A comprehensive seismic prediction method for high-quality gas reservoirs

Abstract

The present invention discloses a comprehensive seismic prediction method for high-quality gas-bearing reservoirs. The method comprises: manually collecting and processing seismic data to obtain a volume; analyzing and obtaining seismic elastic parameters sensitive to high-quality gas-bearing reservoirs; fine well-seismic calibration to obtain time-depth relationships and seismic wavelets; extracting high-frequency attenuation areas through time-frequency analysis to qualitatively predict gas-bearing areas; constructing a low-frequency fusion model that conforms to the sedimentary characteristics of high-quality gas-bearing reservoirs; performing simultaneous pre-stack seismic inversion to obtain a sensitive parameter inversion volume; constructing a high-quality gas-bearing reservoir identification map through probabilistic analysis, quantitatively predicting the distribution of high-quality gas-bearing reservoirs, and comprehensively selecting favorable areas for high-quality gas-bearing reservoirs. Compared with existing methods, this method addresses the challenges of low gas-bearing area exploration, a small number of wells, uneven well location distribution, and strong lateral reservoir heterogeneity. By combining well logging, stratigraphic, and pre-stack and post-stack seismic data, it can obtain more accurate spatial distribution characteristics of high-quality gas-bearing reservoirs and reduce the ambiguity of prediction results.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas exploration and development, and more specifically, to a comprehensive seismic prediction method for high-quality gas-bearing reservoirs, specifically targeting scenarios with low exploration levels, few or unevenly distributed wells, and drastic lateral changes in gas-bearing reservoirs. Background Art

[0002] With the continuous growth of global energy demand, natural gas, as a clean and efficient energy source, is attracting increasing attention from the oil exploration and development industry. To ensure a stable energy supply, finding high-quality gas-bearing reservoirs is becoming increasingly important. Effective detection and assessment of gas-bearing reservoirs not only improves the success rate of traditional oil and gas exploration but also promotes the development of unconventional resources such as shale gas and coalbed methane. Traditional geological exploration methods often face complex geological environments and uncertainties. Seismic exploration, as a key geophysical technique, offers the advantages of high precision and efficiency. By leveraging the propagation characteristics of seismic waves in the subsurface, seismic exploration can provide rich information, helping to identify and characterize the geometry and physical properties of gas-bearing reservoirs. However, with the continuous advancement of oil and gas exploration and development, the difficulty of discovering oil and gas is increasing, and single seismic methods often fail to meet actual exploration needs. To address reservoir prediction challenges such as low exploration depth in gas-bearing areas, a small number of wells, uneven well distribution, and strong lateral reservoir heterogeneity, single methods such as seismic attribute analysis and seismic inversion have certain shortcomings, resulting in unsatisfactory prediction results. Therefore, it is urgent to establish a comprehensive seismic prediction method for high-quality gas-bearing reservoirs, explore the distribution patterns of high-quality reservoirs, and improve the overall success rate of exploration wells. Summary of the Invention

[0003] The purpose of the embodiments of the present invention is to provide a comprehensive seismic prediction method for high-quality gas-bearing reservoirs, so as to improve the accuracy of prediction of high-quality gas-bearing reservoirs.

[0004] To achieve the above objectives, an embodiment of the present invention provides a comprehensive seismic prediction method for high-quality gas-bearing reservoirs, the present invention comprising:

[0005] Manually collect 3D seismic data and process them indoors to obtain pre-stack and post-stack pure wave seismic data volumes and seismic velocity volumes;

[0006] The histogram statistics method is used to correct the systematic error among multiple wells by fine-tuning the main peak position.

[0007] Analyze logging data to obtain seismic elastic parameters E that are sensitive to high-quality gas-bearing reservoirs a and E b ;

[0008] Use well logging and prestack seismic data to perform fine well-seismic calibration to obtain time-depth relationships and prestack seismic wavelets;

[0009] Using the horizon and post-stack seismic data, based on the wavelet transform time-frequency analysis method, the attenuation area attributes of the high-frequency band of the spectrum are extracted to qualitatively predict the gas-bearing area A;

[0010] Analyze the spectrum characteristics of seismic data, velocity volume data and well logging data, and determine the low-frequency compensation spectrum range of different data;

[0011] Under the constraints of the fine seismic grid, based on the mathematical interpolation algorithm, the logging data are interpolated and extrapolated to obtain a conventional well interpolation full-band high-resolution model;

[0012] Quality control of the seismic velocity body and its correlation with the well logging curve, extraction of the P-wave velocity curve at the well point, intersecting and fitting it with the actual well logging P-wave velocity, and correction of the deviation between the seismic velocity body and the well logging data;

[0013] Extract the seismic correction velocity volume well point velocity, establish relationship with logging density and shear wave velocity, and convert them into longitudinal wave impedance, shear wave impedance and density model;

[0014] The unified compaction trend in the region is statistically analyzed by using the P-wave impedance of multiple wells after well seismic calibration and time intersection, and the compaction trend body of the study area and the compaction trend curve on the well are obtained.

[0015] The decompaction trend curve on the well is obtained by subtracting the logging longitudinal wave impedance from the compaction trend curve, and the decompaction trend body is interpolated by the interpolation and extrapolation algorithm;

[0016] The compaction trend body and the decompaction trend body of the study area are added together to obtain the compaction trend correction longitudinal wave impedance model. The density and shear wave impedance compaction trend correction model is constructed in the same way.

[0017] Based on the frequency band range of various data types, the well interpolation model, seismic velocity field conversion model and compaction trend correction body are combined in the frequency domain to construct a low-frequency fusion model that conforms to the sedimentary characteristics of high-quality gas-bearing reservoirs.

[0018] Carry out seismic pre-stack simultaneous inversion to obtain E sensitive to high-quality reservoirs a and E b Inversion data volume;

[0019] Using E a and E b The inversion volume and the classification conclusion of the gas-bearing reservoir in the well logging are used to conduct the probability analysis of the gas-bearing reservoir and construct the high-quality reservoir identification plate;

[0020] Based on the high-quality reservoir identification plate, the spatial distribution of high-quality gas-bearing reservoirs is quantitatively predicted to obtain gas-bearing area B;

[0021] Using the qualitative and quantitative prediction results, favorable areas of high-quality gas-bearing reservoirs are selected. Combining all the above processes, a comprehensive seismic prediction method for high-quality gas-bearing reservoirs is formed.

[0022] The embodiments of the present invention have the following advantages: Compared with attribute analysis methods using post-stack seismic data, the method proposed by the present invention combines seismic inversion methods to reduce the ambiguity of prediction results. Compared with conventional pre-stack seismic inversion methods, this method first uses time-frequency domain seismic attribute analysis methods to qualitatively detect gas-bearing zones, and then performs seismic inversion based on a low-frequency fusion model, integrating more geological information, improving the accuracy and resolution of prediction results in areas far from the wellbore, and improving the accuracy of identifying lateral features of the target layer in the near-wellbore control area, thereby improving the overall prediction effect and greatly enhancing the interpretability of prediction results for high-quality gas layers. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0024] Figure 1 This is a flow chart of a comprehensive seismic prediction method for high-quality gas-bearing reservoirs according to an embodiment of the present invention.

[0025] Figure 2 This is a cross-sectional view of the original seismic data of the target layer in an embodiment of the present invention.

[0026] Figure 3 This is a plane distribution diagram of the qualitative prediction of the gas-bearing zone A according to an embodiment of the present invention.

[0027] Figure 4 This is a cross-sectional view of the well line of the longitudinal wave impedance low-frequency fusion model according to an embodiment of the present invention.

[0028] Figure 5 This is a plan view of the longitudinal wave impedance low-frequency fusion model of an embodiment of the present invention.

[0029] Figure 6 This is a high-quality reservoir identification chart according to an embodiment of the present invention.

[0030] Figure 7 This is a quantitative prediction of the planar distribution of the gas-bearing area B according to an embodiment of the present invention.

[0031] Figure 8 This is a comprehensive predicted distribution map of high-quality gas-bearing reservoir development areas according to an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0033] Figure 1 Flowchart of a comprehensive seismic prediction method for high-quality gas-bearing reservoirs according to an embodiment of the present invention.

[0034] like Figure 1 As shown, the method includes the following steps:

[0035] S101. Collect a set of 3D seismic data in the study area. Specifically, select a certain area in a certain place where an underground exploration target has a 3D seismic data obtained by 3D acquisition, and demonstrate the application of the embodiment of the present invention. The area is a deep-water area in a certain sea in my country. The underground exploration target includes a set of shallow natural gas segments, which are continental slope deposits with multi-layer reservoirs developed from bottom to top. The seismic data is affected by the law of shallow gas enrichment, with rapid lateral energy changes and strong heterogeneity. Intercept the seismic data of the target shallow gas segment, Figure 2 This is the seismic data of the shallow gas section in an embodiment of the present invention. The locations of the wells are marked above. There are only three wells in the area, and all are located in a corner of the study area, with extremely uneven distribution.

[0036] S102: Using a histogram statistical method, correct the systematic error between the three wells by fine-tuning the main peak position. In this embodiment, the well logging data includes longitudinal wave impedance, shear wave impedance, and density curves.

[0037] S103. Analyze the logging data to obtain the seismic elastic parameter E that is sensitive to high-quality gas-bearing reservoirs. a and Eb. In this example, the longitudinal wave impedance cannot effectively distinguish between reservoirs and non-reservoirs, so the seismic post-stack inversion method is not applicable, and pre-stack inversion should be selected. The sensitive parameter analysis results of the embodiment of the present invention show that the longitudinal and transverse wave velocity ratio (Vp / Vs) and the Lame constant multiplied by density (λρ) are extremely sensitive to gas layers. Low Vp / Vs and low λρ point to high-quality gas-bearing reservoirs with high porosity, high gas saturation, high permeability and low mud content. Vp / Vs and λρ are preferably used as sensitive elastic parameters of high-quality gas-bearing reservoirs for simultaneous pre-stack inversion.

[0038] S104: Perform fine well-seismic calibration using well logging and pre-stack seismic data to obtain time-depth relationships and seismic wavelets. In this embodiment, the pre-stack seismic data is three partial angle stack data, namely 0-20 degrees, 20-40 degrees, and 40-60 degrees, corresponding to three comprehensive wavelets.

[0039] S105. Using the horizon and post-stack seismic data, based on the wavelet transform time-frequency analysis method, extract the attenuation area attribute of the high frequency band of the spectrum and qualitatively predict the gas-bearing area A, such as Figure 3 shown.

[0040] S106. Analyze the spectral characteristics of seismic data, velocity volume data, and well logging data to determine the spectral ranges for low-frequency compensation for each data type. In this example, the shallow gas reservoir is shallowly buried, resulting in a strong low-frequency component in the seismic data signal, with an effective frequency band of 7-60 Hz, and missing low-frequency information below 7 Hz. The effective frequency band of the velocity field is 0-10 Hz. To combine the advantages of various data types, the seismic velocity field conversion model, the compaction trend correction model, and the conventional well interpolation model compensate for low-frequency energy of 0-4 Hz, 4-6 Hz, and 6-10 Hz, respectively.

[0041] S107. Within the constraints of the fine seismic grid, a mathematical interpolation algorithm is used to interpolate and extrapolate the well logging data to obtain a high-resolution, full-band model for conventional wells. In this example, data from only three wells is available, so a locally weighted interpolation algorithm is used to obtain P-wave impedance, S-wave impedance, and density models.

[0042] S108, quality control velocity volume and its correlation with well logging curves, extract the P-wave velocity curve at the well point, intersect and fit it with the actual well logging P-wave velocity, and correct the error between the seismic velocity volume data and the well logging data. In this example, the relationship is a linear function:

[0043] y=825.221+0.66794x (Formula 1) where x is the wellpoint velocity field velocity, and y is the measured longitudinal wave velocity, both in units of m / s.

[0044] S109, extract the well point velocity of the seismic correction velocity volume, and establish a relationship with the density and shear wave velocity of the logging data. In this example, the relationship is in the form of a linear function:

[0045] y=1.86778+0.00015075x1 (Formula 2)

[0046] y=195.681+0.366869x² (Formula 3)

[0047] Where x1 is the density in g / cm 3 , x2 is the shear wave velocity, y is the corrected volume well point velocity, all in m / s. Using formulas (1)-(3), the longitudinal wave impedance, shear wave impedance, and density model are obtained.

[0048] S110. Utilize the multi-well P-wave impedance and time intersection after well seismic calibration to calculate the uniform compaction trend in the region. In this example, the compaction trend relationship is:

[0049] y=2035.23+3013.76t (Formula 4)

[0050] Where t is the double layer travel time in seconds and y is the longitudinal wave impedance in g / cm 3 m / s; the compaction trend body in the study area and the compaction trend curve on the well are calculated according to formula (4).

[0051] S111, subtracting the well logging longitudinal wave impedance from the compaction trend curve to obtain a decompaction trend curve on the well, and interpolating it into a decompaction trend body through an interpolation and extrapolation algorithm;

[0052] S112. Add the compaction trend body and the decompaction trend body of the study area to calculate the compaction trend correction longitudinal wave impedance model. The density and shear wave impedance compaction trend correction model is constructed in the same way. In this example, the density, shear wave impedance and time intersection fitting relationship is:

[0053] y1=1.89931+0.314055t (Formula 5)

[0054] y2=348.158+1940.9t (Formula 6)

[0055] Where t is the round-trip travel time in seconds and y1 is the density in g / cm 3 , y2 is the shear wave impedance, unit is g / cm 3 ·m / s.

[0056] S113. Based on the frequency band range of various data types, the well interpolation model, seismic velocity field conversion model, and compaction trend correction body are merged in the frequency domain to construct a low-frequency fusion model that conforms to the sedimentary characteristics of high-quality gas-bearing reservoirs. Figure 4 The longitudinal wave impedance low-frequency fusion model constructed in the embodiment of the present invention is connected to the well line profile. The fusion model incorporates more geological information, the vertical characteristics are more reasonable, the well and well curves are highly consistent, and the wells are consistent with the regional change trend. Figure 5 A plan view of the low-frequency fusion model of longitudinal wave impedance constructed for an embodiment of the present invention. The fusion model reduces the lateral patterning and uncertainty caused by extrapolation of the far-well area, improves the "bull's eye" phenomenon around the well in the compaction trend correction model, and improves the accuracy and rationality of the initial inversion model.

[0057] S114. Perform pre-stack simultaneous inversion to obtain Vp / Vs and λρ inversion data volumes that are sensitive to high-quality gas-bearing reservoirs.

[0058] S115. Use Vp / Vs and λρ inversion volume and well logging gas reservoir classification conclusion to perform gas reservoir probability analysis, and build a high-quality reservoir identification chart based on Bayesian fuzzy discriminant theory. In this example, considering the well logging gas reservoir classification conclusion of five types of lithology (type 1 high-quality reservoir, type 2 high-quality reservoir, type 3 high-quality reservoir, dry layer and mudstone), an intersection analysis is carried out based on the two elastic parameters Vp / Vs and λρ, and the Gaussian function is used to fit the two-dimensional probability density function of type 1 high-quality reservoir and other layers, corresponding to Figure 6 The closer the two-color ellipse is to the center of the ellipse, the greater the possibility of lithofacies or fluid development, and a high-quality reservoir identification chart is established.

[0059] S116. Based on the high-quality reservoir identification plate, quantitatively predict the spatial distribution of gas-bearing reservoirs to obtain gas-bearing area B. In this example, the high-quality reservoir identification plate is applied to the Vp / Vs and λρ data volumes obtained by prestack inversion to calculate the probability volume of high-quality reservoirs. A high-quality gas-bearing reservoir is characterized by a probability range of 0.7-1. Figure 7 To quantitatively predict the spatial distribution of gas-bearing area B.

[0060] S117. Using the qualitative prediction results and the quantitative prediction results, the favorable areas of high-quality gas-bearing reservoirs are selected to form a comprehensive seismic prediction method for high-quality gas-bearing reservoirs. In this example, the qualitative prediction result A and the quantitative prediction result B are superimposed and displayed. The area B included in the range of A is selected as the comprehensive high-quality gas-bearing reservoir development area ( Figure 8 ).

[0061] Although the present invention has been described in detail above using general descriptions and specific embodiments, it will be apparent to those skilled in the art that modifications and improvements may be made thereto. Therefore, such modifications and improvements, without departing from the spirit of the present invention, are intended to be within the scope of protection claimed herein.

Claims

1. A comprehensive seismic prediction method for high-quality gas-bearing reservoirs, characterized by The steps include: (1) Manually collect 3D seismic data and process them indoors to obtain pre-stack and post-stack pure wave seismic data volumes and seismic velocity volumes; (2) Using the histogram statistics method, the systematic error among multiple wells is corrected by fine-tuning the main peak position; (3) Analyze logging data to obtain seismic elastic parameters E that are sensitive to high-quality gas-bearing reservoirs a and E b ; (4) Use well logging and prestack seismic data to perform fine well-seismic calibration to obtain time-depth relationships and prestack seismic wavelets; (5) Using the horizon and post-stack seismic data, based on the wavelet transform time-frequency analysis method, the attenuation area attributes of the high-frequency band of the spectrum are extracted to qualitatively predict the gas-bearing area A; (6) Analyze the spectrum characteristics of seismic data, velocity volume data, and well logging data to determine the low-frequency compensation spectrum range of different data; (7) Under the constraints of the fine seismic grid, based on the mathematical interpolation algorithm, the logging data are interpolated and extrapolated to obtain a conventional well interpolation full-band high-resolution model; (8) Quality control of the seismic velocity volume and its correlation with the well logging curve, extraction of the P-wave velocity curve at the well point, cross-fitting with the actual well logging P-wave velocity, and correction of the deviation between the seismic velocity volume data and the well logging data; (9) Extract the seismic correction velocity volume well point velocity, establish a relationship with the well logging density and shear wave velocity, and convert them into the longitudinal wave impedance, shear wave impedance and density model; (10) Using the multi-well P-wave impedance and time intersection after well seismic calibration to calculate the unified compaction trend in the region, the compaction trend body of the study area and the compaction trend curve on the well are obtained; (11) Subtracting the well logging longitudinal wave impedance from the compaction trend curve to obtain the decompaction trend curve on the well, and interpolating it into the decompaction trend body through the interpolation and extrapolation algorithm; (12) The compaction trend body and the decompaction trend body of the study area are added together to obtain the compaction trend correction longitudinal wave impedance model. The density and shear wave impedance compaction trend correction model is constructed in the same way. (13) Based on the frequency band range of various data, the well interpolation model, seismic velocity field conversion model and compaction trend correction body are merged in the frequency domain to construct a low-frequency fusion model that conforms to the sedimentary characteristics of high-quality gas-bearing reservoirs; (14) Carry out pre-stack seismic simultaneous inversion to obtain E that is sensitive to high-quality reservoirs a and E b Inversion data volume; (15) Using E a and E b Inversion data and well logging gas reservoir classification conclusions are used to perform gas reservoir probability analysis and construct high-quality reservoir identification charts; (16) Based on the high-quality reservoir identification plate, quantitatively predict the spatial distribution of high-quality gas-bearing reservoirs and obtain gas-bearing area B; (17) Using the qualitative and quantitative prediction results, favorable areas of high-quality gas-bearing reservoirs are selected to form a comprehensive seismic prediction method for high-quality gas-bearing reservoirs.

2. The method according to claim 1, wherein the earthquake in step (1) is an artificially induced earthquake for the purpose of exploration; When artificial seismic data is collected in the field using a seismograph, three-dimensional data is collected on the ground, and the seismic data obtained is three-dimensional data; The process of processing 3D seismic data includes preprocessing, field static correction, denoising, deconvolution, velocity analysis, dynamic correction, residual static correction, counter-motion correction and migration imaging steps.

3. The method according to claim 1 is characterized in that the correction processing described in step (2) is performed separately on the longitudinal wave impedance curve, the shear wave impedance curve and the density curve.

4. The method according to claim 1 is characterized in that the high-quality gas-bearing reservoir described in step (3) is a gas-bearing reservoir with high porosity, high gas saturation, high permeability and low mud content.

5. The method according to claim 1, characterized in that the high-quality reservoir identification plate in step (15) consists of a cross-plot and two histograms, and the E a 、E b Inversion data volume and logging gas reservoir classification conclusions.

Images