A method for spatial-temporal characterization of lateral consistency of a tight sandstone reservoir

Abstract

The application discloses a method for characterizing the spatial and temporal consistency of a tight sandstone reservoir, comprising the following steps: (1) inputting seismic data s(t, x), wherein t is time and x is the number of channels; (2) constructing a kernel function ker(t, x, f, theta) varying with strata, wherein f is frequency and theta is a strata dip angle introduced; (3) setting a strata dip angle discrete sequence {theta k} k＝1,2,…,N , N is the number of discrete points; (4) calculating the spatial and temporal consistency short-time Fourier transform V k (t, f, x, theta s ) under each theta k ; (5) determining the optimal strata dip angle (6) obtaining the time-frequency-space characterization result of the spatial and temporal consistency optV s (t, f, x); (7) performing feature extraction at a frequency f0 to obtain a single-frequency profile optV s (t, f0, x). The application introduces a strata dip angle parameter and constructs a kernel function varying with strata on the basis of a short-time Fourier transform (or other equivalent transform and improved time-frequency transform method), can better utilize seismic data to characterize the spatial distribution characteristics of a tight sandstone reservoir and has good noise robustness.

Description

TECHNICAL FIELD

[0001] The present application relates to a signal processing method, and proposes a transverse consistency space-time characterization method for tight sandstone reservoirs. BACKGROUND

[0002] China's oil and gas exploration and development has entered a new era of unconventional oil and gas, among which tight oil and gas resources are rich and have great development potential. The complex superimposed tight channel reservoir is one of the most potential tight oil and gas reservoir types in China's continental sedimentation. Effective characterization and efficient exploration of the reservoir are related to the success or failure of well deployment and rolling exploration and development. However, due to the problems of multi-period superimposition in the vertical direction, thin sand body thickness, rapid lateral change and dense lithology, it is difficult to depict the spatial of the channel sand body and characterize the internal heterogeneity of the reservoir, which further restricts the exploration and development precision of the channel sandstone reservoir.

[0003] For the identification of tight sandstone reservoirs, time-frequency analysis is an effective tool. It reflects the different response characteristics between different scale geological bodies at different frequencies by analyzing non-stationary seismic signals, and further reveals important geological information hidden in seismic data. Common time-frequency analysis methods include short-time Fourier transform, wavelet transform, S transform and Wigner-Ville distribution. The processing of seismic data by these methods can be generally summarized as continuous time-frequency analysis of time points for each trace to obtain its frequency spectrum. However, this trace-by-trace analysis does not consider the lateral heterogeneity or lateral discontinuity of geological bodies, thereby ignoring the lateral spatial distribution of reservoir characterization, resulting in poor characterization of the continuity of lateral changes in seismic data. SUMMARY

[0004] In view of the above problems in the prior art, the present application proposes a transverse consistency space-time characterization method for tight sandstone reservoirs. In this method, the stratum is regarded as locally approximately flat, and based on short-time Fourier transform, a stratum dip angle parameter is introduced and a kernel function varying with stratum is constructed, so that better time-frequency-space distribution characteristics can be obtained.

[0005] To achieve the above purpose, the technical scheme adopted by the present application is as follows: a transverse consistency space-time characterization method for tight sandstone reservoirs, comprising the following steps:

[0006] (1) inputting seismic data s(t,x), wherein t is time and x is trace number;

[0007] (2) constructing a kernel function ker(t,x,f,θ) varying with stratum, wherein f is frequency and θ is the introduced stratum dip angle;

[0008] (3) setting a stratum dip angle discrete sequence {θ k} k＝1,2,…,N , N is the number of discrete points;

[0009] (4) Calculate the lateral consistency short-time Fourier transform V k (t,f,x,θ s ) for each θ k ;

[0010] (5) Determine the optimal formation dip angle θ

[0011] (6) Obtain the lateral consistency time-frequency-space representation result optV s (t,f,x) for the optimal formation dip angle θ

[0012] (7) Perform feature extraction at frequency f0, and obtain the single-frequency profile optV s (t,f0,x) for the optimal formation dip angle θ

[0013] As a preferred, in step (2), the specific formation-varying kernel function is constructed as:

[0014] ker(t,x,f,θ) = h(t,x)e j2πf(t-xtanθ) ,

[0015] where h(t,x) is a two-dimensional Gaussian window function, which can be in the form of:

[0016]

[0017] where σ t , σ x are the standard deviations of the Gaussian window;

[0018] As a preferred, in step (3), the set of formation dip angle discrete sequences {θ k} k＝1,2,…,N is:

[0019]

[0020] As a preferred, in step (4), the defined lateral consistency short-time Fourier transform is:

[0021]

[0022] Other equivalent transforms and improved time-frequency transform methods can all obtain lateral consistency transform results on the basis of the present application.

[0023] As a preferred, in step (5), the determination method of the optimal formation dip angle is:

[0024]

[0025] As a preferred, in step (6), the specific lateral consistency time-frequency-space representation result is:

[0026]

[0027] Compared with the prior art, the method has the advantages that by introducing the stratum dip angle parameter and constructing a kernel function varying with the stratum, a spatial and temporal characterization method of the tight sandstone reservoir is provided. The method can better capture the spatial distribution characteristics of the reservoir in the seismic data, and has stronger robustness to noise. BRIEF DESCRIPTION OF DRAWINGS

[0028] Figure 1 is a flowchart of the present application;

[0029] Figure 2 is a profile of a certain tight sandstone gas field in western Sichuan, China obtained by inputting seismic data;

[0030] Figure 3 is a single frequency profile with a frequency of 32Hz and 60Hz respectively extracted after processing the tight sandstone gas field by the short-time Fourier transform;

[0031] Figure 4 is a single frequency profile with a frequency of 32Hz and 60Hz respectively extracted after processing the tight sandstone gas field by the method of the present application. DETAILED DESCRIPTION

[0032] The present application will be further described below with reference to the accompanying drawings.

[0033] Example 1: see Figure 1 A spatial and temporal characterization method of seismic data by lateral consistency short-time Fourier transform, comprising the following steps:

[0034] (1) inputting seismic data s(t,x), wherein t is time and x is trace number;

[0035] (2) constructing a kernel function varying with the stratum:

[0036] ker(t,x,f,θ)=h(t,x)e j2πf(t-xtanθ) ,

[0037] wherein f is the frequency of the kernel function, θ is the introduced stratum dip angle, h(t,x) is a two-dimensional Gaussian window function, which can be in the form of:

[0038]

[0039] wherein σ t , σ x are the standard deviations of the Gaussian window;

[0040] (3) setting a stratum dip angle discrete sequence {θ k} k＝1,2,…,N , N is the discrete point number:

[0041]

[0042] (4) Calculate the short-time Fourier transform of each θ k

[0043]

[0044] Other equivalent transform and improved time-frequency transform method, can be based on the invention of the horizontal consistency transform results.

[0045] (5) Determine the optimal formation dip angle:

[0046]

[0047] (6) Get the horizontal consistency of time-frequency-space representation results:

[0048]

[0049] (7) Feature extraction at frequency f0, get single frequency profile optV s (t,f0,x);

[0050] The first step of the present application is to input seismic data, the second step is to introduce the formation dip angle structure kernel function, the third step and the fourth step are to set the formation dip angle discrete sequence and calculate the short-time Fourier transform of each θ k

[0051] Example 2:

[0052] Figure 2 The profile of a certain tight sandstone gas field in western Sichuan, China is obtained by inputting data.

[0053] Figure 3 The single frequency profile of the tight sandstone gas field is extracted by short-time Fourier transform at frequency of 32Hz and 60Hz respectively.

[0054] Figure 4 The single frequency profile of the tight sandstone gas field is extracted by the method of the present application at frequency of 32Hz and 60Hz respectively.

[0055] It can be seen that in Figure 3 ​​The short-time Fourier transform is weak in depicting the continuity of the lateral variation of the seismic data, is serious in smearing in the vertical direction, and has poor frequency resolution. Meanwhile, the extracted 60Hz single frequency profile is greatly affected by noise, is not stable in the high frequency case, and is difficult to make reasonable seismic interpretation in the analysis of actual seismic data. Compared with the processing result of the short-time Fourier transform, Figure 4 The seismic data processed by the method of the present application has similar time-frequency spectrum, which indicates the effectiveness of the proposed method. Meanwhile, the result of the method has obvious clearer lateral consistency depiction, the energy distribution is more continuous, the result is more reasonable, the 60Hz single frequency profile is less affected by noise compared with the processing result of the short-time Fourier transform, which indicates that the method of the present application has robustness to noise and can suppress the influence of noise to a certain extent.

[0056] The above examples are only used to illustrate the present application, wherein the steps of the method can be changed, and any equivalent transformation and improvement based on the technical scheme of the present application should not be excluded from the protection scope of the present application.

Claims

1. A method for spatial-temporal characterization of lateral consistency of tight sandstone reservoirs, comprising the following steps: (1) inputting seismic data s(t, x), wherein t is time and x is trace number; (2) constructing a kernel function ker(t, x, f, θ) varying with strata, wherein f is frequency and θ is a strata dip angle introduced; 2. The method according to claim 1, wherein in step (2), the kernel function varying with strata is constructed as follows: (3) setting a stratigraphic dip angle discrete sequence {θ k} k＝1,2,…,N , N is the number of discrete points; (4) Calculate the lateral consistency short-time Fourier transform V k (t,f,x,θ s k under each θ. The specific calculation formula is: (5) determining the optimal formation dip (6) obtaining a time-frequency-space representation result optV that is consistent across the lateral direction s (t,f,x); (7) Perform feature extraction at frequency f0, obtaining single-frequency profile optV s (t, f0, x). wherein h(t, x) is a two-dimensional Gaussian window function, and is in the form of:

3. The method according to claim 1, wherein in step (4), the strata dip angle θ is determined as follows: ker(t,x,f,θ) = h(t,x)e j2πf(t-xtanθ) , 4. The method according to claim 1, wherein in step (5), the optimal strata dip angle is determined as follows: where σ t , σ x is the standard deviation of the Gaussian window.

5. The method according to claim 1, wherein in step (6), the specific spatial-temporal characterization result of lateral consistency is as follows: In step (3), the set formation dip angle discrete sequence {θ k} k＝1,2,…,N is: ​ ​ ​ ​

Images