A low-permeability reservoir gas sweet spot prediction method based on lame impedance

Abstract

The application discloses a low-permeability reservoir gas sweet spot prediction method based on Lame impedance, which comprises the following steps: based on rock physical analysis, multiple fluid identification factors are constructed, and the best sweet spot factor is selected; based on logging data, quantitative interpretation models between Lame impedance and porosity and between Lame impedance and gas saturation are established through regression analysis; prestack simultaneous inversion is carried out to obtain three-dimensional data bodies of longitudinal wave impedance, transverse wave impedance and density; sweet spot attribute bodies are calculated, three-dimensional prediction is carried out, and three-dimensional space quantitative description of reservoir sweet spots is completed. Through the method, the high-order high-sensitivity sweet spot factor Lame impedance is systematically selected for gas sweet spot identification and three-dimensional prediction, and combined with the high-precision prestack simultaneous inversion technology, the prediction accuracy of fluid properties and reservoir quality in the low-permeability reservoir is significantly improved.

Description

Technical Field

[0001] This invention belongs to the field of oil and gas development technology, specifically relating to a method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance. Background Technology

[0002] In oil and gas development, accurately identifying gas-bearing "sweet spots" (i.e., high-yield, high-abundance areas) in reservoirs is crucial for improving drilling success rates and development efficiency. For reservoirs with low permeability and strong heterogeneity, due to their complex pore structure and weak fluid response, traditional geophysical identification methods (such as single wave impedance inversion and Poisson's ratio analysis) often struggle to effectively distinguish between gas layers, water layers, and low-yield dry layers, resulting in highly ambiguous and inaccurate prediction results.

[0003] CN120028861A utilizes pre-stack high-resolution inversion technology under multi-dimensional information constraints to obtain elastic property volumes that accurately characterize geological and engineering sweet spots, significantly improving inversion resolution. CN115016003B employs an artificial neural network method to establish a geological attribute discrimination model for coalbed methane-rich areas, solving the technical problem of inaccurate prediction results due to the strong ambiguity of existing methods for predicting coalbed methane content using single seismic attributes or parameters. CN119828218B improves the accuracy of predicting geological sweet spots in low-permeability reservoirs by extracting multi-dimensional key low-permeability reservoir information and based on data processing, pattern recognition, and machine learning technologies.

[0004] Currently used fluid identification factors, such as the P-wave / S-wave velocity ratio (Vp / Vs) and Poisson's ratio, are zero-power terms of wave impedance, which have limited sensitivity to fluids and are not effective in distinguishing complex reservoirs. While first-power parameters such as P-wave impedance (Ip) and S-wave impedance (Is) can distinguish between reservoirs and non-reservoirs, they are insufficient in identifying fluid types.

[0005] Therefore, there is an urgent need for a higher-order and more sensitive rock physics parameter as a sweet spot factor, combined with advanced pre-stack seismic inversion technology, to achieve a detailed characterization of gas-bearing sweet spots in low-permeability reservoirs. However, no research has been reported on using this technology to predict gas-bearing sweet spots in low-permeability reservoirs. Summary of the Invention

[0006] This invention is proposed to solve the problems existing in the prior art, and its purpose is to provide a method for predicting gas sweet spots in low-permeability reservoirs based on Lamé impedance.

[0007] This invention is achieved through the following technical solution: A method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance includes the following steps: S1. Based on rock physics analysis, a variety of fluid identification factors are constructed, and the best sweet spot factor is selected. S2. Based on well logging data, quantitative interpretation models of the relationship between Lamé impedance and porosity, and between Lamé impedance and gas saturation, are established through regression analysis. S3. Perform pre-stack simultaneous inversion to obtain three-dimensional data volumes of longitudinal wave impedance, transverse wave impedance, and density. S4. Calculate the dessert attribute volume and perform three-dimensional prediction to complete the three-dimensional quantitative characterization of the reservoir dessert.

[0008] In the above technical solution, the fluid identification factor includes a zero-dimensional wave impedance factor, a first-dimensional wave impedance factor, and a second-dimensional wave impedance factor. The dimensionless factor of wave impedance includes the longitudinal wave velocity ( ) and transverse wave impedance ( The ratio of ) = / ), longitudinal wave impedance ( ) and transverse wave impedance ( The difference between the squared ratio of ( ) and the adjustment constant 2 is ( ). ) and the first Lamé coefficient ( ) and twice the first Lamé coefficient ( ) and the second Lamé coefficient ( The ratio between ) and ( ); The first-order factor of the wave impedance dimension includes longitudinal wave impedance and transverse wave impedance; The quadratic factor of the wave impedance dimension includes shear impedance and Lamé impedance; shear impedance is the square of transverse wave impedance; Lamé impedance is the difference between the square of longitudinal wave impedance and the square of transverse wave impedance.

[0009] In the above technical solution, the optimal method for selecting the sweet spot factor in step S1 is to analyze the logging data of known wells in the study area and use a combination of cross-plots and probability distribution plots to comprehensively evaluate the ability of each fluid identification factor to distinguish between gas layers, water layers, dry layers and surrounding rocks.

[0010] In the above technical solution, the optimal dessert factor is Lamé impedance.

[0011] In the above technical solution, the quantitative interpretation model of the relationship between Lamé impedance and porosity is expressed as follows: The quantitative interpretation model for the relationship between Lamé impedance and gas saturation is expressed as follows: In the above formula: All are regression coefficients. Lamé impedance, in units of (m / s)2 • (g / cc); Porosity, expressed as % This represents gas saturation, expressed in percent.

[0012] In the above technical solution, step S3 specifically includes the following steps: S31. Based on the pre-stack common reflection point gather data of the target layer, the angle is divided to generate partial superimposed data volumes of near, middle and far angles. S32. Through precise well-seismic calibration, extract the seismic wavelets corresponding to the superimposed data volumes from three angles; S33. Using the seismic interpretation horizons of the top and bottom of the target reservoir as a framework, and combining the stacked velocity spectrum information of the work area with the low-frequency components of the P-wave impedance, S-wave impedance and density curves of multiple exploration wells, a three-dimensional low-frequency model of P-wave impedance, S-wave impedance and density is constructed by spatial interpolation. S34. A constrained sparse pulse inversion algorithm based on the Knott-Zoeppritz exact equation is used to synchronously invert the superimposed data volume from three angles to obtain a high-fidelity three-dimensional data volume of longitudinal wave impedance, transverse wave impedance, and density.

[0013] In the above technical solution, the pre-stack simultaneous inversion step S34 uses Gardner's empirical formula to apply soft constraints to the density inversion term. The Gardner empirical formula is as follows: In the formula: Density, unit: g / cm³ 3 ; These are empirical coefficients or proportionality constants, and are dimensionless. The longitudinal wave velocity is expressed in m / s. It is an empirical index, dimensionless.

[0014] In the above technical solution, step S4 specifically includes the following steps: S41. Using the three-dimensional data volume of longitudinal wave impedance and the three-dimensional data volume of transverse wave impedance obtained in step S3, the Lamé impedance data volume of the entire study area is calculated. S42. Using the quantitative interpretation model established in step S2, convert the Lamé impedance data volume into a porosity data volume and a gas saturation data volume. S43. Through three-dimensional visualization technology, the spatial distribution range of low Lamé impedance, high porosity, and high gas saturation is comprehensively displayed, and finally the three-dimensional spatial quantitative characterization of the reservoir sweet spot is completed.

[0015] The beneficial effects of this invention are: This invention provides a method for predicting gas-bearing sweet spots in low-permeability reservoirs based on a specific rock physical parameter—Lame impedance. It abandons the traditional low-order fluid identification factor and innovatively adopts Lamé impedance, which is the square root of wave impedance. As a sweet spot factor, it exhibits higher sensitivity. This factor amplifies the changes in P-wave impedance caused by fluids (especially gas-bearing elements) through squared operations, while suppressing the influence of the framework. It is extremely sensitive to gas layers in low-permeability reservoirs and its ability to distinguish between gas, water, and dry layers is significantly better than that of other factors. Conventional parameters such as Poisson's ratio are used; through a systematic integrated process of "factor optimization - quantitative modeling - high-precision inversion", geological understanding, well logging information and seismic information are deeply integrated, resulting in higher prediction accuracy and lower ambiguity. Multiple elastic parameters are simultaneously acquired during pre-stack inversion, reducing the uncertainty of inversion. The final sweet spot prediction results not only include geophysical properties ( Supported by ) and further supported by geological parameters ( , The quantitative description of ) makes the prediction results more reliable.

[0016] This invention not only provides qualitative identification of sweet spots, but also realizes semi-quantitative-quantitative prediction of porosity and gas saturation through the established transformation model. It is highly practical and can directly output three-dimensional sweet spot attribute volume, clearly showing the spatial shape, scale and internal quality changes of sweet spots, providing a direct basis for well location deployment, horizontal well trajectory design and reserve calculation. Attached Figure Description

[0017] Figure 1 This is a schematic diagram comparing and analyzing the sweet spot factor of different power powers of wave impedance in this invention. Figure 2 The Lamé impedance in this invention ( ) and porosity ( ), gas saturation ( Scatter plot of quantitative relationship; Figure 3 This is a flowchart of the pre-stack simultaneous inversion technique in this invention; Figure 4 It is the longitudinal wave impedance volume obtained by simultaneous inversion before stacking in this invention ( ), transverse wave impedance body ( ) and cross-sectional view of the density volume; Figure 5 The Lamé impedance calculated based on the inversion results in this invention is ( A cross-sectional view comparing the predicted effects of Naoi Dessert with those of Naoi Dessert. Figure 6 The Lamé impedance calculated based on the inversion results in this invention is ( A cross-sectional view comparing the dessert prediction results of horizontal wells with those of horizontal wells is shown. Figure 7 This is a predicted distribution map of gaseous desserts based on Lamé impedance in this invention.

[0018] For those skilled in the art, other related figures can be obtained from the above figures without any creative effort. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solution of the present invention, the technical solution of the present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0020] like Figure 1 As shown, A method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance is characterized by comprising the following steps: S1. Constructing and Optimizing Sweet Spot Factors: Based on rock physics analysis, multiple fluid identification factors with different power-law wave impedance dimensions are constructed. Through intersection analysis and probability distribution analysis, their ability to distinguish gas layers, water layers, dry layers, and mudstone in the target reservoir is compared, and the Lamé impedance characterized by the square of the wave impedance dimension is selected. () as the best dessert factor; The fluid identification factor includes multiple power factors with different wave impedance dimensions. The fluid identification factor includes a zero-power wave impedance factor, a first-power wave impedance factor, and a second-power wave impedance factor; the zero-power wave impedance factor is the zeroth power of the wave impedance dimension, the first-power wave impedance factor is the first power of the wave impedance dimension, and the second-power wave impedance factor is the square of the wave impedance dimension. The dimensionless factor of wave impedance includes the longitudinal wave velocity ( ) and transverse wave impedance ( The ratio of ) = / ), longitudinal wave impedance ( ) and transverse wave impedance ( The difference between the squared ratio of ( ) and the adjustment constant 2 is ( ). ) and the first Lamé coefficient ( ) and twice the first Lamé coefficient ( ) and the second Lamé coefficient ( The ratio between ) and ( ); The first-order factor of the wave impedance dimension includes longitudinal wave impedance ( ) and transverse wave impedance ( Longitudinal wave impedance (); ) is the longitudinal wave velocity ( ) and density ( The product of ), i.e. Transverse wave impedance ( ) is the transverse wave velocity ( ) and density ( The product of ), i.e. ; The quadratic factor of the wave impedance dimension includes shear impedance ( ) and Lamé impedance ( ); shear resistance ( ) is the square of the transverse wave impedance ( Lamé impedance (); ) is the difference between the square of the longitudinal wave impedance and the square of the transverse wave impedance. ); The criteria for the optimal sweet spot factor are that it can effectively distinguish between gas layers, water layers, dry layers and non-reservoir mudstone at the same time, and has high sensitivity to fluids, that is, the factor has the smallest distribution overlap area and the largest mean difference in the four types of samples: gas, water, dry and mud. Lamé resistance as the best dessert factor ( It combines the sensitivity of longitudinal wave impedance to fluid and the stability of transverse wave impedance to the skeleton, exhibiting the highest comprehensive distinction between fluid and lithology.

[0021] S2. Establish a quantitative interpretation model: Based on well logging data, establish Lamé impedance through regression analysis (…). ) and porosity ( ) and Lamé impedance ( ) and gas saturation ( Quantitative explanation models between ) The regression analysis includes linear regression analysis or nonlinear regression analysis; Specifically, this involves using cored wells or well-interpreted logging data from multiple systems within the study area to extract Lamé impedance data for the target formation. ) values, measured or interpreted porosity and gas saturation values; through mathematical regression analysis, establish Lamé impedance ( ) and porosity ( ) and Lamé impedance ( ) and gas saturation ( A quantitative interpretation model between geophysical parameters and reservoir geological parameters is used to achieve the conversion from geophysical parameters to reservoir geological parameters.

[0022] The Lamé impedance ( ) and porosity ( The quantitative explanation model between them can be expressed as: The Lamé impedance ( ) and gas saturation ( The quantitative explanation model between them can be expressed as: In the above formula: All are regression coefficients. Lamé impedance, in units of (m / s) 2 • (g / cc); Porosity, expressed as % This represents gas saturation, expressed in percent.

[0023] S3. Perform pre-stack simultaneous inversion to obtain three-dimensional data volumes of longitudinal wave impedance, transverse wave impedance, and density. like Figure 3 As shown, the specific steps include: S31. Based on the pre-stack common reflection point (CRP) gather data of the target layer, perform angle division to generate partial stacked data volumes of near, middle and far angles; S32. Through precise well-seismic calibration, extract the seismic wavelets corresponding to the superimposed data volumes from three angles; S33. Using the seismic interpretation horizons of the top and bottom of the target reservoir as a framework, and combining the stacked velocity spectrum information of the work area with the low-frequency components of the P-wave impedance, S-wave impedance and density curves of multiple exploration wells, a three-dimensional low-frequency model of P-wave impedance, S-wave impedance and density is constructed by spatial interpolation. S34. Using a constrained sparse pulse inversion algorithm based on the Knott-Zoeppritz exact equation, the superimposed data volume from three angles is simultaneously inverted before stacking to obtain a high-fidelity three-dimensional data volume of longitudinal wave impedance, transverse wave impedance, and density. The pre-stack simultaneous inversion uses Gardner's empirical formula to apply soft constraints to the density inversion term in order to improve the stability of the density inversion. The Gardner empirical formula is as follows: In the formula: Density, unit: g / cm³ 3 ; These are empirical coefficients or proportionality constants, and are dimensionless. The longitudinal wave velocity is expressed in m / s. It is an empirical index, dimensionless.

[0024] S4. Calculate the dessert attribute volume and perform 3D prediction: Specifically, the following steps are included: S41. The longitudinal wave impedance obtained from step S3 is inverted ( ) Three-dimensional data volume and transverse wave impedance ( Substituting the three-dimensional data volume into the Lamé impedance definition determined in step S1, the Lamé impedance of the entire study area is calculated. Data body; S42. Using the quantitative interpretation model established in step S2, Lamé impedance ( Data volume converted to porosity ( Data volume and gas saturation ( Data body; S43. Through three-dimensional visualization technology, the spatial distribution range of low Lamé impedance (indicating sweet spots), high porosity, and high gas saturation is comprehensively displayed, and the three-dimensional spatial quantitative characterization (quantitative prediction and delineation) of reservoir sweet spots is finally completed.

[0025] To ensure the accuracy of the quantitative characterization of reservoir sweet spots, the results of the quantitative characterization of reservoir sweet spots need to be compared with the sweet spot range predicted by the Lamé impedance data volume using the drilling results of verification wells or horizontal wells that were not involved in the inversion modeling, in order to verify the accuracy of the prediction results.

[0026] Example 1 The reservoir properties and gas-water distribution patterns of the "Huangyi Section" low-permeability sandstone gas reservoir in the DF-A area of ​​the Yinggehai Basin in the South China Sea are complex, making it difficult to predict favorable gas-bearing sweet spots.

[0027] Taking this area as an example, a method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance is characterized by the following steps: S1. Construct and optimize dessert factors: Specifically, the following steps are included: S11. Starting from rock physics, analyze the characteristics of P-wave velocity / impedance being sensitive to fluid and S-wave velocity / impedance being sensitive to the skeleton in the study area, and derive and construct candidate sweet spot factors with different dimensions: The D gas field is developed by gravity flow sedimentation, with lithology mainly consisting of fine sandstone. The reservoir clay content is mostly greater than 15%, which is a typical fine-grained, high-clay, low-permeability reservoir. The reservoir properties in the study area are controlled by the grain size of the sedimentary rocks. The coarser the grain size, the better the properties.

[0028] The fluid identification factor includes a zero-power wave impedance factor, a first-power wave impedance factor, and a second-power wave impedance factor; the zero-power wave impedance factor is the zeroth power of the wave impedance dimension, the first-power wave impedance factor is the first power of the wave impedance dimension, and the second-power wave impedance factor is the square of the wave impedance dimension. The dimensionless factor of wave impedance includes the longitudinal wave velocity ( ) and transverse wave impedance ( The ratio of ) ), longitudinal wave impedance ( ) and transverse wave impedance ( The difference between the squared ratio of ( ) and the adjustment constant 2 is ( ). ) and the first Lamé coefficient ( ) and the second Lamé coefficient ( The ratio between ) / ); The first-order factor of the wave impedance dimension includes longitudinal wave impedance ( ) and transverse wave impedance ( Longitudinal wave impedance (); ) is the longitudinal wave velocity ( ) and density ( The product of ), i.e. Transverse wave impedance ( ) is the transverse wave velocity ( ) and density ( The product of ), i.e. ; The quadratic factor of the wave impedance dimension includes shear impedance ( ) and Lamé impedance ( ); shear resistance ( ) is the square of the transverse wave impedance ( Lamé impedance (); ) is the difference between the square of the longitudinal wave impedance and the square of the transverse wave impedance. ); S12. Conduct sensitivity analysis of candidate sweet spot factors, selecting 11 wells (including gas, water, and dry layers) with known fluid properties in the eastern and western areas of the study area, and extracting the well logging calculation values ​​of each factor in the target interval; plotting the relationship between each factor and P-wave impedance (P-wave impedance). The intersection graph of ) and their respective probability distribution histograms (e.g. Figure 1 The optimal results show that... , The numerical ranges overlap significantly across air, water, and dry layers, making effective differentiation impossible. (Longitudinal wave impedance...) ) and transverse wave impedance ( It can distinguish between reservoirs and non-reservoirs relatively well, but it has difficulty distinguishing fluid types; / There is some distinction between the dry layer and the gas-water layer, but the distinction is still not obvious. Ultimately, the Lamé impedance ( It exhibits the best distinguishing ability, with its value being significantly lower in gas layers, clearly distinguishing it from water layers, dry layers, and high-value mudstone, and showing a stable threshold value (such as gas layers). ); S2. Establish a quantitative interpretation model: Using seven wells with reliable porosity and gas saturation interpretation results as modeling wells, the Lamé impedance calculated from the logging of the target formation was read. ) value and corresponding porosity ( ), gas saturation ( The values ​​are used to perform linear regression analysis to obtain a quantitative explanatory model (such as...). Figure 2 ): The Lamé impedance ( ) and porosity ( The quantitative explanation model between them can be expressed as: The Lamé impedance ( ) and gas saturation ( The quantitative explanation model between them can be expressed as: These two models are for Lamé impedance ( The conversion of the data volume into a geological parameter data volume provides a mathematical basis; S3. Perform pre-stack simultaneous inversion to obtain three-dimensional data volumes of longitudinal wave impedance, transverse wave impedance, and density. Specifically, the following steps are included: S31. Prepare the seismic data using 101 pre-stack CRP gathers from the study area. For the target segment, divide the offset into three parts: near-field (200-1000m), middle-field (1000-1800m), and far-field (1800-2600m), generating three parts of the stacked data volume; S32. Perform wavelet extraction and calibration. Taking DF-10 well as an example, perform fine well-seismic calibration to ensure that the correlation between the synthetic record and the actual seismic trace reaches more than 80%, and extract the wavelets of the near, middle and far stacked data volumes respectively. S33. Construct a low-frequency model, establishing a structural framework with the T30 and T31 seismic interpretation horizons as the top and bottom boundaries, and including the data from 11 wells. , , The low-frequency trend of the curve (usually <8Hz) is loaded into the framework, and a three-dimensional low-frequency model is established using the Kriging interpolation algorithm; S34. Perform pre-stack simultaneous inversion using commercial inversion software (such as Jason, Hampson, Russell). Input near-, mid-, and far-stacked data volumes, corresponding wavelet models, and low-frequency models. Select the Knott-Zoeppritz equation as the AVO forward model, enable Gardner constraints, and set appropriate sparse pulse weighting coefficients λ. Run the inversion (see inversion procedure). Figure 3 ), ultimately achieving high resolution across the entire work area. , , Three-dimensional data volume ( Figure 4 ).

[0029] S4. Calculate the dessert attribute volume and perform 3D prediction: Specifically, the following steps are included: S41. Using Lamé's impedance definition The inverted longitudinal wave impedance ( Data body and transverse wave impedance ( The data volume is processed one channel at a time to generate Lamé impedance (LAI). Data body; S42. Using the quantitative interpretation model established in step S2, Lamé impedance ( Data volume converted to porosity ( Data volume and gas saturation ( Data body; S43. Verification was performed using actual wells and inversion profiles.

[0030] like Figure 5 In the Lamé impedance profile of wells DF-4 to DF-10 shown, the gas layer encountered by well DF-4 (red section) perfectly matches the low λρ region; the gas layer in well DF13-1-10 also corresponds to a low λρ region. The value is higher, while the upper dry layer corresponds to a higher value. The predicted values ​​are highly consistent with the actual drilling results. The DFP-7 horizontal well encountered a 403-meter gas layer in the horizontal section, and its trajectory is shown in the λρ space profile (…). Figure 6 The data consistently traverses the low λρ (sweet spot) region, further confirming the accuracy of the prediction. The final output includes maps such as the Lamé impedance (λρ) plane distribution map, porosity plane map, and gas saturation plane map. Figure 7 (and data results). These results are directly used to support the development of development plans for the study area.

[0031] In summary, this invention, through the optimization and application of Lamé impedance, a highly sensitive sweet spot factor, combined with advanced pre-stack simultaneous inversion technology, forms a complete sweet spot prediction solution for low-permeability reservoirs. Its effectiveness has been verified through practical application in actual work areas.

[0032] Existing technologies mostly use multi-information fusion to predict reservoirs, and pre-stack seismic inversion often uses a single elastic parameter directly as a sweet spot indicator. This invention quantitatively compares the distinguishing ability of different factors, selects the rock physical parameter Lamé impedance as the sweet spot factor, introduces well logging constraints, and combines advanced pre-stack seismic inversion technology with high-precision P-wave and S-wave impedance volumes to achieve a fine characterization of gas-bearing sweet spots in low-permeability reservoirs.

[0033] This invention, through systematic comparison and sensitivity analysis of various wave impedance dimensions (zero, first, and second power sweet spot factors), selects Lamé impedance as the optimal identification factor; establishes a quantitative conversion model between Lamé impedance and reservoir properties and gas content; performs multi-angle simultaneous inversion based on pre-stack seismic gather data to obtain high-precision P-wave and S-wave impedance volumes; and then calculates the Lamé impedance data volume, ultimately achieving effective identification of gas layers, water layers, dry layers, and non-reservoir layers in low-permeability reservoirs and quantitative three-dimensional characterization of sweet spots. This significantly improves the accuracy of reservoir fluid identification and the reliability of sweet spot prediction, providing key technical support for the efficient development of low-permeability gas reservoirs.

[0034] The applicant declares that the above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention fall within the protection and disclosure scope of the present invention.

Claims

1. A method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance, characterized in that: Includes the following steps: S1. Based on rock physics analysis, a variety of fluid identification factors are constructed, and the best sweet spot factor is selected. S2. Based on well logging data, quantitative interpretation models of the relationship between Lamé impedance and porosity, and between Lamé impedance and gas saturation, are established through regression analysis. S3. Perform pre-stack simultaneous inversion to obtain three-dimensional data volumes of longitudinal wave impedance, transverse wave impedance, and density. S4. Calculate the dessert attribute volume and perform three-dimensional prediction to complete the three-dimensional quantitative characterization of the reservoir dessert.

2. The method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance according to claim 1, characterized in that: The fluid identification factor includes a zero-dimensional wave impedance factor, a first-dimensional wave impedance factor, and a second-dimensional wave impedance factor. The zero-power wave impedance factor includes the ratio of longitudinal wave velocity to transverse wave impedance, the difference between the square of the ratio of longitudinal wave impedance to transverse wave impedance and the adjustment constant 2, and the ratio between the first Lamé coefficient and twice the first Lamé coefficient and the second Lamé coefficient. The first-order factor of the wave impedance dimension includes longitudinal wave impedance and transverse wave impedance; The quadratic factor of the wave impedance dimension includes shear impedance and Lamé impedance; shear impedance is the square of transverse wave impedance; Lamé impedance is the difference between the square of longitudinal wave impedance and the square of transverse wave impedance.

3. The method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance according to claim 1, characterized in that: The optimal method for selecting the sweet spot factor in step S1 is to analyze the logging data of known wells in the study area and use a combination of cross-plots and probability distribution plots to comprehensively evaluate the ability of each fluid identification factor to distinguish between gas layers, water layers, dry layers and surrounding rocks.

4. The method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance according to claim 1, characterized in that: The optimal dessert factor is Lamé resistance.

5. The method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance according to claim 1, characterized in that: The quantitative interpretation model of the relationship between Lamé impedance and porosity is expressed as follows: The quantitative interpretation model for the relationship between Lamé impedance and gas saturation is expressed as follows: In the above formula: All are regression coefficients. Lamé impedance, in units of (m / s) 2 • (g / cc); Porosity is a dimensionless quantity. The gas saturation is dimensionless.

6. The method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance according to claim 1, characterized in that: Step S3 specifically includes the following steps: S31. Based on the pre-stack common reflection point gather data of the target layer, the angle is divided to generate partial superimposed data volumes of near, middle and far angles. S32. Through precise well-seismic calibration, extract the seismic wavelets corresponding to the superimposed data volumes from three angles; S33. Using the seismic interpretation horizons of the top and bottom of the target reservoir as a framework, and combining the stacked velocity spectrum information of the work area with the low-frequency components of the P-wave impedance, S-wave impedance and density curves of multiple exploration wells, a three-dimensional low-frequency model of P-wave impedance, S-wave impedance and density is constructed by spatial interpolation. S34. A constrained sparse pulse inversion algorithm based on the Knott-Zoeppritz exact equation is used to synchronously invert the superimposed data volume from three angles to obtain a high-fidelity three-dimensional data volume of longitudinal wave impedance, transverse wave impedance, and density.

7. The method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance according to claim 6, characterized in that: The pre-stack simultaneous inversion step S34 uses Gardner's empirical formula to apply soft constraints to the density inversion term. The Gardner empirical formula is as follows: In the formula: Density, unit: g / cm³ 3 ; These are empirical coefficients or proportionality constants, and are dimensionless. The longitudinal wave velocity is expressed in m / s. It is an empirical index, dimensionless.

8. The method for predicting gas-bearing sweet spots in low-permeability reservoirs based on Lamé impedance according to claim 1, characterized in that: Step S4 specifically includes the following steps: S41. Using the three-dimensional data volume of longitudinal wave impedance and the three-dimensional data volume of transverse wave impedance obtained in step S3, the Lamé impedance data volume of the entire study area is calculated. S42. Using the quantitative interpretation model established in step S2, convert the Lamé impedance data volume into a porosity data volume and a gas saturation data volume. S43. Through three-dimensional visualization technology, the spatial distribution range of low Lamé impedance, high porosity, and high gas saturation is comprehensively displayed, and finally the three-dimensional spatial quantitative characterization of the reservoir sweet spot is completed.

Images