A spectral inversion frequency expansion processing method based on stratum structure dimension reduction

Abstract

The application discloses a spectral inversion frequency expansion processing method based on stratum structure dimension reduction, obtains post-stack seismic data after preprocessing such as denoising and zero-phase; designs different simple thin-layer stratum structure forward models, establishes the relationship between the seismic response characteristics of different models and the corresponding waveform phase information; obtains odd and even component data reflecting different simple stratum structure thin-layer information and obtains seismic data after phase decomposition; based on the phase decomposition method, the dimension of the complex stratum structure is reduced by decomposing the seismic data, and the optimal solution is searched in the data space after dimension reduction, and accurate high-resolution seismic data is obtained. Compared with the prior art, the data space size of spectral inversion optimization is reduced, the reliability and stability of the spectral inversion frequency expansion result are improved, and the fine seismic reservoir characterization ability of the spectral inversion result, such as thin-layer identification, is enhanced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of seismic data processing and interpretation of oil, and particularly relates to a spectral inversion frequency expansion processing method. BACKGROUND

[0002] With the gradual improvement of oil and gas exploration and development, the demand for fine exploration of lithologic oil and gas reservoirs is increasing. In recent years, fluvial lithologic oil and gas reservoirs are gradually becoming an important part of reserve replacement and production capacity growth. Exploration practice shows that the channel type composite sand body has great potential and is an important target for lithologic oil and gas reservoir exploration. However, the sand body mainly presents thin layer and interbedded structure, and the superimposed and connected relationship of the sand body is complex, which leads to great difficulty in fine description of the sand body, affects the positioning of the oil and gas reservoir and the design of well location, and seriously restricts the exploration and development of lithologic oil and gas reservoirs. Therefore, how to improve the resolution of seismic data has been an important research direction for fine seismic reservoir characterization such as thin layer identification in the exploration and development of complex lithologic oil and gas reservoirs.

[0003] The seismic spectral inversion method is an effective technical means for identifying thin interbeds and improving the resolution of seismic data. Without prior geologic model, layer and well constraints, based on the reflection coefficient odd-even decomposition theory and partial spectral information in seismic, the target function is constructed in the frequency domain, and the reflection coefficient volume is obtained through continuous iteration and inversion, which can remove the influence of interference and tuning effects of seismic waves in the underground propagation process on the underground formation, broaden the frequency band range of the effective seismic signal, especially more accurately broaden the high frequency component, and obtain the reflection coefficient volume with much higher resolution than the original seismic data.

[0004] In the implementation of the traditional seismic spectral inversion method, when the underground formation structure is simple and the signal-to-noise ratio of the seismic data is high, the odd-even decomposition of the reflection coefficient has a unique solution when solving the target function, and the reflection coefficient can be accurately inverted. However, in the complex formation structure of the composite sand body, the quality of the seismic data is usually poor and the noise is obvious, and the conventional seismic spectral inversion is unstable, the odd-even decomposition of the reflection coefficient has multiple solutions when solving the target function, which leads to the reduction of the accuracy of the spectral inversion result and the inability to obtain accurate high-resolution seismic data. Therefore, it is particularly important to develop a spectral inversion method that can accurately improve the resolution of seismic data in complex formation structures. SUMMARY

[0005] The main purpose of the present application is to overcome the above-mentioned shortcomings in the prior art, and to provide a spectral inversion frequency expansion processing method based on dimension reduction of formation structure, which realizes the dimension reduction of the complex formation structure by decomposing the original seismic data, thereby reducing the multiple solutions, and then performing spectral inversion frequency expansion and data fusion reconstruction processing on the odd-even component data reflecting the thin layer information of different formation structures, and finally obtaining accurate high-resolution seismic data.

[0006] The object of the present application is achieved by the following technical solutions.

[0007] A spectral inversion frequency expansion processing method based on stratum structure dimension reduction, comprising the following steps:

[0008] Firstly, obtain post-stack seismic data S(t) after denoising, zero-phase processing and other pretreatments;

[0009] Secondly, comprehensively apply logging, seismic and geological information to establish different simple thin layer stratum structure forward models, form a complex thin interbedded structure geological model by different simple thin layer stratum structure models, analyze seismic response characteristics of different simple thin layer stratum structure forward models, and establish a relationship between seismic response characteristics of different simple thin layer stratum structure models and corresponding waveform phase information;

[0010] Thirdly, perform phase decomposition on the seismic data S(t) to obtain a phase gather S'(theta, t) of the seismic data S(t) and seismic data S'(t) after phase decomposition;

[0011] Fourthly, obtain odd component data S o (t) and even component data S e (t) reflecting thin layer information of different simple stratum structures, and realize dimension reduction of the complex stratum structure; perform conventional spectral inversion processing on the odd component data S o (t) and the even component data S e (t) respectively, and the obtained inversion processing results include a seismic reflection coefficient volume r and r Se (t);

[0012] Fifthly, perform data fusion reconstruction processing on the inversion results obtained in the fourth step to obtain a final high-resolution seismic reflection coefficient volume r(t);

[0013] Sixthly, perform convolution operation on the seismic reflection coefficient volume r(t) obtained in the fifth step and a wide-frequency zero-phase wavelet w wb (t) to obtain wide-frequency amplitude-preserved seismic data S wb (t), which provides a data basis for realizing complex lithologic oil and gas reservoir thin layer identification and reservoir fine characterization.

[0014] The present application has the following advantages due to the above technical solutions:

[0015] 1) The present application analyzes seismic response characteristics and mechanism of the complex stratum structure model by forward simulation technology, establishes a relationship between seismic response characteristics of different stratum structure models and corresponding waveform phase information, and lays a theoretical foundation for solving the problems of multi-solution and low inversion result precision of the conventional spectral inversion method caused by the influence of the complex stratum structure by innovatively using the stratum structure dimension reduction idea.

[0016] 2), the original seismic data is decomposed by adopting the phase decomposition method, and odd component data (data obtained by summing -90° phase component and +90° phase component) and even component data (data obtained by summing 0° phase component and 180° phase component) reflecting different simple layer structure thin layer information are obtained, and the dimension reduction of the complex layer structure is realized in the form of decomposing the original seismic data;

[0017] 3), on the basis of the dimension reduction of the layer structure, the odd component data and the even component data reflecting different layer structure thin layer information are respectively subjected to spectral inversion frequency extension and data fusion reconstruction processing, and finally accurate high-resolution seismic data are obtained. Compared with the conventional spectral inversion method, the technical scheme provided by the present application seeks the optimal solution in the data space after dimension reduction, thereby reducing the data space size of inversion optimization and improving the reliability and stability of the spectral inversion frequency extension result, thereby providing a powerful guarantee for fine lithology exploration and efficient and accurate development of oilfields. BRIEF DESCRIPTION OF DRAWINGS

[0018] Figure 1 is a spectral inversion frequency extension processing method flowchart based on layer structure dimension reduction of the present application;

[0019] Figure 2 is a relationship diagram between the seismic response characteristics of different simple thin layer structure models and the corresponding waveform phase information, wherein (2a) is a low impedance thin layer model relationship diagram, (2b) is a high impedance thin layer model relationship diagram, (2c) is an impedance increasing thin layer model relationship diagram, and (2d) is an impedance decreasing thin layer model relationship diagram;

[0020] Figure 3 is a two-dimensional multi-layer model forward diagram of the present application, wherein (3a) is a two-dimensional multi-layer model, and (3b) is a seismic record obtained by forward simulation based on a convolution model;

[0021] Figure 4 is a diagram showing data reflecting different simple layer structure thin layer information obtained by adopting the phase decomposition method based on the forward seismic record of the present application, wherein (4a) is -90° phase component data, (4b) is +90° phase component data, (4c) is 0° phase component data, and (4b) is 180° phase component data;

[0022] Figure 5 is a comparison diagram of the frequency extension result obtained by the spectral inversion frequency extension processing method based on the layer structure dimension reduction of the present application and the frequency extension result obtained by the conventional spectral inversion method, wherein (5a) is the frequency extension result obtained by the conventional spectral inversion method, and (5b) is the frequency extension result obtained by the method of the present application;

[0023] Figure 6 is the frequency extension processing result diagram of the seismic profile of well A in the research area by using the method of the present application, wherein (6a) is the original well-to-well seismic profile, and (6b) is the frequency extension seismic profile obtained by using the method of the present application. DETAILED DESCRIPTION

[0024] The technical solutions of the present application will be further described in detail below with reference to the drawings.

[0025] As shown in Figure 1 , the overall flowchart of a spectral inversion frequency extension processing method based on stratum structure dimension reduction of the present application. The method comprises the following steps:

[0026] Firstly, obtain the post-stack seismic data S(t) after pre-processing such as de-noising and zero-phase;

[0027] Secondly, comprehensively apply well logging, seismic and geological information to establish different simple thin layer stratum structure forward models, and establish a complex thin interbedded structure geological model composed of different simple thin layer stratum structure models, analyze the seismic response characteristics of different simple thin layer stratum structure forward models through forward simulation technology, and establish the relationship between the seismic response characteristics of different simple thin layer stratum structure models and the corresponding waveform phase information;

[0028] In order to analyze the seismic response characteristics of different simple thin layer stratum structures, four typical simple thin layer geological models are established according to the sedimentary characteristics of fluvial facies reservoirs, which are low impedance thin layer model, high impedance thin layer model, impedance increasing thin layer model and impedance decreasing thin layer model. The high impedance thin layer model has opposite polarities of reflection coefficients, the low impedance thin layer model has opposite polarities of reflection coefficients, the impedance increasing thin layer model has the same polarity of two reflection coefficients and is positive, and the impedance decreasing thin layer model has the same polarity of two reflection coefficients and is negative. Therefore, the low impedance thin layer model and the high impedance thin layer model are antipolar thin layers, and the impedance increasing thin layer model and the impedance decreasing thin layer model are homopolar thin layers. Based on the convolution model, low-frequency zero-phase Ricker wavelet is selected for forward simulation. The four typical simple thin layer geological models and their corresponding forward seismic records are shown in Figure 2 , and it can be seen that the seismic response characteristics of different simple thin layer stratum structure models have different waveform phase information, i.e. the seismic response characteristics of the low impedance thin layer are represented by -90° phase waveform, the seismic response characteristics of the high impedance thin layer are represented by +90° phase waveform, the seismic response characteristics of the impedance increasing thin layer are represented by 0° phase waveform, and the seismic response characteristics of the impedance decreasing thin layer model are represented by 180° phase waveform, which lays a theoretical foundation for solving the problems of multi-solution and low precision of conventional spectral inversion method caused by the influence of complex stratum structure by using the subsequent innovative stratum structure dimension reduction idea.

[0029] Thirdly, the seismic data S(t) is phase decomposed to obtain a phase gather S'(θ, t) of the seismic data S(t) and the seismic data S'(t) after phase decomposition;

[0030] Firstly, a high-resolution and high-frequency stable phase spectrum is obtained by a seismic complex spectrum decomposition algorithm, then all frequency components are integrated to form a phase gather, and finally, stable decomposition and reconstruction of different phase components of the seismic data are realized.

[0031] Further, the expression of the phase gather S'(θ, t) of the seismic data S(t) after phase decomposition is as follows:

[0032]

[0033] Wherein, f1 and f2 represent the frequency range used for integration, S'(f, θ, t) represents the result obtained by using the seismic complex spectrum decomposition method to perform time-frequency analysis on the original seismic data S(t), f represents frequency, θ represents phase, t represents time, S'(θ, t) represents the phase gather obtained by phase decomposing the seismic data S(t), and represents the distribution of the seismic amplitude and phase with time. Further, the phase gather can be directly explained as the relationship between the amplitude and time of a single phase component in the seismic data. The seismic data containing only specified phase information can be obtained by stacking and reconstruction according to actual needs;

[0034] Thus, the seismic data S'(t) after phase decomposition is obtained, and the expression is as follows:

[0035]

[0036] Wherein, θ1 and θ2 represent the phase range used for stacking the phase gather; when all phases and frequencies are included in the calculation range, S'(t) = S(t), that is, the original seismic data can be reconstructed by formula (2) without loss;

[0037] The phase decomposition method separates the seismic reflection with specific geological significance from the original seismic data. The main implementation process is as follows: firstly, a high-resolution and high-frequency stable phase spectrum is obtained by a seismic complex spectrum decomposition algorithm, then all frequency components are integrated to form a phase gather, and finally, stable decomposition and reconstruction of different phase components of the seismic data are realized.

[0038] Fourthly, the odd component data S o (t) and the even component data S e (t) reflecting the thin layer information of different simple layer structures are obtained to realize dimension reduction of the complex layer structure; the odd component data S o (t) and the even component data S e(t) is obtained by the following method: and r Se (t) is obtained by the following method:

[0039] wherein the odd component data S o (t) reflecting thin layer information of different stratum structures, and even component data S e (t) are obtained by the following method:

[0040] The original seismic data is phase-decomposed to obtain +90° phase component, -90° phase component, 0° phase component and 180° phase component. When two layers are relatively close, the +90° phase component and the -90° phase component reflect two equal-amplitude and opposite-polarity composite waveforms of reflected waves, while the 0° phase component and the 180° phase component reflect two equal-amplitude and same-polarity composite waveforms of reflected waves. Therefore, according to the definition of signal even-odd decomposition, the data obtained by summing the -90° phase component and the +90° phase component is called odd component data S o (t), and the data obtained by summing the 0° phase component and the 180° phase component is called even component data S e (t), which represents the reflected information of different stratum structure thin layers.

[0041] The main factors affecting the effect of the conventional spectral inversion method are the signal-to-noise ratio of the seismic data and the accuracy of the extracted initial wavelet, so in actual application, corresponding denoising processing should be performed to ensure that the seismic data has a high signal-to-noise ratio, and a more practical wavelet extraction method should be used to extract a more accurate seismic wavelet, which is a key technical link for the conventional spectral inversion method to achieve good results.

[0042] In the fifth step, the inversion result obtained in the fourth step is subjected to data fusion reconstruction processing to obtain a final accurate high-resolution seismic reflection coefficient body r(t), and the expression is as follows:

[0043]

[0044] wherein r(t) represents the final obtained accurate high-resolution seismic reflection coefficient body, r Se (t) represents the seismic reflection coefficient body obtained by spectral inversion using the reduced dimension odd component data, and r wb (t) represents the seismic reflection coefficient body obtained by spectral inversion using the reduced dimension even component data.

[0045] In the sixth step, the seismic reflection coefficient body r(t) obtained in the fifth step is subjected to convolution operation with a wide-frequency zero-phase wavelet w wb(t) provides a data foundation for the identification of thin layers and the fine characterization of reservoirs in complex lithological oil and gas reservoirs.

[0046] Broadband amplitude-preserving seismic data S wb The expression for (t) is as follows:

[0047] S wb (t)=r(t)*w wb (t) (4)

[0048] In practical applications, the reflection coefficient can be further processed, such as obtaining relative wave impedance data through trace integral methods or color inversion to predict thin reservoirs.

[0049] To better illustrate the effects of the above-described specific implementation methods, a concrete example is given below:

[0050] Based on actual geological conditions and well logging and seismic data, a two-dimensional multi-layer geological model was designed (e.g. Figure 3 (a) is shown as the implementation model for spectral inversion and overlay processing, where the reservoir lithology is low-impedance sandstone, and the non-reservoir lithologies are low-impedance mudstone and high-impedance mudstone, respectively. Table 1 gives the specific parameters of this two-dimensional multilayer geological model.

[0051] Table 1

[0052]

[0053] Based on the model parameters and the convolution model, a zero-phase Ricker wavelet with a dominant frequency of 30Hz was selected for forward modeling. Figure 3 b represents the seismic record obtained through forward modeling. Further, the forward modeling seismic record is decomposed using phase decomposition to obtain odd and even component data reflecting thin-layer information of different simple layer structures, such as... Figure 4 As shown. Comparison Figure 4 a and Figure 4 As can be seen from b, since the designed forward model can be viewed as a complex stratigraphic structure composed of low-impedance and high-impedance thin layers, the energy of the decomposed data is mainly concentrated in the odd component data. Therefore, the research approach of reducing the dimensionality of complex stratigraphic structures by decomposing the original seismic data is feasible.

[0054] In this theoretical model test, to further highlight the advantages and effectiveness of the spectral inversion and frequency extension processing method based on stratigraphic structure dimensionality reduction, the conventional spectral decomposition method was first used to... Figure 3 The forward modeled seismic records shown in b are subjected to frequency upscaling. Figure 5a is the seismic profile obtained after frequency expansion processing by using conventional spectral inversion method, from the figure, it can be seen that in the complex stratum structure, the conventional seismic spectral inversion is unstable, the odd-even decomposition of the reflection coefficient has multiple solutions when solving the objective function, resulting in the reduction of the accuracy of the spectral inversion frequency expansion result, and some false images appear on the profile. Figure 5 b is the seismic profile obtained after frequency expansion processing by using the spectral inversion frequency expansion processing method based on stratum structure dimension reduction, the frequency expansion result is stable and reliable, and the seismic resolution is significantly improved.

[0055] Further, in order to test the high stability and reliability of the spectral inversion frequency expansion processing method based on stratum structure dimension reduction, the method is applied to the actual data processing in the research area according to the flowchart of the spectral inversion frequency expansion processing method based on stratum structure dimension reduction. Figure 6 As shown in the figure, the frequency expansion processing result of the seismic profile of well A in the research area by using the method of the application. Figure 6 a is the original seismic profile, Figure 6 b is the seismic profile obtained after frequency expansion processing by using the method of the application. Drilling reveals that well A drilled multiple reservoirs in the lower member of the Neogene system, and most of them are mainly thin interbeds, which are limited by the resolution of seismic data and affected by seismic reflection tuning interference, Figure 6 the original seismic data of a cannot finely depict the thin interbed structure reservoir, and Figure 6 the wideband data of b processed by the method of the application has a higher degree of agreement with the actually drilled reservoir on the well, which shows that it is better than the original seismic data in resolution and accuracy, has better thin layer description ability, and embodies the practical application advantages of the application.

[0056] In summary, the application analyzes the seismic response characteristics and mechanism of the complex stratum structure model through forward simulation technology, establishes the relationship between the seismic response characteristics of different stratum structure models and the corresponding waveform phase information. A spectral inversion frequency expansion processing method based on stratum structure dimension reduction is innovatively proposed, the original seismic data is decomposed by using a phase decomposition method to obtain odd and even component data reflecting thin layer information of different simple stratum structures, the dimension of the complex stratum structure is reduced in the form of decomposing the original seismic data, then the odd and even component data with relatively simple stratum structure are respectively subjected to spectral inversion and data fusion reconstruction processing to obtain a stable and reliable seismic reflection coefficient volume, and finally the reflection coefficient volume is subjected to convolution operation with a wideband zero-phase wavelet to obtain accurate high-resolution seismic data. Compared with the conventional spectral inversion method, the new method finds the optimal solution in the data space after dimension reduction, thereby reducing the data space size of the inversion optimization and improving the reliability and stability of the spectral inversion frequency expansion result, which plays an important role in improving the thin layer identification and other fine seismic reservoir characterization capabilities of the seismic spectral inversion result, and provides a strong guarantee for fine lithology exploration and efficient and accurate development of oilfields.

[0057] The above merely describes the preferred embodiments of the present application, and is intended to explain the technical concept and characteristics of the present application, and is not intended to limit the present application in any form. Any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present application still falls within the scope of the technical solutions of the present application.

Claims

1. A spectral inversion and overlay processing method based on dimensionality reduction of stratigraphic structure, characterized in that, Includes the following steps: The first step is to acquire post-stack seismic data after denoising and zero-phase preprocessing. ; The second step involves comprehensively applying well logging, seismic, and geological information to establish forward modeling models of different simple thin-layered stratigraphic structures. These models are then combined to form a complex thin interbedded geological model. The seismic response characteristics of the forward modeling models of different simple thin-layered stratigraphic structures are analyzed, and the relationship between the seismic response characteristics of the models and the corresponding waveform phase information is established. The third step is to analyze the earthquake data. Phase decomposition is performed to obtain seismic data. Phase gather And the seismic data S'(t) after phase decomposition; The fourth step is to obtain odd component data that reflects information about thin layers with different simple layer structures. Coupled component data This enables dimensionality reduction of complex geological structures. For odd component data respectively Even component data Performing conventional spectral inversion processing yields inversion results including seismic reflection coefficients. and ; The fifth step involves performing data fusion and reconstruction on the inversion results obtained in step four to obtain the final high-resolution seismic reflection coefficient volume. ; Step 6: Analyze the seismic reflection coefficients obtained in Step 5. With broadband zero-phase wavelet Perform convolution operations to obtain broadband amplitude-preserving seismic data. This provides a data foundation for the identification of thin layers and the fine characterization of reservoirs in complex lithological oil and gas reservoirs.

2. The spectral inversion and overlay processing method based on dimensionality reduction of stratigraphic structure according to claim 1, characterized in that, In the second step, the different simple thin-layer geological models specifically include four types of thin-layer geological models: high-resistivity thin-layer model, low-resistivity thin-layer model, impedance-increasing thin-layer model, and impedance-decreasing thin-layer model; the low-resistivity thin-layer model and the high-resistivity thin-layer model are reverse polarity thin layers, and the impedance-increasing thin-layer model and the impedance-decreasing thin-layer model are same polarity thin layers.

3. The spectral inversion and overlay processing method based on dimensionality reduction of stratigraphic structure according to claim 2, characterized in that, The seismic response characteristics of the low-impedance thin layer are exhibited as a -90° phase waveform, the seismic response characteristics of the high-impedance thin layer are exhibited as a +90° phase waveform, the seismic response characteristics of the impedance-increasing thin layer are exhibited as a 0° phase waveform, and the seismic response characteristics of the impedance-decreasing thin layer model are exhibited as a 180° phase waveform.

4. The spectral inversion and overlay processing method based on dimensionality reduction of stratigraphic structure according to claim 1, characterized in that, In the fourth step, the odd component data This is the data obtained by summing the -90° phase component and the +90° phase component; the even component data... It is the data obtained by summing the 0° phase component and the 180° phase component.

5. The spectral inversion and overlay processing method based on dimensionality reduction of stratigraphic structure according to claim 1, characterized in that, In the third step, the seismic data Phase gather The expression is as follows: （1）； Where f1 and f2 represent the frequency range used for integration, This indicates the use of the seismic complex spectrum decomposition method to process raw seismic data. The results obtained from time-frequency analysis Indicates frequency, Indicates phase, Indicates time, Representing earthquake data The phase gathers obtained after phase decomposition represent the relationship between the amplitude and time of a single phase component in the seismic data.

6. The spectral inversion and overlay processing method based on dimensionality reduction of stratigraphic structure according to claim 1, characterized in that, In the third step, the expression for the phase-decomposed seismic data S'(t) is as follows: （2）； Where θ1 and θ2 represent the phase range used when stacking phase gathers; when all phases and frequencies are included in the calculation range, the original seismic data can be reconstructed without loss using formula (2). .

7. The spectral inversion and overlay processing method based on dimensionality reduction of stratigraphic structure according to claim 1, characterized in that, In the fifth step, the high-resolution seismic reflection coefficient volume The expression is as follows: （3）； in, This represents the final accurate high-resolution seismic reflection coefficient volume. This represents the volume of seismic reflection coefficients obtained after spectral inversion using the dimensionality-reduced odd component data. This represents the volume of seismic reflection coefficients obtained by spectral inversion using the even component data after dimensionality reduction.

8. The spectral inversion and overlay processing method based on dimensionality reduction of stratigraphic structure according to claim 1, characterized in that, In the sixth step, the broadband amplitude-preserving seismic data The expression is as follows: （4）。 9. The spectral inversion and overlay processing method based on stratigraphic structure dimensionality reduction according to claim 1, characterized in that, In the sixth step, the seismic reflection coefficient is also selected. By processing the data using trace integral methods or color inversion methods, relative wave impedance is obtained, providing a data foundation for thin-layer identification and fine reservoir characterization of complex lithological oil and gas reservoirs.

Images