A deep formation petrophysical modeling method in a complex lithology background

Abstract

The present application relates to the technical field of geological modeling, in particular to a deep stratum petrophysical modeling method under complex lithology background, comprising the following steps: well logging data preprocessing, mineral model optimization processing, petrophysical modeling and shear wave prediction, and verification of the accuracy of the petrophysical model in predicting shear wave. The present application adopts four mineral models of quartz-clay model, calcite-clay model, coal and bauxite, respectively, and performs optimal multi-mineral model processing in stratigraphic sections, so as to obtain the petrophysical modeling of deep coal measure stratum complex lithology combination and pore structure under shear wave prediction, and improve the accuracy of shear wave prediction.

Description

Technical Field

[0001] This invention relates to the field of geological modeling technology, specifically to a method for physical modeling of deep strata rocks in complex lithological backgrounds. Background Technology

[0002] Due to its great depth, the Ordos Basin has historically been considered more promising for exploration, with shallow and intermediate layers yielding similar natural gas production. Consequently, exploration and development have long focused on shallow and intermediate coalbed methane, neglecting large-scale exploration of natural gas in deep coal seams within the basin. However, recent explorations have revealed that natural gas production in deep coal seams is significantly higher than in shallow and intermediate layers, demonstrating promising exploration and development prospects. Therefore, rock physical modeling of deep coal-bearing strata is of paramount importance.

[0003] Deep coal-bearing strata possess complex lithological assemblages. In the deltaic front depositional environment of the Taiyuan Formation, generally in the transitional zone between marine and terrestrial sedimentary areas, thick mudstone forms the main coal-mudstone gas-accumulating assemblage, exhibiting good sealing and favorable gas accumulation conditions. In the shallow marine depositional environment of the Taiyuan Formation, in the dense limestone zone, a coal-limestone gas-accumulating assemblage forms, with the overlying dense limestone providing excellent capping. Deep coal strata are generally dual-porosity media with matrix pores and fractures. In the matrix of coal #8 in the Ordos Basin, vesicles, residual plant tissue pores, and inorganic mineral pores such as intercrystalline pores, intergranular pores, and intragranular pores are mainly developed. In addition, cleavage and microfractures are also abundant.

[0004] In summary, a single rock physics model cannot predict shear waves under complex lithological combinations and pore structures in deep coal-bearing strata. If only a single model is used, the rock physics modeling results will show obvious errors in complex lithological combinations such as limestone, coal, and bauxite. Summary of the Invention

[0005] The purpose of this invention is to provide a deep strata rock physics modeling method under complex lithological backgrounds, so as to solve the technical problems of low accuracy of shear wave prediction under complex lithological combinations and pore structures in deep coal-bearing strata in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution:

[0007] A method for deep stratigraphic rock physics modeling in complex lithological settings includes the following steps:

[0008] Well logging data preprocessing: Select high-quality standard wells and marker layers to preprocess the raw well logging data to obtain well logging data for establishing mineral models;

[0009] Mineral model optimization: The well logging data is combined with well logging and core data to determine a multi-mineral model. By analyzing lithological analysis data and whole-rock X-ray diffraction data, the multi-mineral model is optimized by layer to obtain the optimal multi-mineral model. In the optimal multi-mineral model, lithological identification is performed on special lithologies, well sections with special lithologies are delineated, and then the original data is backfilled and secondary processed separately to obtain the rock physical parameters used to establish the rock physical model.

[0010] Rock physics modeling and shear wave prediction: Based on the Gassman equation, the rock physics parameters, and the rock elastic response characteristics under different lithologies, physical properties, and fluid properties, a rock physics model based on a multi-mineral model was constructed, and shear wave prediction was performed through repeated iterations.

[0011] Verification of the accuracy of shear wave prediction by rock physics model: The P-wave and S-wave velocities and the P-wave-S-wave velocity ratio calculated by the rock physics model based on the multi-mineral model are compared with the measured data from well logging, and the accuracy of the shear wave prediction is verified based on the average error between the two.

[0012] As a preferred embodiment of the present invention, the preprocessing includes outlier removal, depth correction, curve stitching, environmental correction, and consistency correction, wherein:

[0013] Outlier removal: Identifying and removing samples that significantly deviate from other observations in the dataset using statistical or physical methods;

[0014] Depth correction: By comparing the logging curves of the reference layer or marker layer, and combining core analysis data, the depths of different logging curves are adjusted to be aligned with each other using sliding window matching and manual verification.

[0015] Curve splicing: The overlapping area matching method is used to eliminate abrupt changes at the seams of the segmented curve data, and the feature points are matched by correlation analysis to smoothly transition in the overlapping area for splicing.

[0016] Environmental correction: The curve with good identification effect of sandstone and mudstone is used as the sensitive parameter curve for lithological logging interpretation. The sonic transit time curve of adjacent well sections of the collapsed section is intersected with the sensitive parameter curve for lithological logging interpretation to obtain the fitted sonic transit time curve. Then, the logging curve of the collapsed section is replaced by the fitted sonic transit time curve and the fitted density curve to obtain the corrected logging curve.

[0017] Consistency correction: Select a preset number of key wells and determine the standard layer. Then compare the distribution of logging data of the treatment wells with the distribution of corresponding data of the key wells to determine the correlation and degree of difference between the two. Then, calculate a set of transformation values ​​required for correction and obtain unified inter-well data based on the standard wells.

[0018] As a preferred embodiment of the present invention, the method for determining the multi-mineral model includes:

[0019] The preprocessed logging data was depth-aligned and standardized with the logging lithology description and core test data.

[0020] The formation lithology was initially determined using well logging data and core descriptions. Mineral categories were classified based on well logging curve characteristics. A mineral database was established using core experimental data as a basis for multi-mineral model calibration.

[0021] We selected logging parameters that are sensitive to minerals, used statistical methods to establish a model that corresponds to the logging response and mineral content, and verified the accuracy of the multi-mineral model using core data.

[0022] By combining the physical property constraints of mineral assemblages, a multi-mineral model is constructed. The model's predictive effect is verified by core samples or oil test data that are not involved in the modeling, and the model parameters are adjusted based on the predictive effect.

[0023] As a preferred embodiment of the present invention, the multi-mineral model optimization processing method includes:

[0024] The well logging data is processed into different segments, and the strata are divided into different segments based on the lithological changes and geological characteristics of the strata.

[0025] Within each layer, through in-depth analysis of lithological analysis data and whole-rock X-ray diffraction data, four mineral models—quartz-clay model, calcite-clay model, coal, and bauxite—were used to optimize the multi-mineral model.

[0026] The focus is on identifying special lithologies and accurately delineating well sections with special lithologies, including coal, limestone, and bauxite.

[0027] For well sections with special lithology, the original data is backfilled and secondary processed separately to obtain porosity, gas saturation, and percentage content of multiple mineral components.

[0028] As a preferred embodiment of the present invention, the rock physical modeling method includes:

[0029] Step 1: Input the porosity, gas saturation, percentage content of multiple mineral components obtained from the optimized multi-mineral model, and the elastic modulus of clay, quartz, calcite, and coal obtained in the laboratory;

[0030] Step 2: Based on the composition of rock minerals, calculate the rock matrix modulus using the Gassman equation or Biot theory, calculate the mixed fluid bulk modulus using the Wood formula, give the water saturation, and construct the modulus of the dry rock skeleton.

[0031] Step 3: Using the rock matrix modulus, mixed fluid bulk modulus, and dry rock skeleton modulus obtained in Step 2 as inputs, establish the relationship between the input quantities based on the Gassman equation;

[0032] Step 4: Using density, clay content, porosity, and water saturation curves as basic data, and based on the obtained rock matrix modulus, mixed fluid bulk modulus, and dry rock skeleton modulus, a rock physics model based on a multi-mineral model is obtained through the Gassman equation.

[0033] As a preferred embodiment of the present invention, the method for predicting shear waves includes:

[0034] Rock physical parameters were input into a rock physical model improved based on a multi-mineral model. Through repeated iterative simulation calculations, the P-wave and S-wave velocities and the P-wave / S-wave velocity ratio were predicted.

[0035] As a preferred embodiment of the present invention, the original logging data includes 11 logging curves, namely, well diameter curve, drill bit diameter curve, natural gamma curve, deep and shallow resistivity curves, compensated neutron curve, P-wave transit time curve, S-wave transit time curve, bulk density curve, lithological density curve, and well inclination curve.

[0036] As a preferred embodiment of the present invention, the selection rules for the standard well include:

[0037] Standard wells have complete and accurate data of all types, including logging data, geological data, well logging data, and core test data;

[0038] Among them, the quality control of logging data mainly relies on the evaluation of the measured data. The evaluation includes whether the well diameter has been enlarged and whether there is a good correspondence between each curve in the lithological stable section.

[0039] Compared with the prior art, the present invention has the following advantages:

[0040] This invention addresses complex lithological backgrounds by processing multiple optimized mineral models and rock physics models separately for different lithologies and combining the results. The results are then repeatedly verified during alternating simulation and substitution calculations to achieve accurate prediction of shear wave velocity in coal-bearing strata under complex lithological backgrounds. Attached Figure Description

[0041] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0042] Figure 1 Flowchart of a method for petrophysical modeling of deep coal-bearing strata under complex lithological backgrounds provided in this embodiment of the invention;

[0043] Figure 2 This is a comparison chart of the original logging data before and after correction provided in an embodiment of the present invention;

[0044] Figure 3 This is a diagram showing the optimized multi-mineral model processing results provided in this embodiment of the invention.

[0045] Figure 4 Comparison chart of the optimized multi-mineral rock physics model processing results provided in the embodiments of the present invention;

[0046] Figure 5 Error analysis diagram of rock physics model calculation results provided in the embodiments of the present invention. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0048] like Figure 1 As shown, this invention provides a method for rock physics modeling of deep strata under complex lithological backgrounds, comprising the following steps:

[0049] Well logging data preprocessing: Select high-quality standard wells and marker layers to preprocess the raw well logging data, including outlier removal, depth correction, curve splicing, environmental correction, and consistency correction, to obtain well logging data for establishing mineral models;

[0050] In this invention, the selection of standard wells is crucial. Standard wells should represent the general characteristics of the oil-bearing reservoirs in the target area, including geological structure, reservoir thickness, and reservoir properties. When selecting high-quality standard wells, it is necessary to ensure that the standard wells have complete and accurate data of various types, including logging, geological, well logging, and core test data. Moreover, the quality control of logging data is mainly based on the evaluation of the measured data, such as whether the well diameter has been enlarged and whether there is a good correspondence between each curve in the lithologically stable section.

[0051] The original logging data in this invention includes 11 logging curves, namely, well diameter curve, drill bit diameter curve, natural gamma curve, deep and shallow resistivity curves, compensated neutron curve, P-wave transit time curve, S-wave transit time curve, bulk density curve, lithological density curve, and well inclination curve.

[0052] Outliers are samples that deviate significantly from other observations in a dataset and are typically identified and removed using statistical or physical methods.

[0053] Depth correction is used to correct depth deviations between different logging series. By comparing logging curves of reference or marker layers and combining core analysis data, the depths of different logging curves are adjusted to align them. This is generally achieved by a combination of sliding window matching and manual verification.

[0054] Curve splicing aims to eliminate abrupt changes at the seams of segmented curve data. It generally employs the overlapping area matching method, using correlation analysis to match feature points and perform splicing with a smooth transition in the overlapping area.

[0055] Environmental correction is used to eliminate the influence of environmental factors such as wellbore collapse, bottom hole temperature, pressure, and mud intrusion on measurement data. Generally, a reference value calibration method is used. A curve with good identification effect on sandstone and mudstone is selected as the sensitive parameter curve for lithological logging interpretation. The sonic transit time curves of adjacent sections of the collapsed section are intersected with the sensitive parameter curve for lithological logging interpretation to obtain a fitted sonic transit time curve. Finally, the fitted sonic transit time curve and the fitted density curve replace the logging curve of the collapsed section to obtain the corrected logging curve.

[0056] Consistency correction ensures the consistency of logging data across different times, locations, and environments. The general approach involves selecting a number of key wells and defining standard layers. Then, the distribution of logging data from the treated wells is compared with the distribution of corresponding data from the key wells to determine their correlation and degree of difference. This allows for the calculation of a set of transformation values ​​required for correction, ultimately resulting in unified inter-well data based on the standard wells.

[0057] like Figure 2 As shown, through comparative analysis of the P-wave time-volume density cross-plots before and after preprocessing, the data points after preprocessing correction show that some divergent outliers have been eliminated, resulting in more concentrated data points and better consistency. This invention, through the above series of data processing techniques, ultimately obtains a set of reliable well logging data, providing an accurate data foundation for subsequent mineral model processing and rock physics modeling.

[0058] Mineral model optimization: The multi-mineral model is determined by combining well logging data with core data. By analyzing lithological analysis data and whole-rock X-ray diffraction data, the multi-mineral model is optimized by layer to obtain the optimal multi-mineral model. In the optimal multi-mineral model, lithological identification is performed on special lithologies, well sections with special lithologies are delineated, and then the original data is backfilled and secondary processed separately to obtain the rock physical parameters used to establish the rock physical model.

[0059] Mineral models are mathematical or statistical models established by integrating well logging, drilling, and core data. They are used to quantitatively describe the type, content, and physical properties (such as density and sonic velocity) of mineral components (e.g., quartz, calcite, coal, bauxite, clay, etc.) in formations. They are fundamental tools for reservoir evaluation and rock mechanics analysis.

[0060] The steps for determining a multi-mineral model are as follows:

[0061] Step 1, Data Integration and Standardization: The preprocessed logging data (such as sonic logging, density logging, neutron logging, natural gamma logging, etc.) are deeply aligned and standardized with logging lithology descriptions and core experimental data (such as XRD mineral analysis). For example, the combination of logging parameters related to lithology is selected by cross plotting.

[0062] Step 2, Lithology Identification and Mineral Assemblage Analysis: Using well logging data and core descriptions, the formation lithology (e.g., sandstone, limestone, mudstone) is initially determined. Mineral categories are then classified based on well logging curve characteristics (e.g., low natural gamma indicator sandstone). A mineral database is established using core experimental data (e.g., mineral content percentage) as the basis for model calibration.

[0063] Step 3: Model Parameter Selection and Optimization: Select logging parameters that are sensitive to minerals. Use statistical methods (such as multiple regression and principal component analysis) to establish a model corresponding to the logging response and mineral content, and verify the model accuracy using core data.

[0064] Step 4: Model Building and Validation: Construct a multi-mineral model by incorporating the physical property constraints of the mineral assemblages. Validate the model's predictive performance using core samples not involved in the modeling or oil test data, and adjust parameters to improve accuracy.

[0065] This invention, through quantitative analysis of mineral composition, can accurately calculate parameters such as porosity, permeability, and hydrocarbon saturation, reducing errors caused by insufficient core sampling. It obtains continuous and reliable porosity, gas saturation, and percentage content of multiple mineral components, which can then be used as input datasets for rock physics modeling. Figure 3 As shown.

[0066] Optimization methods for multi-mineral models include:

[0067] Well logging data is processed into different segments, and the strata are divided into different segments based on the lithological changes and geological characteristics of the strata.

[0068] Within each layer, through in-depth analysis of lithological analysis data and whole-rock X-ray diffraction data, four mineral models—quartz-clay model, calcite-clay model, coal, and bauxite—were used to optimize the multi-mineral model.

[0069] The focus is on identifying special lithologies and accurately delineating well sections with special lithologies, including coal, limestone, and bauxite.

[0070] For well sections with special lithology, the original data is backfilled and secondary processed separately to obtain rock physical parameters such as porosity, gas saturation, and percentage content of multiple mineral components.

[0071] Rock physics modeling and shear wave prediction: Based on the Gassman equation, rock physics parameters, and simulation of the elastic response characteristics of rocks under different lithologies, physical properties (described by pore structure), and fluid properties, a rock physics model based on a multi-mineral model was constructed, and shear wave prediction was performed through repeated iterations.

[0072] In seismic data-driven reservoir and hydrocarbon prediction techniques, elastic parameters and combinations related to rock elastic response characteristics play a crucial role in establishing fundamental interpretation criteria. This invention, through analysis of rock physics experiments and well logging data, selects parameters and methods highly sensitive to seismic properties, thereby enhancing the predictive capabilities for lithology and fluids.

[0073] Rock physics modeling methods include:

[0074] Step 1: Input the porosity, gas saturation, percentage content of multiple mineral components obtained from the optimized multi-mineral model, and the elastic modulus of minerals such as clay, quartz, calcite, and coal obtained in the laboratory;

[0075] Step 2: Based on the composition of rock minerals, calculate the rock matrix modulus using the Gassman equation or Biot theory, calculate the mixed fluid bulk modulus using the Wood formula, give the water saturation, and construct the modulus of the dry rock skeleton.

[0076] Step 3: Using the rock matrix modulus, mixed fluid bulk modulus, and dry rock skeleton modulus obtained in Step 2 as inputs, establish the relationship between the input quantities based on the Gassman equation;

[0077] Step 4: Using density, clay content, porosity, and water saturation curves as basic data, and based on the obtained rock matrix modulus, mixed fluid bulk modulus, and dry rock skeleton modulus, a rock physics model based on a multi-mineral model is obtained through the Gassman equation.

[0078] Methods for predicting shear waves include:

[0079] By inputting rock physical parameters into a rock physical model improved based on a multi-mineral model, and through repeated iterative simulation calculations, the P-wave, S-wave velocities, and the P-wave-S-wave velocity ratio are predicted.

[0080] like Figure 4As shown, if only a single quartz-clay model is used for rock physics modeling, the results will show obvious errors in the modeling results for limestone, coal, and bauxite. This invention uses four mineral models, namely quartz-clay model, calcite-clay model, coal, and bauxite, to perform detailed rock physics modeling in layers. The results of multi-model combination and iterative processing are more reasonable and have a higher fitting degree with the measured longitudinal waves.

[0081] This invention establishes the interaction between the well logging formation evaluation model and the rock elastic properties through rock physics modeling. At the same time, through repeated iterations, the correction results of the original well logging data are corrected, providing a relatively complete and reliable dataset for subsequent wavelet extraction and low-frequency model construction, and obtaining a relatively accurate rock physics model prediction of shear waves.

[0082] This invention employs four mineral models—quartz-clay, calcite-clay, coal, and bauxite—and performs multi-mineral modeling optimization in layers to obtain a rock physical model that realizes the prediction of shear waves under the complex lithological combination and pore structure of deep coal-bearing strata. This improves the accuracy of shear wave prediction and provides a relatively complete and reliable dataset for wavelet extraction and low-frequency model construction in subsequent pre-stack seismic inversion.

[0083] Verification of the accuracy of shear wave prediction by rock physics model: The P-wave and S-wave velocities and the P-wave-S-wave velocity ratio calculated by the rock physics model based on the multi-mineral model are compared with the measured data from well logging, and the accuracy of the shear wave prediction is verified based on the average error between the two.

[0084] This invention compares the predicted shear wave velocity with actual well logging data to verify the model's accuracy. If the predicted results match the actual data well, the model is considered effective; otherwise, the model parameters need to be adjusted or the model structure modified.

[0085] like Figure 5 As shown, by comparing the P-wave and S-wave velocities and the P-wave-S-wave velocity ratio calculated by the multi-mineral model rock physics model with the well logging data, the average error between the two is less than 5%, which indicates the accuracy of the rock physics model.

[0086] This invention addresses complex lithological backgrounds by processing multiple optimized mineral models and rock physics models separately for different lithologies and combining the results. The results are then repeatedly verified during alternating simulation and substitution calculations to achieve accurate prediction of shear wave velocity in coal-bearing strata under complex lithological backgrounds.

[0087] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method for rock physical modeling of deep strata under complex lithological backgrounds, characterized in that, Includes the following steps: Well logging data preprocessing: Select high-quality standard wells and marker layers to preprocess the raw well logging data to obtain well logging data for establishing mineral models; Mineral model optimization: The well logging data is combined with well logging and core data to determine a multi-mineral model. By analyzing lithological analysis data and whole-rock X-ray diffraction data, the multi-mineral model is optimized by layer to obtain the optimal multi-mineral model. In the optimal multi-mineral model, lithological identification is performed on special lithologies, well sections with special lithologies are delineated, and then the original data is backfilled and secondary processed separately to obtain the rock physical parameters used to establish the rock physical model. Rock physics modeling and shear wave prediction: Based on the Gassman equation, the rock physics parameters, and the rock elastic response characteristics under different lithologies, physical properties, and fluid properties, a rock physics model based on a multi-mineral model was constructed, and shear wave prediction was performed through repeated iterations. Verification of the accuracy of rock physics model prediction of shear waves: The P-wave and S-wave velocities and the P-wave-S-wave velocity ratio calculated by the rock physics model based on the multi-mineral model are compared with the well logging data, and the accuracy of the shear wave prediction is verified based on the average error of the two. Optimization methods for multi-mineral models include: Well logging data is processed into different segments, and the strata are divided into different segments based on the lithological changes and geological characteristics of the strata. Within each layer, through in-depth analysis of lithological analysis data and whole-rock X-ray diffraction data, four mineral models—quartz-clay model, calcite-clay model, coal, and bauxite—were used to optimize the multi-mineral model. The focus is on identifying special lithologies and accurately delineating well sections with special lithologies, including coal, limestone, and bauxite. For well sections with special lithology, the original data is backfilled and secondary processed separately to obtain porosity, gas saturation, and percentage content of multiple mineral components.

2. The method for deep stratigraphic rock physical modeling under complex lithological backgrounds according to claim 1, characterized in that: The preprocessing includes outlier removal, depth correction, curve stitching, environmental correction, and consistency correction, wherein: Outlier removal: Identifying and removing samples that significantly deviate from other observations in the dataset using statistical or physical methods; Depth correction: By comparing the logging curves of the reference layer or marker layer, and combining core analysis data, the depths of different logging curves are adjusted to be aligned with each other using sliding window matching and manual verification. Curve splicing: The overlapping area matching method is used to eliminate abrupt changes at the seams of the segmented curve data, and the feature points are matched by correlation analysis to smoothly transition in the overlapping area for splicing. Environmental correction: The curve with good identification effect of sandstone and mudstone is used as the sensitive parameter curve for lithological logging interpretation. The sonic transit time curve of adjacent well sections of the collapsed section is intersected with the sensitive parameter curve for lithological logging interpretation to obtain the fitted sonic transit time curve. Then, the logging curve of the collapsed section is replaced by the fitted sonic transit time curve and the fitted density curve to obtain the corrected logging curve. Consistency correction: Select a preset number of key wells and determine the standard layer. Then compare the distribution of logging data of the treatment wells with the distribution of corresponding data of the key wells to determine the correlation and degree of difference between the two. Then, calculate a set of transformation values ​​required for correction and obtain unified inter-well data based on the standard wells.

3. The method for deep stratigraphic rock physical modeling under complex lithological backgrounds according to claim 1, characterized in that: The method for determining the multi-mineral model includes: The preprocessed logging data was depth-aligned and standardized with the logging lithology description and core test data. The formation lithology was initially determined using well logging data and core descriptions. Mineral categories were classified based on well logging curve characteristics. A mineral database was established using core experimental data as a basis for multi-mineral model calibration. We selected logging parameters that are sensitive to minerals, used statistical methods to establish a model that corresponds to the logging response and mineral content, and verified the accuracy of the multi-mineral model using core data. By combining the physical property constraints of mineral assemblages, a multi-mineral model is constructed. The model's predictive effect is verified by core samples or oil test data that are not involved in the modeling, and the model parameters are adjusted based on the predictive effect.

4. The method for deep stratigraphic rock physical modeling under complex lithological backgrounds according to claim 1, characterized in that: Rock physics modeling methods include: Step 1: Input the porosity, gas saturation, percentage content of multiple mineral components obtained from the optimized multi-mineral model, and the elastic modulus of clay, quartz, calcite, and coal obtained in the laboratory; Step 2: Based on the composition of rock minerals, calculate the rock matrix modulus using the Gassman equation or Biot theory, calculate the mixed fluid bulk modulus using the Wood formula, give the water saturation, and establish the modulus of the rock skeleton. Step 3: Using the rock matrix modulus, mixed fluid bulk modulus, and dry rock skeleton modulus obtained in Step 2 as inputs, establish the relationship between the input quantities based on the Gassman equation; Step 4: Using density, clay content, porosity, and water saturation curves as basic data, and based on the obtained rock matrix modulus, mixed fluid bulk modulus, and dry rock skeleton modulus, a rock physics model based on a multi-mineral model is obtained through the Gassman equation.

5. The method for deep stratigraphic rock physical modeling under complex lithological backgrounds according to claim 1, characterized in that: The method for predicting shear waves includes: By inputting rock physical parameters into a rock physical model improved based on a multi-mineral model, and through repeated iterative simulation calculations, the P-wave, S-wave velocities, and the P-wave-S-wave velocity ratio are predicted.

6. The method for deep stratigraphic rock physical modeling under complex lithological backgrounds according to claim 1, characterized in that: The original logging data includes 11 logging curves, namely, well diameter curve, drill bit diameter curve, natural gamma curve, deep and shallow resistivity curves, compensated neutron curve, P-wave transit time curve, S-wave transit time curve, bulk density curve, lithological density curve, and well inclination curve.

7. The method for deep stratigraphic rock physical modeling under complex lithological backgrounds according to claim 1, characterized in that: The selection rules for the standard wells include: Standard wells have complete and accurate data of all types, including logging data, geological data, well logging data, and core test data; Among them, the quality control of well logging data mainly relies on the evaluation of the measured data. The evaluation includes whether the well diameter has been enlarged and whether there is a good correspondence between each curve in the lithological stable section.

Images