New method for characterization of dense channel reservoirs based on two-dimensional transient extraction superposition

Abstract

The application discloses a two-dimensional transient extraction complex superimposed dense channel reservoir characterization method, which comprises the following steps: S1, inputting a two-dimensional seismic profile d(t, x), wherein t is time and x is trace number; S2, performing short-time Fourier transform S(t, omega, x, k) on the seismic profile by using a two-dimensional Gaussian window function, wherein omega is frequency and k is wave number; S3, obtaining group delay t0(t, omega, x, k) in time and group delay x0(t, omega, x, k) in space by respectively taking partial derivatives of S(t, omega, x, k) with respect to frequency and wave number; S4, obtaining a two-dimensional extraction operator TDTEO(t, omega, x, k); S5, further obtaining a two-dimensional transient extraction transform TDTET(t, omega, x, k) of seismic data; S6, determining the wave number by energy maximization, obtaining a time-frequency-space domain three-dimensional data body TDTET(t, omega, x, k(t, x, omega)), and obtaining corresponding single-frequency profiles according to required frequencies to perform morphology characterization on the single-frequency profiles. The application can improve energy focusing, effectively eliminate interference phenomenon and be better applied to complex superimposed dense channel characterization.

Description

Technical Field

[0001] This invention relates to a method for processing petroleum geophysical signals, and proposes a two-dimensional transient extraction method for characterizing complex superimposed tight channel reservoirs. Background Technology

[0002] However, these reservoirs typically exhibit characteristics such as multiple vertical stacking phases, thin sand bodies, rapid lateral variations, and dense lithology, leading to significant challenges in exploration and development. Therefore, developing high-precision characterization methods for complex, stacked, and dense channel reservoirs is urgently needed.

[0003] In the past, many scholars have used spectral decomposition techniques to identify channel reservoirs, but these techniques mostly focus on processing seismic data in a single channel, ignoring the spatial distribution of reservoirs in the longitudinal and transverse directions. Traditional spectral decomposition techniques are limited in accuracy in characterizing complex superimposed tight channel reservoirs due to the constraints of the Heisenberg uncertainty principle. At the same time, interference phenomena caused by thin sand bodies make it impossible to characterize them in detail. Summary of the Invention

[0004] The purpose of this invention is to propose a two-dimensional transient extraction method for characterizing complex, superimposed tight channel reservoirs by introducing a spatial window based on transient extraction transformation. In this method, we derive a two-dimensional extraction operator that can realistically characterize the boundaries of complex, superimposed tight channel reservoirs and, to some extent, improve the interference phenomenon caused by thin sand bodies.

[0005] To achieve the above objectives, the technical solution adopted by this invention is as follows: a two-dimensional transient extraction method for characterizing complex superimposed tight channel reservoirs, comprising the following steps:

[0006] S1. Input a two-dimensional seismic profile d(t, x), where t is time and x is the number of traces;

[0007] S2. Perform a short-time Fourier transform S(t, ω, x, k) on the seismic profile using a two-dimensional Gaussian window function, where ω is the frequency and k is the wave number.

[0008] S3. Take the partial derivatives of S(t, ω, x, k) with respect to frequency and wavenumber respectively to obtain the corresponding group delay t0(t, ω, x, k) in time and group delay x0(t, ω, x, k) in space.

[0009] S4. Obtain the two-dimensional extraction operator TDTEO(t, ω, x, k);

[0010] S5. Finally, the two-dimensional transient extraction transform of the seismic data, TDTET(t, ω, x, k), is obtained.

[0011] S6. Determine the wave number by maximizing its energy, thereby obtaining the three-dimensional data volume TDTET(t, ω, x, k(t, x, ω)) in the time-frequency-spatial domain of TDTET. Then, obtain the corresponding single-frequency profile according to the required frequency to characterize its shape.

[0012] Preferably, in step S2, the seismic profile is processed using a two-dimensional Gaussian window function, with the formula S(t, ω, x, k) as follows:

[0013]

[0014] in Let h(t, x) be the imaginary unit of the complex number, and let h(t, x) be a two-dimensional Gaussian window function of the form:

[0015]

[0016] Where σ t σ x These represent the standard deviations of the Gaussian window function in the t and x directions, respectively.

[0017] Preferably, in step S3, the estimation formulas for the temporal group delay t0(t, ω, x, k) and the spatial group delay x0(t, ω, x, k) are:

[0018]

[0019]

[0020] Where th represents the product of the window function and time, and xh represents the product of the window function and the number of channels.

[0021] As a preferred option: the extraction operator defined in step S4 is:

[0022] TDTEO(t,ω,x,k)=δ(t-t0(t,ω,x,k))δ(x-x0(t,ω,x,k))

[0023] Where δ(·) represents the Dirac function.

[0024] As a preferred embodiment: In step S5, the two-dimensional transient extraction transform is obtained as follows:

[0025] TDTET(t,ω,x,k)=S(t,ω,x,k)·δ(t-t0(t,ω,x,k))δ(x-x0(t,ω,x,k))

[0026] As a preferred option: In step S6, the specific expression is as follows:

[0027]

[0028] Thus, the time-frequency-spatial domain three-dimensional data volume TDTET(t, ω, x, k(t, ω, x)) is obtained.

[0029] Compared with existing technologies, the advantages of this invention are: by balancing the longitudinal and lateral transformations of seismic data through a spatial window, it simultaneously obtains estimates of both temporal and spatial group delays, thereby defining a new extraction operator and constructing a two-dimensional transient extraction transformation characterization method. This method can better capture the spatial distribution characteristics of reservoirs in seismic data, effectively eliminate interference phenomena, and improve energy focusing. Attached Figure Description

[0030] Figure 1 This is a flowchart of the present invention;

[0031] Figure 2 This is a seismic profile of the synthesized wedge-shaped model;

[0032] Figure 3 The following are examples of single-frequency profiles extracted from the wedge model using other methods and this method, with a frequency of 35Hz: (a) is the profile after short-time Fourier transform processing, (b) is the profile after two-dimensional short-time Fourier transform processing, (c) is the profile after time rearrangement synchronous compression transform processing, (d) is the profile after two-dimensional time rearrangement synchronous compression transform processing, (e) is the profile after transient extraction transform processing, and (f) is the profile after two-dimensional transient extraction transform processing.

[0033] Figure 4 In order to be in Figure 3 Based on this, the single-channel time-frequency diagrams of one-eighth of the wavelength are obtained: (a) is the time-frequency diagram after STFT processing, (b) is the time-frequency diagram after TDSTFT processing, (c) is the time-frequency diagram after TSST processing, (d) is the time-frequency diagram after TDTSST processing, (e) is the time-frequency diagram after TET processing, and (f) is the time-frequency diagram after TDTET processing. Detailed Implementation

[0034] The invention will now be further described with reference to the accompanying drawings.

[0035] Example 1: See Figure 1 A two-dimensional transient extraction method for characterizing complex superimposed tight channel reservoirs includes the following steps:

[0036] S1. Input a two-dimensional seismic profile d(t, x), where t is time and x is the number of traces;

[0037] S2. The short-time Fourier transform S(t, ω, x, k) of the seismic profile using a two-dimensional Gaussian window function is as follows:

[0038]

[0039] in Let h(t, x) be the imaginary unit of the complex number, and let h(t, x) be a two-dimensional Gaussian window function of the form:

[0040]

[0041] Where σ t σ x These represent the standard deviations of the Gaussian window function in the t and x directions, respectively.

[0042] S3. Taking the partial derivatives of S(t, ω, x, k) with respect to frequency and wavenumber respectively, we obtain the corresponding group delay t0(t, ω, x, k) in time and group delay x0(t, ω, x, k) in space as follows:

[0043]

[0044]

[0045] Where th represents the product of the window function and time, and xh represents the product of the window function and the number of channels.

[0046] S4. Obtain the two-dimensional extraction operator TDTEO(t, ω, x, k):

[0047] TDTEO(t,ω,x,k)=δ(t-t0(t,ω,x,k))δ(x-x0(t,ω,x,k))

[0048] Where δ(·) represents the Dirac function.

[0049] S5. Finally, the two-dimensional transient extraction transform TDTET(t, ω, x, k) of the seismic data is obtained:

[0050] TDTET(t,ω,x,k)=S(t,ω,x,k)·δ(t-t0(t,ω,x,k))δ(x-x0(t,ω,x,k))

[0051] S6. The wavenumber is determined by maximizing its energy. The wavenumber expression is:

[0052]

[0053] This yields the time-frequency-spatial domain three-dimensional data volume TDTET(t, ω, x, k(t, ω, x)), and then the corresponding single-frequency profile is obtained according to the required frequency to characterize its shape.

[0054] See Figures 2 to 4 Specifically, we take the synthesized wedge-shaped model seismic profile as an example, such as... Figure 2Using short-time Fourier transform, two-dimensional short-time Fourier transform, time redistribution synchronous compression transform, two-dimensional time redistribution synchronous compression transform, instantaneous extraction transform, and two-dimensional instantaneous extraction transform, a single-frequency profile of 35Hz was extracted from the wedge model. Figure 3 As can be seen, while Figures (a)-(d) show good separation results, their energy focusing is poor. Figure (e), although exhibiting good energy focusing, shows significant interference. Figure (f) shows the result processed by the method of this invention, effectively eliminating interference while maintaining high energy accuracy. Figure 4 exist Figure 3 Based on this, taking a single-channel time-frequency diagram of one-eighth of the wavelength can more clearly demonstrate the effectiveness of the method of the present invention.

[0055] The above examples are only used to illustrate the present invention. The implementation steps of the method can be changed. All equivalent transformations and improvements made on the basis of the technical solution of the present invention should not be excluded from the protection scope of the present invention.

Claims

1. A two-dimensional transient extraction method for characterizing complex superimposed tight channel reservoirs, characterized in that, Includes the following steps: S1, Input 2D seismic profile ,in For time, For the number of paths; S2. Perform short-time Fourier transform on the seismic profile using a two-dimensional Gaussian window function. ,in For frequency, Wave number; S3, will Taking the partial derivatives with respect to frequency and wavenumber respectively, we obtain the corresponding group delay in time. and spatial group delay ; S4. Obtain the two-dimensional extraction operator. ,in Represents the Dirac function; S5, thereby obtaining the two-dimensional transient extraction transformation of seismic data. ; S6. Determine the wavenumber by maximizing its energy to obtain three-dimensional data in the time-frequency-spatial domain. Based on the required frequency, the corresponding single-frequency profile is obtained to characterize its shape.

2. The two-dimensional transient extraction method for characterizing complex superimposed tight channel reservoirs according to claim 1, characterized in that: In step S2, a two-dimensional Gaussian window function is used to analyze the seismic profile. for: ； in The imaginary part of a complex number is the unit. It is a two-dimensional Gaussian window function, in the form of: ； in , These represent the Gaussian window function at... direction and Standard deviation in direction.

3. The two-dimensional transient extraction method for characterizing complex superimposed tight channel reservoirs according to claim 1, characterized in that: In step S3, the group delay in time and spatial group delay The estimation formula is: ； ； in This represents the product of the window function and time. This represents the product of the window function and the number of channels.

4. The two-dimensional transient extraction method for characterizing complex superimposed tight channel reservoirs according to claim 1, characterized in that: In step S6, the specific expression is: ； Thus obtain Time-frequency-spatial domain three-dimensional data volume Then, based on the required frequency, the corresponding single-frequency profile is obtained to characterize its shape.

Images