High-resolution seismic inversion identification method for thin and small reservoir of oil reservoir

Abstract

The invention relates to the technical field of oil and gas geophysical exploration, in particular to a high-resolution seismic inversion recognition method for a thin and small reservoir of an oil reservoir. The method comprises the following steps: performing digital core model construction on a thin and small reservoir core sample to obtain a three-dimensional digital core model; carrying out multi-mineral component quantitative analysis on the thin and small reservoir core sample to obtain mineral components and distribution data; performing fracture geometric analysis on the three-dimensional digital core model to obtain fracture geometric analysis data; constructing a microscopic rock physical model according to the fracture geometric analysis data and the mineral composition and distribution data, and calculating physical model parameters to obtain microscopic rock physical parameters; and calculating matrix pore elastic parameters according to the microscopic rock physical parameters to obtain pore-containing dry rock elastic parameters. According to the method, the problems that a traditional method is low in resolution and influences of reservoir fluid and fractures are ignored are effectively solved, and high-precision recognition and fine description of the thin and small reservoir are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas geophysical exploration, and particularly to a high-resolution seismic inversion identification method for thin and small oil reservoir layers. Background Art

[0002] Seismic inversion can provide physical property information of reservoirs, such as porosity, fluid saturation, fracture properties, etc., by relating seismic data to elastic parameters of underground media. However, with the advancement of oil and gas exploration towards complex geological conditions, especially the exploration of thin and small reservoirs, traditional seismic inversion methods are facing increasingly significant challenges.

[0003] Thin and small reservoirs refer to oil and gas reservoirs with small thickness and limited spatial scale. These reservoirs usually have relatively complex geological structures, such as developed fractures, complex multi-mineral compositions, and various fluid combinations. Due to the fact that the seismic wavelength is much larger than the reservoir thickness, thin and small reservoirs often appear relatively blurred in conventional seismic data and are difficult to clearly identify. The demand for high-resolution seismic inversion is increasing continuously. Traditional seismic inversion methods rely on layered geological models and have limited resolution for reservoirs, making it difficult to fully reflect the fine structure and physical property changes of thin and small reservoirs. In this case, how to accurately identify and characterize thin and small reservoirs through higher-resolution seismic inversion methods has become an important issue in oil and gas exploration.

[0004] 1. Low resolution of traditional methods, difficult to identify thin and small reservoirs: Traditional seismic inversion methods are difficult to effectively distinguish reservoirs with small thickness under seismic data with long wavelengths. The thickness of thin and small reservoirs is usually much smaller than the seismic wavelength, resulting in insufficient resolution obtained by seismic inversion. The inversion results often cannot clearly reflect the true structural information of the reservoirs. This insufficient resolution not only affects the accurate prediction of reservoir physical properties but also leads to misjudgment or missed judgment of reservoirs.

[0005] 2. Traditional methods ignore the influence of reservoir fluids and fractures: Complexity of fractures and pores: Fractures and pores in thin and small reservoirs are often very complexly developed. The presence of fractures will significantly change the elastic properties of the reservoirs. Traditional methods usually regard the reservoir as a homogeneous medium, ignoring the anisotropic influence of fractures and being unable to accurately depict the influence of fractures on seismic wave propagation, resulting in inaccurate inverted elastic parameters and reduced reservoir identification accuracy.

[0006] Influence of fluids is simplified: Traditional methods often adopt simple rock physics models, ignoring the complex influence of different types of fluids (such as oil, water, gas) on rock elastic parameters, especially the coupling effect between fluids and rock pore structures under different saturations. This simplified treatment leads to inaccurate identification of reservoir fluids and affects the overall inversion effect of the reservoirs. Summary of the Invention

[0007] Based on this, it is necessary to provide a high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs to solve at least one of the above technical problems.

[0008] To achieve the above object, a high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs includes the following steps: Step S1: Construct a digital core model for the thin and small reservoir core samples to obtain a three-dimensional digital core model; perform quantitative analysis of multi-mineral components on the thin and small reservoir core samples to obtain mineral composition and distribution data; perform fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; construct a microscopic rock physics model based on the fracture geometry analysis data and the mineral composition and distribution data, and perform physical model parameter calculation to obtain microscopic rock physics parameters; Step S2: Calculate matrix pore-elastic parameters based on the microscopic rock physics parameters to obtain pore-containing dry rock elastic parameters; calculate fracture elastic parameters based on the pore-containing dry rock elastic parameters and the microscopic rock physics parameters to obtain fracture-containing dry rock elastic parameters; calculate fluid-saturated rock elastic parameters based on the fracture-containing dry rock elastic parameters, and perform composite elastic parameter output to obtain composite elastic parameters; Step S3: Determine the fluid type of the thin and small reservoir and determine the property parameters to obtain reservoir fluid parameters; perform fluid substitution based on the Gassmann equation according to the composite elastic parameters and the reservoir fluid parameters to obtain rock elastic parameters with different saturations; perform fluid-coupled elastic parameter output based on the rock elastic parameters with different saturations to obtain fluid-coupled elastic parameters; Step S4: Obtain actual seismic data; perform seismic wavefield simulation according to the microscopic rock physics parameters and the fluid-coupled elastic parameters to obtain simulated seismic wavefield data; perform convolution calculation according to the actual seismic data and the simulated seismic wavefield data, and perform high-resolution seismic response output to obtain high-resolution seismic response data; Step S5: Use the high-resolution seismic response data to train a non-linear inversion model to obtain a non-linear inversion model; use the non-linear inversion model and the actual seismic data to perform reservoir parameter inversion and perform inversion result display processing to obtain the reservoir physical property inversion result of the thin and small reservoir.

[0009] By constructing a digital core model, analyzing mineral compositions and fracture geometric features, and combining this information to build a microscopic rock physics model, the present invention can obtain more accurate microscopic rock physics parameters, providing more reliable basic data for subsequent rock physics analysis and seismic inversion. By gradually calculating the elastic parameters of dry rocks with pores, dry rocks with fractures, and initially fluid-saturated rocks, and finally obtaining the composite elastic parameters, the influence of the mineral matrix, pores, fractures, and fluids on the elastic properties of rocks can be comprehensively considered, thereby more accurately describing the elastic characteristics of the reservoir. By determining reservoir fluid parameters and using the Gassmann equation for fluid substitution, the elastic parameters of rocks at different saturations can be simulated, and the influence of saturation changes on the elastic parameters can be analyzed. Finally, the fluid-coupled elastic parameters can be obtained, providing more accurate rock physics parameters for seismic wavefield simulation. By constructing a detailed geological model, performing multi-scale finite-difference forward modeling, extracting seismic wavelets, and performing convolution calculations, high-resolution seismic response data can be obtained. This data can better reflect the characteristics of thin and small reservoirs, providing high-quality training data for subsequent non-linear inversion. By using the high-resolution seismic response data to train a non-linear inversion model and preprocessing and inverting the actual seismic data, more accurate prediction results of the physical properties of thin and small reservoirs can be obtained. Through post-processing and visualization, the fine characterization and effective identification of thin and small reservoirs can be finally achieved. Therefore, the present invention provides a high-resolution seismic inversion and identification method for thin and small reservoirs in oil reservoirs. By combining techniques such as digital core technology, microscopic rock physics modeling, fluid substitution, seismic wavefield simulation, and non-linear inversion, the problems of low resolution and neglect of the influence of reservoir fluids and fractures in traditional methods are effectively solved, and the high-precision identification and fine characterization of thin and small reservoirs are realized.

[0010] Preferably, step S1 includes the following steps: Step S11: Collect core image data at different scales of the thin and small reservoir core samples through high-resolution CT scanning technology to obtain multi-scale core image data; Step S12: Perform image segmentation processing on the multi-scale core image data to obtain an image segmentation result; construct a digital core model based on the image segmentation result to obtain a three-dimensional digital core model; Step S13: Characterize the pore results and fracture network based on the three-dimensional digital core model to obtain pore-fracture network parameters; Step S14: Conduct mineralogical analysis on the thin and small reservoir core samples to obtain a mineralogical analysis result; perform quantitative analysis of the content and spatial distribution of different mineral components based on the three-dimensional digital core model and the mineralogical analysis result to obtain mineral composition and distribution data; Step S15: Conduct fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; construct a microscopic rock physics model based on the fracture geometry analysis data, pore-fracture network parameters, and mineral composition and distribution data, and calculate physical model parameters to obtain microscopic rock physics parameters.

[0011] By acquiring multi-scale core image data, the present invention can capture the microscopic structure information of core samples more comprehensively, covering pore, fracture, and mineral distribution characteristics at different scales from the micron level to the centimeter level, providing a high-precision data basis for the subsequent construction of digital core models, thereby improving the reliability of inversion results. Through image segmentation processing, different mineral components, pores, and fractures are distinguished from the core images, providing basic data for constructing a three-dimensional digital core model. Accurate image segmentation results can ensure the accuracy of the digital core model, and further improve the accuracy of subsequent rock physics analysis and seismic inversion. Pore results and fracture network characterization can quantitatively describe the pore structure and fracture network characteristics of core samples, such as parameters like porosity, permeability, fracture density, connectivity, etc. These parameters are key input parameters for constructing a microscopic rock physics model, helping to more accurately describe the elastic properties and seepage characteristics of rocks. The acquisition of mineral composition and distribution data can accurately determine the content and spatial distribution of different mineral components in the core, providing important basic data for the subsequent construction of rock physics models and helping to understand the influence of mineral composition on the elastic properties of rocks. Through fracture geometry analysis and microscopic rock physics model construction, more accurate microscopic rock physics parameters can be obtained, such as mineral modulus, fracture compliance, etc. These parameters can more accurately reflect the rock physics characteristics of thin and small reservoirs, providing more reliable input data for subsequent seismic wave field simulation and reservoir parameter inversion, and ultimately improving the inversion accuracy.

[0012] Preferably, step S15 includes the following steps: Step S151: Calculate porosity and fracture density based on the three-dimensional digital core model to obtain porosity and fracture density data; Step S152: Obtain single-mineral elastic parameter data; conduct a literature search for single-mineral elastic parameters based on the mineral composition and distribution data to obtain single-mineral elastic parameters; calculate mineral modulus using the single-mineral elastic parameters to obtain mineral modulus data; Step S153: Conduct fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; Step S154: Calculate anisotropic parameters based on the fracture geometry analysis data to obtain anisotropic parameter data; Step S155: Construct a microscopic rock physics model based on the porosity and fracture density data, mineral modulus data, fracture geometry analysis data, and anisotropic parameter data to obtain a microscopic rock physics model; Step S156: Calculate micro rock physical parameters using a micro rock physical model to obtain micro rock physical parameters.

[0013] Through accurate calculation of porosity and fracture density, the present invention can provide key structural information for the subsequent construction of a micro rock physical model. Porosity and fracture density directly affect the elastic properties and permeability of rocks, and accurate calculation results help improve the accuracy of the rock physical model. Obtaining single mineral elastic parameters and calculating mineral moduli provide basic data for the construction of the rock physical model. The elastic properties of different minerals vary greatly, and accurate mineral modulus data can more realistically reflect the elastic characteristics of the rock matrix, thereby improving the prediction accuracy of the rock physical model. Conducting fracture geometry analysis on the three-dimensional digital core model can obtain key parameters such as the length, aperture, and direction of fractures. These parameters can more comprehensively describe the geometric characteristics of fractures, providing a basis for the subsequent establishment of fracture models and the calculation of anisotropic parameters. Calculating anisotropic parameters based on fracture geometry analysis data can quantify the anisotropy of the elastic properties of rocks caused by fractures. This is crucial for understanding the influence of fractures on seismic wave propagation and helps construct a more accurate rock physical model. Comprehensively using porosity and fracture density data, mineral modulus data, fracture geometry analysis data, and anisotropic parameter data to construct a micro rock physical model can more comprehensively consider the influence of various factors on the elastic properties of rocks, thereby constructing a more accurate rock physical model and improving the prediction ability of the model. Calculating micro rock physical parameters using the constructed micro rock physical model can obtain more accurate rock elastic parameters, such as dry rock modulus, pore modulus, and fracture compliance. These parameters are key input parameters for subsequent fluid substitution and seismic wave field simulation, and their accuracy directly affects the reliability of the final inversion results.

[0014] Preferably, step S153 is specifically as follows: Extract a three-dimensional fracture network from the three-dimensional digital core model to obtain a three-dimensional fracture network model; perform skeletonization on the three-dimensional fracture network model to obtain a fracture skeleton model; Segment the fractures of the fracture skeleton model to obtain a segmented fracture model; Calculate the geometric parameters of single fractures for the segmented fracture model to obtain a list of single fracture geometric parameters; Perform stress field simulation on the three-dimensional digital core model to obtain stress field distribution data; based on the stress field distribution data, perform fracture deformation calculation on the segmented fracture model to obtain fracture deformation data; According to the list of single fracture geometric parameters and the fracture deformation data, calculate the dynamic geometric parameters of each fracture segment under different stress states to obtain dynamic fracture geometric parameters; Perform data statistical analysis on the list of dynamic fracture geometric parameters to obtain fracture geometric analysis data.

[0015] By extracting the three-dimensional fracture network model, the present invention can completely retain the spatial morphology and connectivity information of fractures, which is more accurate than traditional two-dimensional fracture analysis methods and can better reflect the actual formation conditions, providing a more reliable basis for subsequent fracture parameter analysis. The skeletonization process can simplify the geometric morphology of the fracture network, remove redundant information such as fracture thickness, facilitate subsequent fracture segmentation and geometric parameter calculation, improve the calculation efficiency, and reduce the data storage volume. Dividing the fracture skeleton model into multiple independent fracture segments enables individual geometric parameter calculation and deformation analysis for each segment, thereby more precisely characterizing the geometric features of fractures and considering the interaction between fractures. Calculating geometric parameters such as the length, width, aperture, dip angle, and azimuth angle of each fracture segment can more detailedly describe the geometric features of fractures, providing more accurate input data for subsequent fracture deformation calculation and rock physics modeling. Through stress field simulation, the stress distribution inside the core can be obtained, which helps to understand the deformation of fractures under different stress states and provides the necessary stress conditions for fracture deformation calculation. Considering the influence of the stress field on fracture deformation can more realistically simulate the dynamic changes of fractures in the underground environment, thereby more accurately predicting the impact of fractures on the elastic properties of rocks. Calculating the dynamic fracture geometric parameters can more accurately reflect the actual geometric morphology of fractures under different stress states, thereby more precisely evaluating the impact of fractures on the elastic properties of rocks and improving the accuracy of seismic inversion. By performing statistical analysis on the dynamic fracture geometric parameters, the statistical distribution characteristics of fracture geometric parameters, such as the mean value, standard deviation, etc., can be obtained, which helps to more comprehensively understand the geometric features of fractures and provides more reliable input parameters for subsequent rock physics modeling.

[0016] Preferably, step S2 includes the following steps: Step S21: Calculate the elastic parameters of the mineral matrix according to the microscopic rock physics parameters to obtain the elastic parameters of the dry rock matrix; Step S22: Calculate the matrix pore elastic parameters according to the elastic parameters of the dry rock matrix and the microscopic rock physics parameters to obtain the elastic parameters of the pore-containing dry rock; Step S23: Calculate the fracture elastic parameters according to the elastic parameters of the pore-containing dry rock and the microscopic rock physics parameters to obtain the elastic parameters of the fracture-containing dry rock; Step S24: Calculate the elastic parameters of the fluid-saturated rock according to the elastic parameters of the fracture-containing dry rock and the preset initial fluid saturation to obtain the elastic parameters of the initial fluid-saturated rock; Step S25: Output the composite elastic parameters according to the elastic parameters of the initial fluid-saturated rock to obtain the composite elastic parameters.

[0017] By calculating the elastic parameters of the mineral matrix, the equivalent elastic modulus of the mineral mixture constituting the rock skeleton can be obtained, which is the basis for subsequent calculations of the elastic parameters of porous and fractured rocks, providing a guarantee for accurately describing the elastic properties of the rock matrix. The calculation of the matrix poroelastic parameters takes into account the influence of pores on the elastic properties of the rock, making the elastic modulus of the rock closer to the actual situation and more accurate than the case where only the mineral matrix is considered, thus improving the accuracy of subsequent calculations of fracture elastic parameters. By calculating the fracture elastic parameters, the influence of fractures on the elastic properties of the rock, especially anisotropy, can be further considered. This is crucial for accurately describing the elastic characteristics of fractured reservoirs and provides a more accurate dry rock skeleton model for subsequent fluid substitution. Calculating the elastic parameters of the initially fluid-saturated rock introduces the influence of the fluid on the elastic properties of the rock, making the rock elastic parameters closer to the actual situation of the reservoir. Using the preset initial saturation value can provide a reasonable starting point for subsequent fluid substitution. Outputting the composite elastic parameters comprehensively considers the influence of the mineral matrix, pores, fractures, and initial fluid saturation on the elastic properties of the rock, providing more comprehensive and accurate rock elastic parameters, laying a solid foundation for subsequent fluid substitution and seismic wavefield simulation.

[0018] Preferably, step S23 includes the following steps: Step S231: Extract the fracture geometric parameters from the microscopic rock physical parameters to obtain the fracture geometric parameters; Step S232: Calculate the fracture compliance parameters based on the fracture geometric parameters to obtain the fracture compliance parameters; Step S233: Construct the anisotropic compliance matrix using the Schoenberg linear slip model based on the fracture compliance parameters and the fracture geometric parameters to obtain the anisotropic compliance matrix; Step S234: Calculate the compliance matrix of the fractured dry rock for the anisotropic compliance matrix and the elastic parameters of the porous dry rock to obtain the compliance matrix of the fractured dry rock; Step S235: Perform matrix inversion calculation on the compliance matrix of the fractured dry rock to obtain the elastic parameters of the fractured dry rock.

[0019] The present invention extracts fracture geometric parameters, such as fracture density, aperture, and orientation, providing necessary input data for subsequent calculation of fracture compliance parameters and construction of the anisotropic compliance matrix, ensuring the accuracy and reliability of subsequent calculations. Calculating fracture compliance parameters can quantify the influence degree of fractures on the elastic properties of rocks, providing key parameters for constructing the anisotropic compliance matrix and helping to more accurately describe the elastic characteristics of fractured reservoirs. Using the Schoenberg linear slip model to construct the anisotropic compliance matrix can effectively describe the anisotropy of rock elastic properties caused by fractures, and relate the geometric parameters and compliance parameters of fractures, providing an effective method for calculating the elastic parameters of dry fractured rocks. Combining the anisotropic compliance matrix with the elastic parameters of dry porous rocks (converted into a compliance matrix) to calculate the compliance matrix of dry fractured rocks can comprehensively consider the influence of pores and fractures on the elastic properties of rocks and obtain more accurate rock compliance parameters. Performing matrix inversion calculation on the compliance matrix of dry fractured rocks to obtain the elastic parameters of dry fractured rocks, converting the compliance matrix into elastic parameters, facilitating subsequent fluid substitution calculations, and providing more accurate rock physics parameters for seismic wave field simulation.

[0020] Preferably, step S3 includes the following steps: Step S31: Determine the fluid type of the thin and small reservoir and determine the property parameters to obtain the reservoir fluid parameters; Step S32: Perform fluid substitution based on the Gassmann equation according to the composite elastic parameters and the reservoir fluid parameters to obtain the elastic parameters of rocks with different saturations; Step S33: Analyze the influence of saturation change on the elastic parameters according to the elastic parameters of rocks with different saturations to obtain the saturation-elastic parameter relationship curve; Step S34: Output the fluid-coupled elastic parameters according to the elastic parameters of rocks with different saturations and the saturation-elastic parameter relationship curve to obtain the fluid-coupled elastic parameters.

[0021] The present invention provides necessary input data for subsequent fluid substitution based on the Gassmann equation by determining the reservoir fluid type and property parameters. Accurate fluid parameters are the key to ensuring the reliability of fluid substitution results, which is crucial for subsequent analysis of the impact of saturation changes on elastic parameters. Using the Gassmann equation for fluid substitution can calculate the elastic parameters of rocks at different saturations, thereby quantifying the influence of fluid saturation on the elastic properties of rocks. This is crucial for understanding the influence of reservoir fluids on seismic wave propagation and also provides a basis for subsequent inversion of reservoir saturation. Analyzing the impact of saturation changes on elastic parameters and plotting the saturation-elastic parameter relationship curve can intuitively show the variation law of rock elastic parameters with fluid saturation. This helps to deeply understand the relationship between reservoir fluids and rock elastic properties and provides a basis for the output of subsequent fluid-coupled elastic parameters. The output of fluid-coupled elastic parameters comprehensively considers the elastic parameters of rocks at different saturations and the relationship between saturation and elastic parameters, provides more accurate rock physics parameters for subsequent seismic wave field simulation, and improves the accuracy and reliability of reservoir parameter inversion.

[0022] Preferably, step S32 includes the following steps: Step S321: Back-calculate the bulk modulus of the dry rock matrix based on the composite elastic parameters to obtain the bulk modulus of the dry rock matrix; Step S322: Determine the bulk modulus of the fluid mixture at different saturations based on the reservoir fluid parameters and the preset initial fluid saturation to obtain the bulk modulus of the fluid mixture at different saturations; Step S323: Substitute the bulk modulus of the dry rock matrix, the composite elastic parameters, and the bulk modulus of the fluid mixture at different saturations into the Gassmann equation to calculate the bulk modulus of the rock at different saturations and obtain the bulk modulus of the rock at different saturations; Step S324: Calculate the shear modulus of the rock at different saturations for the bulk modulus of the rock at different saturations to obtain the shear modulus of the rock at different saturations; Step S325: Calculate the density of the rock at different saturations based on the reservoir fluid parameters and the microscopic rock physics parameters to obtain the density of the rock at different saturations; Step S326: Calculate the elastic parameters of the rock at different saturations based on the bulk modulus of the rock at different saturations, the shear modulus of the rock at different saturations, and the density of the rock at different saturations to obtain the elastic parameters of the rock at different saturations.

[0023] By back-calculating the bulk modulus of the dry rock matrix, the present invention can eliminate the influence of the initial fluid saturation, obtain more accurate elastic parameters of the dry rock skeleton, and provide more reliable input parameters for the subsequent application of the Gassmann equation. Determining the bulk modulus of the fluid mixture at different saturations can accurately describe the elastic properties of the pore fluid at different saturations, provide the necessary fluid parameters for the application of the Gassmann equation, and thus improve the accuracy of the calculation results. Using the Gassmann equation to calculate the bulk modulus of the rock at different saturations can effectively simulate the influence of fluid substitution on the bulk modulus of the rock, and thus quantify the influence of saturation change on the elastic properties of the rock. According to the assumptions of the Gassmann equation, the shear modulus of the rock remains unchanged during the fluid substitution process. Therefore, directly using the shear modulus in the composite elastic parameters as the shear modulus of the rock at different saturations simplifies the calculation process and maintains the consistency of the calculation results. Calculating the density of the rock at different saturations can accurately reflect the influence of fluid substitution on the density of the rock, and thus more precisely describe the physical properties of the rock at different saturations, providing more realistic parameters for the subsequent seismic wave field simulation. Organizing the bulk modulus, shear modulus, and density of the rock at different saturations into the elastic parameters of the rock at different saturations can completely describe the elastic properties of the rock at different saturations, providing the necessary data for the subsequent analysis of the influence of saturation on the elastic parameters and the construction of the saturation-elastic parameter relationship curve.

[0024] Preferably, step S4 includes the following steps: Step S41: Obtain geological data; construct a fine geological model including thin and small reservoirs and surrounding formations according to the geological data and microscopic rock physical parameters to obtain a fine geological model; Step S42: Assign the fluid-coupled elastic parameters to the corresponding positions in the fine geological model to obtain a saturation geological model; Step S43: Based on the saturation geological model, use the multi-scale finite difference forward algorithm to perform seismic wave field simulation to obtain simulated seismic wave field data; Step S44: Obtain actual seismic data; extract seismic wavelets from the actual seismic data to obtain seismic wavelet data; perform convolution calculation on the seismic wavelet data and the simulated seismic wave field data to obtain convolved seismic data; Step S45: Output high-resolution seismic responses according to the convolved seismic data to obtain high-resolution seismic response data.

[0025] By constructing a fine geological model that includes thin and small reservoirs and the surrounding formations, the present invention can accurately describe the underground geological structure and the distribution of rock physical properties, providing an accurate model basis for subsequent seismic wavefield simulation, thereby improving the reliability of the simulation results. Assigning fluid-coupled elastic parameters to the fine geological model to construct a saturation geological model can more realistically reflect the fluid distribution in the underground reservoir and its influence on elastic properties, thereby improving the accuracy and precision of seismic wavefield simulation. Using the multi-scale finite difference forward algorithm for seismic wavefield simulation can more accurately simulate the propagation process of seismic waves in complex geological models. Especially for fine structures such as thin and small reservoirs, it can better capture their seismic response characteristics. Extracting seismic wavelets from actual seismic data and convolving the wavelets with the simulated seismic wavefield data can simulate the actual situation of the seismic waves being affected by the wavelets during the underground propagation process, making the simulated seismic data closer to the actual seismic data, thereby improving the reliability of the inversion results. Outputting high-resolution seismic response data provides high-quality training data for subsequent non-linear inversion model training. Since the simulated seismic data has undergone the construction of a fine geological model and multi-scale finite difference forward simulation, the high-resolution seismic response data can better reflect the characteristics of thin and small reservoirs, thereby improving the resolution and precision of the inversion results.

[0026] Preferably, step S5 includes the following steps: Step S51: Using the high-resolution seismic response data as input features and the reservoir parameters in the saturation geological model as output labels, construct a training data set to obtain a seismic-reservoir parameter training data set; Step S52: Use the seismic-reservoir parameter training data set to train a pre-selected neural network model to obtain a non-linear inversion model; Step S53: Perform data and processing of the processing flow and data format on the actual seismic data according to the simulated seismic wavefield data to obtain preprocessed seismic data; Step S54: Use the non-linear inversion model to invert the reservoir parameters of the preprocessed seismic data to obtain the inverted and predicted reservoir parameters; Step S55: Post-process the inverted and predicted reservoir parameters and perform inversion result display processing to obtain the inversion result of the reservoir physical properties of the thin and small reservoir.

[0027] The present invention provides necessary learning samples for the training of a non - linear inversion model by constructing a seismic - reservoir parameter training dataset. Using high - resolution seismic response data as input features and reservoir parameters in the saturation geological model as output labels, a mapping relationship between seismic data and reservoir parameters is established, laying a foundation for training a high - precision inversion model. Training a pre - selected neural network model with the training dataset can obtain a non - linear inversion model that can effectively learn the non - linear relationship between seismic data and reservoir parameters. This enables the inversion model to more accurately predict reservoir parameters, especially in complex geological situations such as thin and small reservoirs. Pre - processing the actual seismic data and making its processing flow and data format consistent with the simulated seismic wavefield data can eliminate noise and other interference factors in the actual seismic data and ensure that the non - linear inversion model can be effectively applied to the actual seismic data. Using the trained non - linear inversion model to invert the reservoir parameters of the pre - processed actual seismic data can obtain high - precision prediction results for the physical properties of thin and small reservoirs. The non - linear inversion model can capture the complex non - linear relationship between seismic data and reservoir parameters, thereby improving the accuracy and reliability of the inversion results. Post - processing and visualizing the inverted and predicted reservoir parameters can further improve the stability and interpretability of the inversion results and present the inversion results in an intuitive way, facilitating interpretation and application by geologists and engineers, and ultimately achieving the fine characterization and effective identification of thin and small reservoirs. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 FIG. is a schematic flow chart of the steps of a high - resolution seismic inversion identification method for thin and small reservoirs in an oil reservoir; Figure 2 is Figure 1 a detailed implementation step flow chart of step S1 in Figure 3 is Figure 1 a detailed implementation step flow chart of step S2 in

[0029] The implementation, functional features, and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED IMPLEMENTATION MANNER

[0030] The technical method of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those skilled in the art within the scope of the present invention without creative efforts belong to the scope of protection of the present invention.

[0031] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus repeated descriptions thereof will be omitted. Some of the block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor methods and / or microcontroller methods.

[0032] It should be understood that although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the associated listed items.

[0033] To achieve the above object, please refer to Figures 1 to 3 , a high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs, comprising the following steps: Step S1: Construct a digital core model for the thin and small reservoir core sample to obtain a three-dimensional digital core model; perform quantitative analysis of multi-mineral components on the thin and small reservoir core sample to obtain mineral component and distribution data; perform fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; construct a microscopic rock physics model based on the fracture geometry analysis data and the mineral component and distribution data, and perform physical model parameter calculation to obtain microscopic rock physics parameters; Step S2: Calculate matrix pore elastic parameters based on the microscopic rock physics parameters to obtain pore-containing dry rock elastic parameters; calculate fracture elastic parameters based on the pore-containing dry rock elastic parameters and the microscopic rock physics parameters to obtain fracture-containing dry rock elastic parameters; calculate fluid-saturated rock elastic parameters based on the fracture-containing dry rock elastic parameters, and perform composite elastic parameter output to obtain composite elastic parameters; Step S3: Determine the fluid type of the thin and small reservoir and determine the property parameters to obtain reservoir fluid parameters; perform fluid substitution based on the Gassmann equation according to the composite elastic parameters and the reservoir fluid parameters to obtain rock elastic parameters with different saturations; perform fluid-coupled elastic parameter output based on the rock elastic parameters with different saturations to obtain fluid-coupled elastic parameters; Step S4: Obtain actual seismic data; perform seismic wavefield simulation based on micro-scale rock physical parameters and fluid-coupled elastic parameters to obtain simulated seismic wavefield data; perform convolution calculation based on the actual seismic data and the simulated seismic wavefield data, and perform high-resolution seismic response output to obtain high-resolution seismic response data; Step S5: Use the high-resolution seismic response data to train a non-linear inversion model to obtain a non-linear inversion model; use the non-linear inversion model and the actual seismic data to perform reservoir parameter inversion and perform inversion result display processing to obtain the reservoir physical property inversion result of the thin and small reservoir.

[0034] In the embodiment of the present invention, with reference to Figure 1 As described above, it is a schematic diagram of the step flow of the high-resolution seismic inversion identification method for thin and small reservoirs in the oil reservoir of the present invention. In this example, the high-resolution seismic inversion identification method for thin and small reservoirs in the oil reservoir includes the following steps: Step S1: Construct a digital core model for the thin and small reservoir core sample to obtain a three-dimensional digital core model; perform quantitative analysis of multi-mineral components on the thin and small reservoir core sample to obtain mineral composition and distribution data; perform fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; construct a micro-scale rock physical model based on the fracture geometry analysis data and the mineral composition and distribution data, and perform physical model parameter calculation to obtain micro-scale rock physical parameters; In the embodiment of the present invention, first, use high-resolution CT scanning technology to obtain multi-scale image data of the thin and small reservoir core sample, and then perform image segmentation and three-dimensional reconstruction to construct a three-dimensional digital core model. Next, use techniques such as X-ray diffraction (XRD) and scanning electron microscopy (SEM) to perform mineralogical analysis on the core sample, and combine with the digital core model for quantitative analysis to obtain mineral composition and distribution data. At the same time, perform fracture network extraction and geometry analysis on the three-dimensional digital core model, calculate parameters such as fracture density, aperture, and direction to obtain fracture geometry analysis data. Finally, based on the mineral composition and distribution data and the fracture geometry analysis data, select a suitable rock physical model (such as the DEM model) to calculate parameters such as mineral modulus and fracture compliance to obtain micro-scale rock physical parameters.

[0035] Step S2: Calculate matrix pore-elastic parameters based on the micro-scale rock physical parameters to obtain pore-containing dry rock elastic parameters; calculate fracture elastic parameters based on the pore-containing dry rock elastic parameters and the micro-scale rock physical parameters to obtain fracture-containing dry rock elastic parameters; calculate fluid-saturated rock elastic parameters based on the fracture-containing dry rock elastic parameters, and perform composite elastic parameter output to obtain composite elastic parameters; In the embodiments of the present invention, based on the microscopic rock physical parameters obtained in step S1, such as mineral modulus and porosity, the elastic parameters (bulk modulus and shear modulus) of the porous dry rock are calculated using the Kuster-Toksöz model or other rock physical models. Then, in combination with the fracture geometric parameters (such as fracture density and direction), the elastic parameters of the fractured dry rock are calculated using the Schoenberg linear slip model. Finally, an initial fluid saturation (such as 50%) is set, and the elastic parameters of the initially fluid-saturated rock are calculated using the Gassmann equation, and these parameters are output as composite elastic parameters.

[0036] Step S3: Determine the fluid type of the thin and small reservoir and determine the property parameters to obtain the reservoir fluid parameters; perform fluid substitution based on the Gassmann equation according to the composite elastic parameters and the reservoir fluid parameters to obtain the elastic parameters of rocks with different saturations; output the fluid-coupled elastic parameters according to the elastic parameters of rocks with different saturations. In the embodiments of the present invention, by analyzing well logging data and geological data, the fluid type (such as oil, gas, water) in the thin and small reservoir is determined, and the property parameters (bulk modulus, density, and viscosity) of each fluid are determined. Then, based on the composite elastic parameters obtained in step S2 and the determined reservoir fluid parameters, fluid substitution is performed using the Gassmann equation to calculate the elastic parameters of the rock at different saturations (such as 0% to 100%, increasing by 10%). Finally, the elastic parameters of the rock at different saturations are sorted and output to obtain the fluid-coupled elastic parameters.

[0037] Step S4: Obtain the actual seismic data; perform seismic wavefield simulation according to the microscopic rock physical parameters and the fluid-coupled elastic parameters to obtain the simulated seismic wavefield data; perform convolution calculation according to the actual seismic data and the simulated seismic wavefield data, and output the high-resolution seismic response to obtain the high-resolution seismic response data. In the embodiments of the present invention, the actual seismic data of the study area is obtained. At the same time, based on the microscopic rock physical parameters obtained in step S1 and the fluid-coupled elastic parameters obtained in step S3, a three-dimensional geological model including the thin and small reservoir and the surrounding strata is constructed, and the multi-scale finite difference forward algorithm is used for seismic wavefield simulation to obtain the simulated seismic wavefield data. Extract the seismic wavelet from the actual seismic data, and convolve the extracted wavelet with the simulated seismic wavefield data to simulate the response of the seismic wave after being affected by the wavelet during underground propagation. Finally, the convolved seismic data is output as the high-resolution seismic response data.

[0038] Step S5: Use the high-resolution seismic response data to train a non-linear inversion model to obtain the non-linear inversion model; use the non-linear inversion model and the actual seismic data to perform reservoir parameter inversion and display processing of the inversion results to obtain the reservoir physical property inversion results of the thin and small reservoirs. In the embodiment of the present invention, the high-resolution seismic response data obtained in step S4 is used as the input feature, and the reservoir parameters (such as porosity, saturation) in the saturation geological model are used as the output labels to construct a training data set. Select a neural network model (such as a convolutional neural network) and use the constructed training data set for training to obtain a non-linear inversion model. Preprocess the actual seismic data to make it have the same data format and processing flow as the simulated seismic data. Then, use the trained non-linear inversion model to perform reservoir parameter inversion on the preprocessed actual seismic data. Finally, post-process (such as smoothing or constraining) and visually display the inversion results to obtain the reservoir physical property inversion results of the thin and small reservoirs.

[0039] Preferably, step S1 includes the following steps: Step S11: Collect core image data at different scales of the thin and small reservoir core samples through high-resolution CT scanning technology to obtain multi-scale core image data; Step S12: Perform image segmentation processing on the multi-scale core image data to obtain an image segmentation result; construct a three-dimensional digital core model based on the image segmentation result; Step S13: Characterize the pore results and fracture network based on the three-dimensional digital core model to obtain pore fracture network parameters; Step S14: Perform mineralogical analysis on the thin and small reservoir core samples to obtain mineralogical analysis results; perform quantitative analysis on the content and spatial distribution of different mineral components based on the three-dimensional digital core model and the mineralogical analysis results to obtain mineral composition and distribution data; Step S15: Perform fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; construct a microscopic rock physics model based on the fracture geometry analysis data, pore fracture network parameters, and mineral composition and distribution data, and perform physical model parameter calculation to obtain microscopic rock physics parameters.

[0040] As an example of the present invention, refer to Figure 2 As shown, in this example, step S1 includes: Step S11: Collect core image data at different scales of the thin and small reservoir core samples through high-resolution CT scanning technology to obtain multi-scale core image data; In the embodiments of the present invention, an industrial CT scanner equipped with a detector with micron-level resolution is used to scan thin and small reservoir core samples. First, the core samples are dried and encapsulated in epoxy resin to prevent deformation during scanning. Then, the core samples are scanned at different magnification factors to obtain core image data at at least three different scales, such as micron scale, millimeter scale, and centimeter scale. Scanning parameters, such as tube voltage, tube current, and exposure time, are adjusted according to the mineral composition and density of the core samples to obtain the best image quality. Finally, grayscale image data sets at different scales are output, and each data set contains hundreds to thousands of two-dimensional slice images, and each slice represents a thin cross-section of the core sample.

[0041] Step S12: Perform image segmentation processing on the multi-scale core image data to obtain an image segmentation result; construct a three-dimensional digital core model based on the image segmentation result. In the embodiments of the present invention, a threshold-based segmentation algorithm and an edge detection algorithm are used to perform image segmentation processing on the multi-scale core image data. First, the grayscale image is denoised, such as median filtering or Gaussian filtering, to remove the noise in the image. Then, according to the grayscale value differences of different mineral components and pores, multiple thresholds are set to segment the image into different regions, and each region represents a mineral or a pore. For fine structures such as fractures, an edge detection algorithm, such as the Canny operator or the Sobel operator, is used to identify the boundaries of the fractures. The segmentation results at different scales are registered and fused to construct a three-dimensional digital core model. This model represents the three-dimensional structure of the core sample in the form of voxels, and each voxel represents a small cube and is assigned the corresponding mineral or pore type.

[0042] Step S13: Characterize the pore results and fracture network based on the three-dimensional digital core model to obtain pore-fracture network parameters. In the embodiments of the present invention, based on the constructed three-dimensional digital core model, a pore network model extraction algorithm is used to extract the pore network structure. Parameters such as the connectivity, porosity, and permeability of the pore network are calculated. A fracture network extraction algorithm is used to extract the fracture network model, and fracture network parameters such as fracture density, fracture length, fracture aperture, and fracture connectivity are calculated. The extracted pore and fracture network parameters are statistically analyzed to obtain pore-fracture network parameters.

[0043] Step S14: Perform mineralogical analysis on the thin and small reservoir core samples to obtain a mineralogical analysis result; perform quantitative analysis on the content and spatial distribution of different mineral components based on the three-dimensional digital core model and the mineralogical analysis result to obtain mineral composition and distribution data. In the embodiments of the present invention, techniques such as X-ray diffraction (XRD) and scanning electron microscopy (SEM) are used to perform mineralogical analysis on thin and small reservoir core samples. XRD analysis can determine the types and relative contents of different mineral components in the core samples. SEM analysis can provide information on the morphology, size, and distribution of mineral particles in the core samples. Combining the three-dimensional digital core model and the results of mineralogical analysis, image analysis software is used to perform quantitative analysis on different mineral components to determine the content and spatial distribution of each mineral in the three-dimensional digital core model, and finally obtain mineral composition and distribution data.

[0044] Step S15: Perform fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; construct a microscopic rock physics model based on the fracture geometry analysis data, pore-fracture network parameters, and mineral composition and distribution data, and perform physical model parameter calculation to obtain microscopic rock physics parameters; In the embodiments of the present invention, based on the three-dimensional digital core model, a fracture network is extracted. Through three-dimensional image processing technology, fractures in the core model are identified and extracted. Geometric analysis is performed on the extracted fracture network to calculate parameters such as fracture length, width, aperture, dip angle, and azimuth angle. Combining the pore-fracture network parameters obtained in step S13 and the mineral composition and distribution data obtained in step S14, a suitable microscopic rock physics model is selected, such as the DEM model, Kuster-Toksöz model, or self-consistent model, etc. Substitute the fracture geometry analysis data, pore-fracture network parameters, and mineral composition and distribution data into the selected model to calculate microscopic rock physics parameters, such as mineral modulus, fracture compliance, dry rock modulus, etc.

[0045] Preferably, step S15 includes the following steps: Step S151: Calculate porosity and fracture density based on the three-dimensional digital core model to obtain porosity and fracture density data; Step S152: Obtain single-mineral elastic parameter data; perform a literature search on single-mineral elastic parameters according to the mineral composition and distribution data to obtain single-mineral elastic parameters; use the single-mineral elastic parameters to calculate mineral modulus to obtain mineral modulus data; Step S153: Perform fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; Step S154: Calculate anisotropy parameters based on the fracture geometry analysis data to obtain anisotropy parameter data; Step S155: Construct a microscopic rock physics model based on the porosity and fracture density data, mineral modulus data, fracture geometry analysis data, and anisotropy parameter data to obtain a microscopic rock physics model; Step S156: Use the microscopic rock physics model to calculate microscopic rock physics parameters to obtain microscopic rock physics parameters.

[0046] In the embodiment of the present invention, based on the three-dimensional digital core model obtained in step S12, the ratio of the number of pore voxels to the total number of voxels is statistically calculated to obtain the porosity. Using three-dimensional image processing technology, the fracture network in the core model is identified and extracted. The ratio of the total fracture length or total area to the core volume is calculated to obtain the fracture density data. The calculated porosity and fracture density values are recorded and stored for the subsequent construction of the microscopic rock physics model.

[0047] By referring to the published rock physics literature or databases (such as "The Rock Physics Handbook"), the elastic parameter data of various single minerals that make up the thin and small reservoir core samples are obtained, including the bulk modulus, shear modulus, and density. Ensure that the obtained elastic parameter data are measured under room temperature and atmospheric pressure conditions. According to the mineral composition and distribution data obtained in step S14, the elastic parameters of each mineral are weighted and averaged to calculate the mineral modulus data, such as the Voigt-Reuss-Hill average model.

[0048] Based on the three-dimensional digital core model generated in step S12, three-dimensional image processing technology is used to identify and extract the fracture network. The extracted fracture network is analyzed to calculate the geometric parameters of the fractures, including fracture length, width, aperture, dip angle, and azimuth angle. Statistical analysis of the geometric parameters of all fractures, such as calculating the average value, standard deviation, and distribution histogram, is performed to obtain the fracture geometric analysis data.

[0049] Based on the fracture geometric analysis data obtained in step S153, using fracture weakening theories, such as the Hudson model or the Schoenberg linear slip model, the anisotropic parameters of the rock are calculated. These parameters include the Thomsen parameters (ε, γ, δ) or the fracture compliance parameters. The geometric parameters of the fractures, such as fracture density, fracture aperture, and fracture direction, will be used to calculate these anisotropic parameters.

[0050] Based on the porosity and fracture density data obtained in step S151, the mineral modulus data obtained in step S152, the fracture geometric analysis data obtained in step S153, and the anisotropic parameter data obtained in step S154, a suitable microscopic rock physics model is selected, such as the Kuster-Toksöz model, the self-consistent model, or the differential effective medium (DEM) model. The above data are substituted into the selected model to construct a microscopic rock physics model, which describes the relationship between the elastic properties of the rock and the mineral composition, porosity, fracture density, and anisotropic parameters.

[0051] Using the microscopic rock physics model constructed in step S155, calculate microscopic rock physics parameters, such as dry rock moduli (bulk modulus and shear modulus), pore modulus, and fracture compliance. Store the calculated microscopic rock physics parameters for subsequent elastic parameter calculation and seismic wave field simulation.

[0052] Preferably, step S153 is specifically as follows: Extract the three-dimensional fracture network from the three-dimensional digital core model to obtain a three-dimensional fracture network model; perform skeletonization on the three-dimensional fracture network model to obtain a fracture skeleton model; Segment the fractures in the fracture skeleton model to obtain a segmented fracture model; Calculate the geometric parameters of individual fractures for the segmented fracture model to obtain a list of individual fracture geometric parameters; Perform stress field simulation on the three-dimensional digital core model to obtain stress field distribution data; based on the stress field distribution data, perform fracture deformation calculation on the segmented fracture model to obtain fracture deformation data; According to the list of individual fracture geometric parameters and the fracture deformation data, calculate the dynamic geometric parameters of each fracture segment under different stress states to obtain dynamic fracture geometric parameters; Perform statistical analysis on the list of dynamic fracture geometric parameters to obtain fracture geometric analysis data.

[0053] In the embodiments of the present invention, use three-dimensional image segmentation techniques, such as threshold-based segmentation or edge detection algorithms, to identify and extract fracture voxels from the three-dimensional digital core model. Connect all the fracture voxels to form a three-dimensional fracture network model, which accurately represents the distribution and morphology of fractures in three-dimensional space.

[0054] Adopt a three-dimensional thinning algorithm, such as medial axis transformation or topological thinning, to perform skeletonization on the three-dimensional fracture network model. Remove the thickness information of the fracture network and simplify it into a fracture skeleton model composed of central line segments, retaining the topological structure and connectivity of the fractures.

[0055] For each fracture segment in the segmented fracture model, calculate its geometric parameters, including length, average width, average aperture, inclination angle, and azimuth angle. Store the geometric parameters of each fracture segment in a list to form a list of individual fracture geometric parameters.

[0056] Use finite element analysis software, such as ABAQUS or COMSOL, to perform stress field simulation on the three-dimensional digital core model. Set boundary conditions and applied loads to simulate the stress distribution inside the core under different stress states. Output stress field distribution data, including the stress tensor on each voxel.

[0057] Map the stress field distribution data onto the segmented fracture model. Calculate the deformation of each fracture segment according to the stress state at the location of the fracture, such as the change in fracture width and aperture. Store the deformation of each fracture segment to form fracture deformation data.

[0058] Combine the single fracture geometric parameter list and the fracture deformation data to calculate the dynamic geometric parameters of each fracture segment under different stress states, such as dynamic width and dynamic aperture. These dynamic geometric parameters reflect the changes in the fracture under stress.

[0059] Conduct statistical analysis on the dynamic geometric parameters of all fracture segments, such as calculating the mean value, standard deviation, and distribution histogram, to obtain fracture geometric analysis data. These data will be used for subsequent petrophysical modeling and seismic inversion.

[0060] Preferably, step S2 includes the following steps: Step S21: Calculate the elastic parameters of the mineral matrix based on the microscopic petrophysical parameters to obtain the elastic parameters of the dry rock matrix; Step S22: Calculate the matrix pore elastic parameters based on the elastic parameters of the dry rock matrix and the microscopic petrophysical parameters to obtain the elastic parameters of the pore-containing dry rock; Step S23: Calculate the fracture elastic parameters based on the elastic parameters of the pore-containing dry rock and the microscopic petrophysical parameters to obtain the elastic parameters of the fracture-containing dry rock; Step S24: Calculate the elastic parameters of the fluid-saturated rock based on the elastic parameters of the fracture-containing dry rock and the preset initial fluid saturation to obtain the elastic parameters of the initially fluid-saturated rock; Step S25: Output the composite elastic parameters based on the elastic parameters of the initially fluid-saturated rock to obtain the composite elastic parameters.

[0061] As an example of the present invention, refer to Figure 3 As shown, in this example, step S2 includes: Step S21: Calculate the elastic parameters of the mineral matrix based on the microscopic petrophysical parameters to obtain the elastic parameters of the dry rock matrix; In the embodiment of the present invention, based on the microscopic petrophysical parameters, such as mineral moduli (including the bulk modulus and shear modulus of various minerals) and mineral volume fractions, the Voigt-Reuss-Hill average model or the Hashin-Shtrikman bounds model is used to calculate the elastic parameters of the dry rock matrix, including the bulk modulus and shear modulus of the dry rock matrix.

[0062] Step S22: Calculate the matrix pore elastic parameters based on the elastic parameters of the dry rock matrix and the microscopic petrophysical parameters to obtain the elastic parameters of the pore-containing dry rock; In the embodiments of the present invention, based on the elastic parameters of the dry rock matrix and the microscopic rock physical parameters, such as porosity, pore shape factor, etc., the elastic parameters of the porous dry rock are calculated by using rock physical models such as the Gassmann equation or the Kuster-Toksöz model, including the bulk modulus and shear modulus of the porous dry rock.

[0063] Step S23: Calculate the elastic parameters of the cracked dry rock according to the elastic parameters of the porous dry rock and the microscopic rock physical parameters to obtain the elastic parameters of the cracked dry rock; In the embodiments of the present invention, based on the elastic parameters of the porous dry rock and the microscopic rock physical parameters, such as fracture density, fracture aperture, fracture direction, etc., the elastic parameters of the cracked dry rock are calculated by using the Schoenberg linear slip model or the Hudson fracture model, including the bulk modulus and shear modulus of the cracked dry rock.

[0064] Step S24: Calculate the elastic parameters of the fluid-saturated rock according to the elastic parameters of the cracked dry rock and the preset initial fluid saturation to obtain the elastic parameters of the initial fluid-saturated rock; In the embodiments of the present invention, based on the elastic parameters of the cracked dry rock obtained in step S23, an initial fluid saturation value is set, such as 50%. According to the preset initial fluid saturation and the elastic parameters of the fluid and the rock matrix (the fluid parameters are determined in step S31), the Gassmann equation is used again to calculate the elastic parameters of the initial fluid-saturated rock, including the bulk modulus and shear modulus.

[0065] Step S25: Output the composite elastic parameters according to the elastic parameters of the initial fluid-saturated rock to obtain the composite elastic parameters.

[0066] In the embodiments of the present invention, the elastic parameters (bulk modulus and shear modulus) of the initial fluid-saturated rock calculated in step S24 are output as the composite elastic parameters. These composite elastic parameters comprehensively consider the effects of the mineral matrix, pores, fractures, and initial fluid saturation, and more accurately reflect the elastic properties of the rock.

[0067] Preferably, step S23 includes the following steps: Step S231: Extract the fracture geometric parameters from the microscopic rock physical parameters to obtain the fracture geometric parameters; Step S232: Calculate the fracture compliance parameters according to the fracture geometric parameters to obtain the fracture compliance parameters; Step S233: Construct the anisotropic compliance matrix by using the Schoenberg linear slip model according to the fracture compliance parameters and the fracture geometric parameters to obtain the anisotropic compliance matrix; Step S234: Calculate the compliance matrix of the cracked dry rock by performing calculations on the anisotropic compliance matrix and the elastic parameters of the porous dry rock, to obtain the compliance matrix of the cracked dry rock; Step S235: Perform matrix inversion calculation on the compliance matrix of the cracked dry rock to obtain the elastic parameters of the cracked dry rock.

[0068] In the embodiments of the present invention, fracture geometric parameters are extracted from the microscopic rock physical parameters, including fracture density, fracture aperture, fracture direction (dip angle and azimuth angle), etc. These parameters describe the geometric characteristics of the fractures and are the basis for calculating the fracture compliance parameters.

[0069] Based on the fracture geometric parameters, using fracture weakening theories, such as the Schoenberg linear slip model or the Hudson fracture model, calculate the fracture compliance parameters. The fracture compliance parameters describe the degree of influence of fractures on the elastic properties of rocks.

[0070] Based on the fracture compliance parameters and the fracture geometric parameters, especially the fracture direction, use the Schoenberg linear slip model to construct an anisotropic compliance matrix. This matrix is a 6x6 matrix that characterizes the compliance of the rock in different directions.

[0071] Superimpose the anisotropic compliance matrix with the elastic parameters of the porous dry rock (converted into a compliance matrix) to calculate the compliance matrix of the cracked dry rock. The superimposing method depends on the fracture model used. For example, the Schoenberg linear slip model uses the direct addition method.

[0072] Perform matrix inversion calculation on the compliance matrix of the cracked dry rock to obtain the elastic parameters of the cracked dry rock, including the bulk modulus and shear modulus of the cracked dry rock, as well as other elastic constants. These elastic parameters reflect the combined effects of pores and fractures.

[0073] Preferably, step S3 includes the following steps: Step S31: Determine the fluid type of the thin and small reservoir and determine the property parameters to obtain the reservoir fluid parameters; Step S32: Perform fluid substitution based on the Gassmann equation according to the composite elastic parameters and the reservoir fluid parameters to obtain the elastic parameters of rocks with different saturations; Step S33: Analyze the influence of saturation change on the elastic parameters according to the elastic parameters of rocks with different saturations to obtain the saturation-elastic parameter relationship curve; Step S34: Output the fluid-coupled elastic parameters according to the elastic parameters of rocks with different saturations and the saturation-elastic parameter relationship curve to obtain the fluid-coupled elastic parameters.

[0074] In the embodiments of the present invention, by analyzing well logging data (such as resistivity logging, acoustic logging, and neutron logging) and geological data, the fluid types in thin and small reservoirs are determined, such as oil, gas, water, or their mixtures. The property parameters of each fluid are determined, including bulk modulus, density, viscosity, etc. For the mixed fluid, the equivalent property parameters of the mixed fluid are calculated according to the volume ratio of each fluid. For example, the bulk modulus of the mixed fluid is calculated using the Wood equation. The determined fluid types and property parameters are recorded as reservoir fluid parameters.

[0075] Based on the composite elastic parameters and reservoir fluid parameters, fluid substitution is performed using the Gassmann equation. A series of different fluid saturation values are set, such as 0%, 10%, 20%,..., 100%. For each saturation value, the elastic parameters of the dry rock, porosity, fluid property parameters, and saturation value are substituted into the Gassmann equation to calculate the elastic parameters of the rock at different saturations, including bulk modulus and shear modulus.

[0076] Based on the series of elastic parameters of the rock at different saturations calculated in step S32, the influence of saturation change on the elastic parameters is analyzed. The relationship curves between saturation and elastic parameters (bulk modulus and shear modulus) are plotted, such as the variation curves of longitudinal wave velocity and transverse wave velocity with saturation. These curves reflect the variation law of the rock elastic parameters with fluid saturation.

[0077] Based on the elastic parameters of the rock at different saturations calculated in step S32 and the saturation-elastic parameter relationship curves plotted in step S33, the fluid-coupled elastic parameters are output. The fluid-coupled elastic parameters can be the elastic parameters of the rock at a specific saturation or the functional expression of the saturation-elastic parameter relationship curve, which are used to describe the relationship between the rock elastic parameters and fluid saturation. These parameters will be used for subsequent seismic wave field simulation and reservoir parameter inversion.

[0078] Preferably, step S32 includes the following steps: Step S321: Back-calculate the bulk modulus of the dry rock matrix according to the composite elastic parameters to obtain the bulk modulus of the dry rock matrix; Step S322: Determine the bulk modulus of the fluid mixture at different saturations according to the reservoir fluid parameters and the preset initial fluid saturation to obtain the bulk modulus of the fluid mixture at different saturations; Step S323: Substitute the bulk modulus of the dry rock matrix, the composite elastic parameters, and the bulk modulus of the fluid mixture at different saturations into the Gassmann equation to calculate the bulk modulus of the rock at different saturations to obtain the bulk modulus of the rock at different saturations; Step S324: Calculate the shear modulus of the rock at different saturations for the bulk modulus of the rock at different saturations to obtain the shear modulus of the rock at different saturations; Step S325: Calculate the densities of rocks with different saturations based on reservoir fluid parameters and microscopic rock physical parameters to obtain the densities of rocks with different saturations. Step S326: Calculate the elastic parameters of rocks with different saturations based on the bulk modulus of rocks with different saturations, the shear modulus of rocks with different saturations, and the densities of rocks with different saturations to obtain the elastic parameters of rocks with different saturations.

[0079] In the embodiment of the present invention, using the composite elastic parameters (including the bulk modulus and shear modulus of rocks with initial fluid saturation), porosity, and the initial fluid saturation and fluid bulk modulus preset in step S31, the bulk modulus of the dry rock matrix is calculated by inverting the Gassmann equation.

[0080] Based on the reservoir fluid parameters determined in step S31, for a series of different saturation values (such as 0%, 10%, 20%,..., 100%), calculate the bulk modulus of the fluid mixture at different saturations. If there are multiple fluids in the reservoir, according to the volume ratio and bulk modulus of each fluid, use the Wood equation or other mixed fluid models to calculate the equivalent bulk modulus of the mixed fluid.

[0081] Substitute the bulk modulus of the dry rock matrix, the composite elastic parameters, and the bulk modulus of the fluid mixture at different saturations into the Gassmann equation to calculate the bulk modulus of the rock at different saturations.

[0082] The Gassmann equation assumes that the shear modulus of the rock remains unchanged during the fluid replacement process. Therefore, the shear modulus of the rock with different saturations is the same as the shear modulus of the composite elastic parameters in step S25.

[0083] According to the reservoir fluid parameters (including fluid density) determined in step S31 and the microscopic rock physical parameters (including mineral density and porosity), calculate the densities of rocks with different saturations. Use the volume weighted average method to calculate the density of the mixture according to the volume content and density of different phases.

[0084] Organize the bulk modulus of the rock with different saturations, the shear modulus of the rock with different saturations, and the density of the rock with different saturations into the elastic parameters of the rock with different saturations. These parameters will be used for the subsequent analysis of the influence of saturation changes on elastic parameters.

[0085] Preferably, step S4 includes the following steps: Step S41: Obtain geological data; construct a fine geological model including thin and small reservoirs and surrounding strata based on the geological data and microscopic rock physical parameters to obtain a fine geological model. Step S42: Assign the fluid-coupled elastic parameters to the corresponding positions in the fine geological model to obtain the saturation geological model; Step S43: Based on the saturation geological model, perform seismic wavefield simulation using the multi-scale finite difference forward algorithm to obtain simulated seismic wavefield data; Step S44: Obtain actual seismic data; extract seismic wavelets from the actual seismic data to obtain seismic wavelet data; perform convolution calculation on the seismic wavelet data and the simulated seismic wavefield data to obtain the convolved seismic data; Step S45: Output high-resolution seismic responses based on the convolved seismic data to obtain high-resolution seismic response data.

[0086] In the embodiments of the present invention, geological data of the study area is collected, including formation thickness, lithology distribution, fault information, tectonic morphology, etc. Using data such as interpreted seismic profiles, logging data, and geological maps, combined with microscopic rock physical parameters, a fine geological model including thin and small reservoirs and surrounding formations is constructed. This model represents the underground geological structure in the form of a three-dimensional grid, and each grid cell is assigned corresponding lithology, porosity, permeability, and other parameters. The fine geological model should accurately reflect the geometric shape and physical properties of the thin and small reservoirs.

[0087] Assign the fluid-coupled elastic parameters to the corresponding positions in the fine geological model. According to the fluid saturation conditions at different positions in the thin and small reservoirs, the corresponding elastic parameters are assigned to the grid cells of the fine geological model to construct the saturation geological model.

[0088] Based on the saturation geological model constructed in Step S42, perform seismic wavefield simulation using the multi-scale finite difference forward algorithm. Set simulation parameters, including source type, source location, receiver location, simulation time length, spatial grid size, time step, etc. By numerically solving the wave equation, simulate the propagation process of seismic waves in the saturation geological model to obtain simulated seismic wavefield data, such as seismic records.

[0089] Obtain the actual seismic data of the study area. Extract seismic wavelets from the actual seismic data, for example, use statistical methods or deconvolution methods to extract seismic wavelets from the seismic data. Perform convolution calculation on the extracted seismic wavelets and the simulated seismic wavefield data obtained in Step S43 to simulate the influence of seismic wavelets on the propagation process of seismic waves underground to obtain the convolved seismic data.

[0090] Output the convolved seismic data obtained in Step S44 as high-resolution seismic response data. Since the simulated seismic wavefield data contains fine information of the thin and small reservoirs and the multi-scale finite difference forward algorithm is used, the obtained high-resolution seismic response data has higher resolution and can better reflect the characteristics of the thin and small reservoirs.

[0091] Preferably, step S5 includes the following steps: Step S51: Using the high-resolution seismic response data as input features and the reservoir parameters in the saturation geological model as output labels, construct a training data set to obtain a seismic-reservoir parameter training data set; Step S52: Use the seismic-reservoir parameter training data set to train a pre-selected neural network model to obtain a non-linear inversion model; Step S53: According to the processing flow and data format of the simulated seismic wavefield data, perform data and processing on the actual seismic data to obtain preprocessed seismic data; Step S54: Use the non-linear inversion model to invert the reservoir parameters of the preprocessed seismic data to obtain the inverted and predicted reservoir parameters; Step S55: Post-process the inverted and predicted reservoir parameters and perform inversion result display processing to obtain the reservoir physical property inversion result of the thin and small reservoir.

[0092] In the embodiment of the present invention, the high-resolution seismic response data obtained in step S45 is used as input features, and the reservoir parameters in the saturation geological model constructed in step S42, such as porosity, saturation, permeability, etc., are used as output labels. Pair the seismic data and the corresponding reservoir parameter data to construct a training data set. To improve the training effect, data augmentation can be performed on the training data set, such as adding random noise or performing random cropping. Finally, a seismic-reservoir parameter training data set is obtained.

[0093] Pre-select a neural network model, such as a convolutional neural network (CNN) or a recurrent neural network (RNN). Use the seismic-reservoir parameter training data set constructed in step S51 to train the pre-selected neural network model. Set training parameters, such as learning rate, batch size, number of iterations, etc. Update the weights and biases of the neural network model through the backpropagation algorithm to minimize the error between the predicted value and the true value. After training, a non-linear inversion model is obtained.

[0094] According to the processing flow and data format of the simulated seismic wavefield data obtained in step S43, preprocess the actual seismic data. The preprocessing steps include denoising, static correction, amplitude compensation, bandpass filtering, etc. Ensure that the preprocessed actual seismic data has the same data format and processing flow as the simulated seismic wavefield data so that the non-linear inversion model can effectively perform predictions.

[0095] Use the non-linear inversion model trained in step S52 to invert the reservoir parameters of the actual seismic data preprocessed in step S53. Input the preprocessed seismic data into the non-linear inversion model, and the model outputs the predicted reservoir parameters, such as porosity, saturation, permeability, etc.

[0096] Post-process the reservoir parameters predicted by the inversion in step S54, such as smoothing or constraint processing, to improve the stability and reliability of the inversion results. Visualize the post-processed inversion results, such as by plotting a plan view, a sectional view, or a three-dimensional model of the reservoir parameters. Finally, obtain the inversion results of the reservoir physical properties of the thin and small reservoir.

[0097] Therefore, from any perspective, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Thus, all changes falling within the meaning and scope of the equivalent elements of the application documents are intended to be embraced within the present invention.

[0098] The above are only specific embodiments of the present invention, enabling those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be obvious to those skilled in the art. The general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather will conform to the broadest scope consistent with the principles and novel features invented herein.

Claims

1. A high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs, characterized in that: The following steps are involved: Step S1: constructing a digital core model for a thin and small reservoir core sample to obtain a three-dimensional digital core model; performing a multi-mineral component quantitative analysis on the thin and small reservoir core sample to obtain mineral component and distribution data; performing a fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; Based on the fracture geometry analysis data and the mineral composition and distribution data, a micro rock physics model is constructed, and the physical model parameters are calculated to obtain the micro rock physics parameters; Step S2: Calculate matrix pore elastic parameters according to microscopic rock physical parameters to obtain elastic parameters of dry rock containing pores; calculate fracture elastic parameters according to the elastic parameters of dry rock containing pores and microscopic rock physical parameters to obtain elastic parameters of dry rock containing fractures; calculate fluid saturated rock elastic parameters according to the elastic parameters of dry rock containing fractures, and output composite elastic parameters to obtain composite elastic parameters; Step S3: Determine the fluid type and property parameters of the thin reservoir to obtain reservoir fluid parameters; perform fluid replacement based on the Gassmann equation according to the composite elastic parameters and the reservoir fluid parameters to obtain rock elastic parameters with different saturations; output fluid coupling elastic parameters according to the rock elastic parameters with different saturations to obtain fluid coupling elastic parameters; Step S4: obtaining actual earthquake data; Perform seismic wave field simulation based on microscopic rock physical parameters and fluid coupling elastic parameters to obtain simulated seismic wave field data; perform convolution calculation based on actual seismic data and simulated seismic wave field data, and perform high-resolution seismic response output to obtain high-resolution seismic response data; Step S5: Use high-resolution seismic response data to train a nonlinear inversion model to obtain a nonlinear inversion model; use the nonlinear inversion model and actual seismic data to invert reservoir parameters and perform inversion result display processing to obtain reservoir property inversion results of thin and small reservoirs.

2. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 1 is characterized in that: Step S1 includes the following steps: Step S11: collecting core images of thin and small reservoir core samples at different scales by high-resolution CT scanning technology to obtain multi-scale core image data; Step S12: performing image segmentation processing on the multi-scale core image data to obtain an image segmentation result; constructing a digital core model according to the image segmentation result to obtain a three-dimensional digital core model; Step S13: characterizing the pore results and fracture network according to the three-dimensional digital core model to obtain pore fracture network parameters; Step S14: performing a mineralogical analysis on the thin reservoir core sample to obtain a mineralogical analysis result; performing a quantitative analysis on the content and spatial distribution of different mineral components based on the three-dimensional digital core model and the mineralogical analysis result to obtain mineral composition and distribution data; Step S15: Perform fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; construct a micro rock physics model based on the fracture geometry analysis data, pore fracture network parameters, and mineral composition and distribution data, and calculate physical model parameters to obtain micro rock physics parameters.

3. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 2 is characterized in that: Step S15 includes the following steps: Step S151: Calculate porosity and fracture density according to the three-dimensional digital core model to obtain porosity and fracture density data; Step S152: Obtain single mineral elastic parameter data; perform literature search on single mineral elastic parameters based on mineral composition and distribution data to obtain single mineral elastic parameters; calculate mineral modulus using the single mineral elastic parameters to obtain mineral modulus data; Step S153: performing fracture geometry analysis on the three-dimensional digital core model to obtain fracture geometry analysis data; Step S154: performing anisotropic parameter calculation according to the fracture geometry analysis data to obtain anisotropic parameter data; Step S155: constructing a micro rock physics model according to the porosity and fracture density data, mineral modulus data, fracture geometry analysis data and anisotropy parameter data to obtain a micro rock physics model; Step S156: Calculate microscopic rock physical parameters using the microscopic rock physical model to obtain microscopic rock physical parameters.

4. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 3 is characterized in that: Step S153 is specifically as follows: Performing three-dimensional fracture network extraction on the three-dimensional digital core model to obtain a three-dimensional fracture network model; performing skeleton processing on the three-dimensional fracture network model to obtain a fracture skeleton model; Segmenting the crack skeleton model to obtain a segmented crack model; Calculate the geometric parameters of a single crack on the segmented crack model and obtain a list of geometric parameters of a single crack; Perform stress field simulation on the three-dimensional digital core model to obtain stress field distribution data; perform crack deformation calculation on the segmented crack model based on the stress field distribution data to obtain crack deformation data; According to the single crack geometric parameter list and crack deformation data, the dynamic geometric parameters of each crack under different stress states are calculated to obtain the dynamic crack geometric parameters; Perform data statistical analysis on the dynamic crack geometry parameter list to obtain crack geometry analysis data.

5. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 1 is characterized in that: Step S2 includes the following steps: Step S21: Calculate the mineral matrix elastic parameters according to the microscopic rock physical parameters to obtain the dry rock matrix elastic parameters; Step S22: Calculate matrix pore elastic parameters according to dry rock matrix elastic parameters and microscopic rock physical parameters to obtain elastic parameters of porous dry rock; Step S23: Calculate the elastic parameters of the fracture according to the elastic parameters of the dry rock containing pores and the microscopic rock physical parameters to obtain the elastic parameters of the dry rock containing fractures; Step S24: calculating the fluid-saturated rock elastic parameters according to the elastic parameters of the dry rock containing cracks and the preset initial fluid saturation to obtain the initial fluid-saturated rock elastic parameters; Step S25: Outputting composite elastic parameters according to the initial fluid-saturated rock elastic parameters to obtain composite elastic parameters.

6. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 5 is characterized in that: Step S23 includes the following steps: Step S231: extracting fracture geometry parameters from microscopic rock physical parameters to obtain fracture geometry parameters; Step S232: Calculating the crack flexibility parameters according to the crack geometric parameters to obtain the crack flexibility parameters; Step S233: constructing an anisotropic flexibility matrix using the Schoenberg linear slip model according to the fracture flexibility parameters and the fracture geometric parameters to obtain the anisotropic flexibility matrix; Step S234: Calculate the flexibility matrix of dry rock with cracks based on the anisotropic flexibility matrix and the elastic parameters of dry rock with pores to obtain the flexibility matrix of dry rock with cracks; Step S235: performing matrix inversion calculation on the flexibility matrix of the dry rock containing cracks to obtain elastic parameters of the dry rock containing cracks.

7. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 1 is characterized in that: Step S3 includes the following steps: Step S31: determining the fluid type of the thin reservoir and determining the property parameters to obtain reservoir fluid parameters; Step S32: performing fluid replacement based on the Gassmann equation according to the composite elastic parameters and reservoir fluid parameters to obtain rock elastic parameters with different saturations; Step S33: analyzing the influence of saturation change on elastic parameters according to rock elastic parameters with different saturations, and obtaining a saturation-elastic parameter relationship curve; Step S34: Output the fluid coupling elastic parameters according to the rock elastic parameters with different saturations and the saturation-elastic parameter relationship curve to obtain the fluid coupling elastic parameters.

8. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 7 is characterized in that: Step S32 includes the following steps: Step S321: back-calculating the bulk modulus of the dry rock matrix according to the composite elastic parameters to obtain the bulk modulus of the dry rock matrix; Step S322: determining the bulk modulus of the fluid mixture at different saturations according to the reservoir fluid parameters and the preset initial fluid saturation, and obtaining the bulk modulus of the fluid mixture at different saturations; Step S323: Substituting the bulk modulus of dry rock matrix, composite elastic parameters and bulk modulus of fluid mixtures with different saturations into the Gassmann equation, the bulk modulus of rock at different saturations is calculated to obtain the bulk modulus of rock at different saturations; Step S324: calculating the shear modulus of rocks with different saturations for the bulk moduli of rocks with different saturations, and obtaining the shear moduli of rocks with different saturations; Step S325: Calculate the density of rocks with different saturations according to reservoir fluid parameters and microscopic rock physical parameters to obtain the density of rocks with different saturations; Step S326: Calculate the elastic parameters of rocks with different saturations according to the bulk modulus of rocks with different saturations, the shear modulus of rocks with different saturations, and the density of rocks with different saturations to obtain the elastic parameters of rocks with different saturations.

9. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 1 is characterized in that: Step S4 includes the following steps: Step S41: Acquire geological data; construct a fine geological model including thin reservoirs and surrounding strata based on the geological data and microscopic rock physical parameters to obtain a fine geological model; Step S42: assigning the fluid coupling elastic parameters to the corresponding positions in the fine geological model to obtain a saturation geological model; Step S43: Based on the saturation geological model, a multi-scale finite difference forward algorithm is used to simulate the seismic wave field to obtain simulated seismic wave field data; Step S44: acquiring actual seismic data; extracting seismic wavelets from the actual seismic data to obtain seismic wavelet data; performing convolution calculation on the seismic wavelet data and simulated seismic wave field data to obtain convolution seismic data; Step S45: Output high-resolution seismic response according to the convolved seismic data to obtain high-resolution seismic response data.

10. The high-resolution seismic inversion identification method for thin and small reservoirs in oil reservoirs according to claim 9, characterized in that: Step S5 includes the following steps: Step S51: using high-resolution seismic response data as input features and reservoir parameters in the saturation geological model as output labels to construct a training data set to obtain a seismic-reservoir parameter training data set; Step S52: using the seismic-reservoir parameter training data set to perform nonlinear inversion model training on the pre-selected neural network model to obtain a nonlinear inversion model; Step S53: Processing the actual seismic data according to the simulated seismic wave field data in terms of the processing flow and data format to obtain pre-processed seismic data; Step S54: using a nonlinear inversion model to perform reservoir parameter inversion on the preprocessed seismic data to obtain reservoir parameters predicted by inversion; Step S55: Post-process the reservoir parameters predicted by the inversion, and perform inversion result display processing to obtain the reservoir property inversion results of the thin and small reservoir.

Images