The value of i can be adjusted according to the strength of the multiple wave development.

[0021] Acquiring three-dimensional seismic data containing multiples includes: acquiring three-dimensional seismic data containing multiples through field seismic acquisition or indoor forward numerical simulation.

[0022] Among them, multiple waves are predicted by directly utilizing information implicit in pre-stack seismic data, without the need for other information about the subsurface medium.

[0023] Among them, surface multiples of all orders were predicted.

[0024] This application also provides a method for multiple wave suppression, including any of the above-mentioned multiple wave prediction methods based on adaptive weighting.

[0025] This includes predicting the multiple wave model record using any of the above-mentioned adaptive weighted multiple wave prediction methods, and then subtracting the multiple waves using an adaptive subtraction method to achieve multiple wave suppression.

[0026] The adaptive weighted multiple prediction method in this application has the following advantages:

[0027] This method proposes a three-dimensional wave equation multiple prediction technique for different offset distances (near, middle, and far). It can improve the accuracy of multiple prediction, especially the accuracy of multiple prediction at near offset distances. Compared with the conventional simple superposition method, it has better prediction effect and higher accuracy for multiples. Attached Figure Description

[0028] Figure 1 This is a schematic diagram of the adaptive weighted multiple prediction method according to an embodiment of this application;

[0029] Figure 2 The weight plane distribution of MCG gathers for simple weighted prediction of multiple waves is shown.

[0030] Figure 3 The weight plane distribution of the MCG gathers for adaptive weighted multiple prediction in this application is shown.

[0031] Figure 4(a) is a schematic diagram of conventional simple weight prediction;

[0032] Figure 4(b) is a schematic diagram of the adaptive weighted multiple prediction of this application;

[0033] Figure 5(a) is a schematic diagram of multiple suppression before it is performed based on adaptive weighted multiple prediction data;

[0034] Figure 5(b) is a schematic diagram of multiple suppression based on adaptive weighted multiple prediction data. Detailed Implementation

[0035] The present application will be further described below with reference to the accompanying drawings and embodiments.

[0036] The following description provides several embodiments of the present invention. Different embodiments can be substituted or combined. Therefore, this application can also be considered to include all possible combinations of the same and / or different embodiments described. Thus, if one embodiment includes features A, B, and C, and another embodiment includes features B and D, then this application should also be considered to include embodiments containing one or more other possible combinations of features A, B, C, and D, even if such embodiments are not explicitly described in the following text.

[0037] Verschuur and Weglein's method for suppressing multiples involves predicting multiple model records from seismic data and then subtracting the multiples using an adaptive subtraction method. However, this method simply overlays the multiple contribution gathers during prediction, failing to fully consider the significant differences in multiple development at different offsets. Typically, multiple development is stronger at closer offsets and weaker at farther offsets. Therefore, this invention proposes a three-dimensional wave equation multiple prediction technique based on adaptive weighted stacking for near-, mid-, and far offsets. This technique improves the accuracy of multiple prediction, thereby providing a high-quality multiple model for the subsequent multiple subtraction stage.

[0038] like Figure 1As shown, the adaptive weighted multiple prediction method of this application includes: S101, acquiring three-dimensional seismic data containing multiples; S103, calculating the three-dimensional multiple contribution gathers based on the differences in the development of multiples at different offsets. These will be described in detail below.

[0039] Existing methods do not fully consider the development of multiples at different offsets, resulting in weaker predicted multiples at near offsets, which does not match the actual situation. This scheme fully considers the differences in the contribution of near, medium and far offsets to multiples, and adopts a weighted superposition method, which can better match the actual multiples.

[0040] The purpose of this invention is to better predict multiples and provide a high-quality multiple model for subsequent matching and subtraction. Multiples are typically stronger at closer offsets and weaker at farther offsets; therefore, an adaptive weighted superposition method can effectively improve the prediction accuracy of multiples.

[0041] like Figure 1 As shown in Figure 5, the wave theory-based multiple prediction technique is a data-driven method for predicting multiples. It directly utilizes information implicit in pre-stack seismic data to predict multiples without requiring any other information about the subsurface medium, and can simultaneously predict surface multiples of all orders. The specific steps are as follows:

[0042] Step 1: Obtain three-dimensional seismic data containing multiples through field seismic acquisition or indoor forward numerical simulation.

[0043] Step 2: The usual method for obtaining the contribution gather of three-dimensional multiple waves.

[0044] Given the detector point (x) r y r The common receiver point set P(x) related to r ,y r ,x k ,y k ,t) and with the source (x s ,y s The common shot point set P(x) related to ) k ,y k ,x s ,y s ,t), using their convolution to predict multiple channels m(x r ,y r ,x s ,y s ,t).

[0045] In the spatial frequency domain, the 3D surface multiple wave prediction formula based on the wave equation can be expressed as:

[0046]

[0047] In the formula, M0 is the predicted multiple wave, and P0(x) r ,y r ,x k ,y k ,ω) denotes the Fourier transform of the original data common receiving point gather, P(x k ,y k ,x s ,y s (x, ω) denotes the Fourier transform of the common shot point gather, (x k y k () indicates the downlink reflection point, and MCG indicates that the epicenter is located at (x) s ,y s ) and the detector point is located at (x r y r The multiple wave contribution gathers for all downlink reflection points (x) k y k ) to perform calculations.

[0048] Step 3: This invention improves upon conventional methods for obtaining three-dimensional multiple wave contribution gathers.

[0049] Existing methods for predicting multiples simply superimpose the contributing gathers of the multiples, failing to fully consider the significant differences in multiple development at different offsets. Typically, multiple development is stronger at near offsets and weaker at far offsets. Therefore, this invention proposes a three-dimensional wave equation multiple prediction technique based on adaptive weighted superposition for near-, mid-, and far offsets. This technique improves the accuracy of multiple prediction, thereby providing a high-quality multiple model for the subsequent multiple subtraction stage. Building upon the second step, this application performs weighted superposition on M0 from the second step to obtain M... weight The specific algorithm is shown in the following formula:

[0050]

[0051] The above formula is an adaptive weighted superposition for MCG gathers, where W is the weighting factor, and the magnitude of this weight is related to the relative distance between the downlink reflection point and the shot-receiver point of the multiple wave to be predicted, and D... SR D represents the distance between the shot and receiver of the multiple waves to be predicted. SK D represents the distance from the shot point to the downlink reflection point. RK This represents the distance from the detector point to the downlink reflection point. The weighting factor W is calculated as follows:

[0052]

[0053] In the formula, i is a natural number greater than 1, which can be adjusted appropriately according to the strength of the multiple wave development.

[0054] The adaptive weighted superposition technique for three-dimensional wave equation multiple prediction proposed in this application for different offset distances (near, middle, and far) can improve the accuracy of multiple prediction, especially the accuracy of multiple prediction at near offset distances. Compared with conventional simple superposition methods, it has better prediction effect and higher accuracy for multiples.

[0055] like Figure 1 As shown in -5, Figure 2 The figure shows the weight plane distribution of the MCG gather for predicting multiple waves using simple weights. As can be seen from the figure, simple weights mean that the weight of each trace in the MCG gather is 1. Figure 3 The figure shows the weight plane distribution of the MCG gather for adaptive weighted multiple prediction. It can be seen that the adaptive weighting is related to the distance between the downlink reflection point and points S and D; the greater the distance, the smaller the weight. The weight between the lines connecting S and D is 1, and all other weights are less than 1. Figure 4(a) shows the effect of conventional simple weight prediction, and Figure 4(b) shows the effect of adaptive weighted multiple prediction with single weight prediction. The comparison shows that the adaptive weight prediction technique is more effective for near-offset multiple prediction and better reflects the actual situation where multiples are strong at near offsets. Figure 5(a) is a schematic diagram before multiple suppression based on adaptive weighted multiple prediction data, and Figure 5(b) shows the effect after multiple suppression based on adaptive weighted multiple prediction data. The figures show that multiples are well suppressed, while the effective signal is well preserved.

[0056] This application also provides a method for suppressing multiple waves, including any of the above-mentioned adaptive weighted multiple wave prediction methods, such as predicting multiple wave model records using any of the above-mentioned adaptive weighted multiple wave prediction methods, and subtracting multiple waves using an adaptive subtraction method to achieve multiple wave suppression.

[0057] This application also provides an adaptive weighted multiple prediction device, comprising: an acquisition unit for acquiring three-dimensional seismic data containing multiples; and a calculation unit for calculating the three-dimensional multiple contribution gathers based on the differences in the development of multiples at different offsets.

[0058] In some embodiments, the calculation unit is used to: know the coordinates of the detector point (x) r y r The common receiver point set P(x) related to r ,y r ,x k ,y k ,t) and with the source (x s ,y s The common shot point set P(x) related to )k ,y k ,x s ,y s ,t), using their convolution to predict multiple channels m(x r ,y r ,x s ,y s ,t);

[0059] In the spatial frequency domain, the 3D surface multiple wave prediction formula based on the wave equation can be expressed as:

[0060]

[0061] In the formula, M0 is the predicted multiple wave, and P0(x) r ,y r ,x k ,y k ,ω) denotes the Fourier transform of the original data common receiving point gather, P(x k ,y k ,x s ,y s (x, ω) denotes the Fourier transform of the common shot point gather, (x k y k () indicates the downlink reflection point, and MCG indicates that the epicenter is located at (x) s ,y s ) and the detector point is located at (x r y r The multiple wave contribution gathers for all downlink reflection points (x) k y k Perform calculations;

[0062] The predicted multiple waves M0 are weighted and superimposed to obtain M. weight The specific algorithm is shown in the following formula:

[0063]

[0064] In the formula, W is the weighting factor, which is related to the relative distance, specifically the distance between the downlink reflection point and the shot-receiver point of the multiple wave to be predicted. The weighting factor W is calculated as follows:

[0065]

[0066] D SR D represents the distance between the shot and receiver of the multiple waves to be predicted. SK D represents the distance from the shot point to the downlink reflection point. RK This represents the distance from the receiver point to the downlink reflection point. i is a natural number greater than 1, and the value of i can be adjusted according to the strength of the multiple wave development.

[0067] The embodiments of the adaptive weighted multiple prediction device and the adaptive weighted multiple prediction method in this application are basically the same. For details, please refer to the introduction of the adaptive weighted multiple prediction method.

[0068] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the adaptive weighted multiple wave prediction method described above. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0069] This application also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of any of the above-described multiple wave prediction methods.

[0070] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for predicting multiple waves based on adaptive weighting, characterized in that, include: Acquire 3D seismic data containing multiples; Based on the differences in the development of multiples at different offsets, the three-dimensional multiple contribution gathers are obtained. Among them, based on the differences in the development of multiples at different offsets, the three-dimensional multiple contribution gather is obtained, including: Known and receiver point Related common receiving point collection and the epicenter Related artillery points Using their convolution to predict multiple channels ; In the spatial frequency domain, the 3D surface multiple wave prediction formula based on the wave equation can be expressed as: In the formula It is a predicted multiple wave. This represents the Fourier transform of the original data common-receiver gather. This represents the Fourier transform of the common gun point gather. Indicates the downward reflection point. Indicates that the epicenter is located at and the detector point is located Multiple wave contribution gathers for all downlink reflection points Perform calculations; For predicted multiple waves By performing weighted summation, we obtain The specific algorithm is shown in the following formula: In the formula It is a weighting factor; Among them, weighting factor It relates to the relative distance, which is the relative distance between the downlink reflection point and the shot-receiver point of the multiple wave to be predicted; Among them, weighting factor The calculation is as follows: This represents the distance between the shot and receiver points for the multiple waves to be predicted. This indicates the distance from the shot point to the downlink reflection point. This indicates the distance from the detector point to the downlink reflection point. It is a natural number greater than 1.

2. The adaptive weighted multiple prediction method according to claim 1, characterized in that, The value can be adjusted according to the strength of the multiple wave development.

3. The adaptive weighted multiple prediction method according to claim 1, characterized in that, Acquiring 3D seismic data containing multiples includes: acquiring 3D seismic data containing multiples through field seismic acquisition or indoor forward numerical simulation.

4. The adaptive weighted multiple wave prediction method according to claim 1, characterized in that, Multiple waves can be predicted directly using information implicit in prestack seismic data, without the need for other information about the subsurface medium.

5. The adaptive weighted multiple prediction method according to claim 1, characterized in that, Predict all orders of surface multiple waves.

6. A method for multiple wave suppression, characterized in that, Includes any one of the adaptive weighted multiple wave prediction methods in claims 1-5.

7. The method for suppressing multiple waves according to claim 6, characterized in that, This includes predicting the multiple wave model record using any one of the adaptive weighted multiple wave prediction methods in claims 1-5, and subtracting the multiple waves using an adaptive subtraction method to achieve multiple wave suppression.

Images