A shear compliance factor calculation method based on constraint solving

Abstract

The present application belongs to the technical field of permeability prediction, and particularly relates to a shear compliance factor calculation method based on constraint solving. Based on the rock physics theory, the present application performs shear compliance factor modeling, and derives a theoretical formula of the shear compliance factor. On the basis of well logging curve splicing, completion and singular value elimination processing, the theoretical formula of the shear compliance factor is derived, a target function of constraint solving of the shear compliance factor is established, the L2 norm regularization least square method is used to solve the established target function, and finally the constraint solving of the shear compliance factor is obtained. The present application realizes the constraint solving of the shear compliance factor, and compared with directly solving the shear compliance factor, the correlation of the constraint solving of the shear compliance factor and the permeability is improved by about 30%, which provides basic data for accurately predicting the permeability.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of permeability prediction, and particularly relates to a shear compliance factor calculation method based on constraint solving. BACKGROUND

[0002] Reservoir permeability prediction using seismic data has always been a frontier and difficult problem of reservoir prediction. Since permeability and porosity are not simply one-to-one corresponding relationship, but also affected by pore structure, the corresponding relationship between porosity and permeability of different pore structure rocks is obviously different. Only by combining reservoir pore structure for permeability prediction can the multi-solution be reduced and the prediction accuracy be improved.

[0003] In recent years, a new idea of combining porosity and skeleton compliance factor for permeability prediction has been successfully applied in carbonate reservoirs such as Texas, Sichuan Basin Puguang Gas Field, etc. The method first calculates the skeleton compliance factor in the logging data, combines the skeleton compliance factor and rock thin section to classify the rock, and then calculates the regression of porosity and permeability in each type of rock.

[0004] Therefore, how to accurately calculate the skeleton compliance factor is one of the keys to improve the permeability prediction. The traditional method is to directly solve the skeleton compliance factor by formula. However, in actual application, it is found that due to various noises in logging data, the direct solution of skeleton compliance factor is very unstable, which further leads to inaccurate prediction of permeability. Therefore, an accurate method for calculating shear compliance factor is needed. SUMMARY

[0005] To achieve the above purpose, the shear compliance factor calculation method based on constraint solving improves the stability and accuracy of shear compliance factor calculation, and adopts the following technical solutions:

[0006] Step 1, based on rock physics theory, shear compliance factor modeling is carried out, and the theoretical formula of shear compliance factor is derived,

[0007] (1)

[0008] wherein, V p represents the longitudinal wave velocity, V s represents the transverse wave velocity, K s represents the bulk modulus, µ s represents the shear modulus, p represents the density;

[0009] (2)

[0010] where, K d denotes the dry rock bulk modulus, K m denotes the rock matrix bulk modulus, µ d denotes the dry rock shear modulus, µ m denotes the rock matrix shear modulus, denotes the porosity, g denotes the bulk skeleton compliance factor, g µ denotes the shear compliance factor;

[0011] Taking the logarithm of both sides of the shear skeleton compliance factor equation (2) gives:

[0012] (3)

[0013] where, µ d denotes the dry rock shear modulus, µ m denotes the rock matrix shear modulus, denotes the porosity, g µ denotes the shear compliance factor;

[0014] Rearranging equation (3) gives:

[0015] (4)

[0016] where, µ d denotes the dry rock shear modulus, µ m denotes the rock matrix shear modulus, denotes the porosity, g µ denotes the shear compliance factor;

[0017] Since the fluid has no effect on the shear modulus of the rock, we have:

[0018] (5)

[0019] µ d denotes the dry rock shear modulus, µ s denotes the shear modulus;

[0020] Finally, we obtain the theoretical equation for directly solving the shear compliance factor g µ :

[0021] (6)

[0022] wherein, g µ denotes the shear compliance factor; V s denotes the S-wave velocity; p denotes the density; m m The solution of the equation employs the Voigt-Reuss-Hill average method, the Voigt model assumes that various minerals constituting the rock are arranged in parallel along the stress direction, and the equivalent modulus is obtained by spatial volume average method; denotes the porosity;

[0023] Step 2, splicing, completing and singular value removing treatment are performed on the well logging curves;

[0024] Step 3, a target function for solving the shear compliance factor is established by deducing the theoretical formula of the shear compliance factor,

[0025] Further arranging the formula (3), the following formula is obtained:

[0026] (10)

[0027] wherein, V s denotes the S-wave velocity; p denotes the density; µ m denotes the shear modulus of the rock matrix, denotes the porosity, g µ denotes the shear compliance factor;

[0028] The above formula is simplified as:

[0029] (11)

[0030] wherein, D =log( V s 2 p ) - log µ m ; A =log( 1- ); X = g µ ; µ m denotes the shear modulus of the rock matrix; V s denotes the S-wave velocity;p Indicates density; Indicates porosity. g µ Indicates the shear compliance factor; n represents noise;

[0031] Solving by optimizing the objective function X Establish the following objective function:

[0032] (12)

[0033] in, D =log( V s 2 p ) - log µ m ; A =log( 1- ); X = g µ ; µ m Indicates the shear modulus of the rock matrix; V s Indicates the transverse wave velocity; p Indicates density; Indicates porosity. g µ Indicates the shear compliance factor;

[0034] By minimizing this objective function, the constrained shear compliance factor is obtained. ;

[0035] Step 4: Solve the established objective function using the L2 norm regularized least squares method to finally obtain the shear compliance factor of the constraint solution; extend the objective function formula (12) as follows:

[0036] (13)

[0037] in: That is, the L2 norm; D =log( V s 2 p ) - log µ m ; A =log( 1- ); X = g µ ;V s represents the shear modulus of the rock matrix; p represents the density; µ m represents the shear modulus of the rock matrix; represents the porosity, g µ represents the shear compliance factor; l is a constant; x is a vector X in the scalar form;

[0038] Derivation of formula (13) is zero, and the final form of the constraint solution of the shear compliance factor is:

[0039] (14)

[0040] in which, D =log( V s 2 p ) - log µ m ; A =log( 1- ); X = g µ ; A T represents the transpose of the matrix A ; µ m represents the shear modulus of the rock matrix; V s represents the shear modulus of the rock matrix; p represents the density; represents the porosity, g µ represents the shear compliance factor; I is a diagonal matrix; l is a constant.

[0041] The present application has the following beneficial effects:

[0042] In view of the problems of low accuracy and poor stability in directly solving the shear compliance factor in the current complex reservoir permeability prediction process, a shear compliance factor calculation method based on constraint solving is invented. The constraint solving of the shear compliance factor is realized for the first time, and the correlation between the constraint solving of the shear compliance factor and the permeability is improved by about 30% compared with the direct solving of the shear compliance factor, which provides basic data for accurately predicting the permeability. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 A flow chart of a method for calculating a shear compliance factor based on constraint solving;

[0044] Figure 2 A schematic diagram of the definition of the pore aspect ratio;

[0045] Figure 3 A schematic diagram of the relationship between the pore aspect ratio, porosity and velocity;

[0046] Figure 4 A schematic diagram of the relationship between the shear compliance factor, permeability and pore aspect ratio;

[0047] Figure 5 A schematic diagram of the pre-processing of well logging curves in a specific embodiment study area;

[0048] Figure 6 Well logging curves required for solving the shear compliance factor;

[0049] Figure 7 Comparison of the effects of different methods for solving the shear compliance factor;

[0050] Figure 8 Comparison of the correlation between different shear compliance factor solving results and permeability;

[0051] Figure 9 A crossplot of the permeability fitted by the constraint solving shear compliance factor and the original permeability. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described clearly and completely below in combination with specific embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present specification, all other embodiments obtained by those of ordinary skill in the art without making any creative labor fall within the scope of protection of the present application.

[0053] The matrix compliance factor, also known as the pore structure factor, is used to describe the influence of pores of different shapes and sizes on the elastic properties of the formation. Pore structure is one of the important factors affecting the permeability of the reservoir. The influence of pore structure on the characteristics of seismic response is mainly reflected in the rock matrix model, and it has a certain sensitivity to permeability. Fluid has a greater influence on bulk modulus, while shear modulus is not affected by fluid. Therefore, fluid has an influence on the bulk matrix compliance factor, but has no influence on the shear matrix compliance factor.

[0054] There are many parameters to characterize the pore structure in geology, and the pore aspect ratio (a) is usually used to characterize the pore structure in seismic. The pore aspect ratio has a clear indication of pore type and pore morphology. In theory, the pore aspect ratio of circular pores is the largest, the pore aspect ratio of pore + fracture is the second, and the pore aspect ratio of fracture is the smallest. For the same porosity, the larger the a, the larger the velocity; for the same velocity, the larger the a, the larger the porosity, as shown in Figure 2 .

[0055] The main influencing factors of shear skeleton flexibility factor include pore size, pore shape, permeability, and rock cementation degree. In theory, the pore type of shear flexibility factor has a good relationship. Generally speaking, the larger the pore aspect ratio, the smaller the shear flexibility factor; the smaller the pore aspect ratio, the larger the shear flexibility factor. The shear flexibility factor has a negative correlation with the pore aspect ratio, and has a positive correlation with the permeability, as shown in Figure 3 . The relationship between the shear flexibility factor, the permeability, and the pore aspect ratio is shown in Figure 4 .

[0056] Taking a research area in XH sag west sub-sag as an example, through physical property analysis, the porosity of the target layer reservoir is between 1.9% and 12%, mainly distributed in 5% to 10%, and the average is 7.4%; the permeability is between 0.029 and 44.1 mD, mainly distributed in 0.05 to 1 mD, and the average is 0.82, belonging to ultra-low porosity and low permeability to ultra-low permeability reservoir, and the reservoir physical property difference is large, the argillaceous content is low, and the local physical property sweet spot is developed, which is one of the target areas of new area production in the next step, the sweet spot distribution prediction is the main restricting factor of new area scheme compilation and implementation scheme optimization, and the permeability is an important sweet spot characteristic parameter, and accurately obtaining the shear flexibility factor is one of the key factors to obtain accurate permeability.

[0057] Step 1, shear flexibility factor modeling

[0058] The seismic parameters representing rock physical characteristics mainly include the elastic modulus of rock, P-wave velocity, S-wave velocity, density and attenuation, etc. It is an important parameter for identifying lithology and oil and gas, and is a bridge for quantitative seismic reservoir description, such as porosity, water saturation, permeability and formation pressure, etc.

[0059] One of the purposes of establishing rock physical model is to build a quantitative relationship between reservoir physical properties, fluid parameters and elastic parameters. The basic research of rock physics and related analysis method is the basis of seismic inversion and attribute analysis, which lays the foundation for further identification of lithology and hydrocarbon, and is also the basis for obtaining the skeleton flexibility factor.

[0060] Based on the rock physics theory, the shear compliance factor modeling is carried out, and the theoretical formula of the shear compliance factor is derived. The skeleton compliance factor is a parameter introduced in the new rock physics model derived by Sun, and mainly includes the bulk skeleton compliance factor and the shear skeleton compliance factor. These skeleton compliance factor parameters can be used to describe the influence of different shapes and sizes of pores on the elastic properties of the formation, wherein the bulk skeleton compliance factor explains the bulk strain under stress conditions, and the shear skeleton compliance factor explains the shear deformation.

[0061] (1)

[0062] wherein: V p denotes the longitudinal wave velocity, V s denotes the transverse wave velocity, K s denotes the bulk modulus, µ s denotes the shear modulus, p denotes the density.

[0063] (2)

[0064] wherein: K d denotes the dry rock bulk modulus, K m denotes the rock matrix bulk modulus, µ d denotes the dry rock shear modulus, µ m denotes the rock matrix shear modulus, denotes the porosity, g denotes the bulk skeleton compliance factor, g µ denotes the shear compliance factor.

[0065] Taking the logarithm of both sides of the following formula in the shear skeleton compliance factor formula (2):

[0066] (3)

[0067] wherein: µ d denotes the dry rock shear modulus, µ m denotes the rock matrix shear modulus, denotes the porosity, g µ denotes the shear compliance factor.

[0068] The formula (3) is arranged as follows:

[0069] (4)

[0070] wherein: µ d Gdry represents the dry rock shear modulus, µ m Gmat represents the rock matrix shear modulus, φ represents the porosity, g µ G represents the shear compliance factor.

[0071] Since the fluid has no effect on the shear modulus of the rock, we have:

[0072] (5)

[0073] µ d Gdry represents the dry rock shear modulus, µ s G represents the shear modulus.

[0074] Finally, we can obtain the theoretical formula for directly solving the shear compliance factor g µ G:

[0075] (6)

[0076] wherein: g µ G represents the shear compliance factor; m m The solution of G employs the Voigt-Reuss-Hill averaging method. The Voigt model assumes that the various minerals that make up the rock are arranged parallel to the force direction, and the equivalent modulus is obtained by spatial volume averaging. V s Vp represents the S-wave velocity; p ρ represents the density; φ represents the porosity.

[0077] (7)

[0078] wherein: G represents the Voigt model, f i G represents the percentage of different components, M i G represents the modulus of different components (quartz, clay, etc.).

[0079] The Ruess model assumes that the various minerals that make up the rock are arranged perpendicular to the force direction, and the equivalent modulus is obtained by spatial volume averaging.

[0080] (8)

[0081] wherein: denotes the Ruess model; f i denotes the percentage of different components, M i denotes different components

[0082] The Voigt-Reuss-Hill average is the arithmetic mean of the Voigt upper limit and the Reuss lower limit:

[0083] (9)

[0084] wherein , m m The Voigt-Reuss-Hill average method is used to solve the problem, the Voigt model assumes that the various minerals constituting the rock are arranged parallel to the force direction, and the equivalent modulus is obtained by spatial volume averaging denotes the Voigt model; denotes the Ruess model.

[0085] Step 2, well logging curve processing

[0086] In the field of oil and gas exploration and development, well logging curves are an important means of obtaining underground geological information. However, in actual operations, due to factors such as instruments and environment, well logging curves often have problems such as segmented measurement, data missing and uneven data distribution. Therefore, well logging curve splicing, missing curve completion and singular value elimination processing become key technical links.

[0087] Well logging curve splicing: During well logging operation, due to instrument tripping or measurement time limit, the same type of well logging curve may be collected in segments and distributed in different depth sections. For example, in complex formations, multiple trips are required for measurement, which leads to segmented curves. In order to meet the needs of subsequent data comprehensive research, the same type of curves in different depth sections need to be spliced into a continuous curve. Traditional manual splicing not only has low efficiency, but also easily introduces human error, affecting data accuracy. With the growth of well logging data, automatic splicing technology has emerged. The automatic splicing algorithm first determines the overlapping area between curves, finds the overlapping part by comparing depth information and curve characteristics. In the overlapping section, the correlation coefficient is calculated by using curve correlation analysis to determine the part with the highest matching degree. Then, in view of the data difference in the overlapping area, methods such as mean correction and trend correction are used to unify the curve amplitude. For example, if the mean values of the two curves in the overlapping section are biased, the bias value is calculated, and one of the curves is translated and corrected to make the data in the overlapping section smooth transition.

[0088] Missing curve completion: During the logging process, due to instrument failure, signal interference, etc., the curve may appear data missing. Missing data will destroy the integrity of the curve and affect the accuracy of subsequent geological analysis. For missing curve completion, interpolation method can be used in simple scenarios. If the missing segment is short, linear interpolation connects the known data points at both ends of the missing segment and estimates the missing value according to the linear relationship. Cubic spline interpolation can construct a smoother curve, which uses known data points to generate a curve with continuous second derivative to estimate the missing value. When the missing segment is long or there are multiple curve correlations, machine learning methods are more advantageous. Taking the random forest algorithm as an example, first construct a feature matrix, select other logging curve data related to the target curve as features, and the non-missing part of the target curve itself as the training label. Divide the data into valid data and missing data index, use valid data to train the random forest model, and after the model learns the relationship between features and target values, predict the missing data to complete the curve.

[0089] Outlier removal processing: During the acquisition of logging curves, due to instrument failure, sudden changes in downhole environment, or signal transmission interference, etc., abnormal values often deviate from the normal fluctuation range. These outliers are represented as data jump points, extreme outliers or sudden changes that do not conform to geological rules. If not removed, it will seriously interfere with the accuracy of reservoir parameter calculation, and even cause deviations in lithology identification, fluid property judgment and other geological interpretation. Outlier removal technology needs to combine statistical characteristics and geological logic. The statistical threshold method calculates the mean and standard deviation of the curve, and marks values outside a certain range as abnormal, which is suitable for normal distribution of conventional logging curves. The sliding window method uses local depth segments as units and removes outliers based on methods such as median and mean within the window, which can better adapt to the trend changes of the curve. For multi-curve joint processing scenarios, the isolated forest algorithm captures isolated points in data distribution by building a binary tree model, which has strong robustness for high-dimensional logging data. The wavelet transform method can separate high-frequency noise and effective signal corresponding to the outlier after decomposition by multiple scales.

[0090] Before calculating the shear compliance factor, logging curve splicing ensures curve continuity, missing curve completion repairs data integrity, and outlier removal, which are mutually complementary and lay a solid foundation for accurately calculating the shear compliance factor. They are indispensable key steps in the logging data processing flow.

[0091] As can be seen from formula (6), the logging curves required to solve the shear compliance factor include the shear wave curve, density curve, porosity curve, and clay content curve. Before obtaining the shear compliance factor, logging curve splicing ensures curve continuity, missing curve completion restores data integrity, and outlier removal is performed. These three aspects work together to lay a solid foundation for accurately obtaining the shear compliance factor. The logging curves in the study area mainly have two problems: outlier removal and missing curve completion. For the outlier removal problem, a combination of median filtering and mean filtering is used to remove outliers. For the missing curve completion problem, a random forest curve fitting method is used to complete the missing curves, such as... Figure 5 As shown, after singular value removal, the singular values ​​of the original curve were effectively removed, and the signal-to-noise ratio of the curve was significantly improved. After missing curve completion, the missing parts of the original curve were also well completed, providing a good data foundation for subsequent calculation of the shear compliance factor. Finally, all the logging curves (shale content, porosity, shear wave velocity, density) required for calculating the shear compliance factor are as follows: Figure 6 As shown.

[0092] Step 3: Establishing the objective function

[0093] By deriving the theoretical formula for the shear compliance factor, an objective function for constrained solution of the shear compliance factor is established.

[0094] The traditional method is to directly calculate the shear compliance factor using formula (6). g µ However, due to various noises in well logging data, such as those derived from measured shear wave velocity and density calculations... m d The Voigt-Reuss-Hill averaging method is used to calculate the average value. m m Solve directly with equal parameters g µ It is ill-posed and can be solved directly using formula (6). g µ The results will be very unstable; therefore, the method of minimizing the objective function is used to constrain the solution. g µ .

[0095] The objective function is the difference between the model's predictions and the actual results, usually expressed as a mathematical formula. The form of the objective function varies depending on the problem, but it is usually a convex function, which means that it has a global minimum in the entire parameter space.

[0096] Further reorganizing formula (3), we get:

[0097] (10)

[0098] wherein: V s represents the S-wave velocity; p represents the density; µ m represents the shear modulus of the rock matrix, represents the porosity, g µ represents the shear compliance factor.

[0099] The above equation is simplified as:

[0100] (11)

[0101] wherein, D = log( V s 2 p ) - log µ m ; A = log( 1- ); X = g µ ; µ m represents the shear modulus of the rock matrix; V s represents the S-wave velocity; p represents the density; represents the porosity, g µ represents the shear compliance factor; n represents noise.

[0102] Solving X by objective function optimization, the following objective function is established:

[0103] (12)

[0104] wherein, D = log( V s 2 p ) - log µ m ; A = log( 1- ); X = g µ ; µ m represents the shear modulus of the rock matrix;V s denotes the shear wave velocity; p denotes the density; denotes the porosity, g µ denotes the shear compliance factor.

[0105] By minimizing the objective function, i.e. g µ .

[0106] Step 4: Shear compliance factor solving

[0107] Shear compliance factor solving is actually an optimization solving of the objective function. Generally speaking, the optimization methods of the objective function mainly include gradient descent method, Newton method, genetic algorithm, simulated annealing algorithm, least squares method, etc. Gradient descent method is an iterative algorithm, which calculates the gradient of the objective function according to the current parameters each time, and then updates the parameters according to the gradient; Newton method is an iterative algorithm, which calculates the second-order derivative of the objective function according to the current parameters each time, and then updates the parameters according to the second-order derivative; genetic algorithm is a random search algorithm, which finds the optimal solution by constantly selecting, crossing and mutating the population; simulated annealing algorithm is a random search algorithm, which gradually reduces the temperature by constantly accepting small improvements, and finally reaches the global minimum value.

[0108] The least squares method is widely used in seismic exploration. In order to further improve the stability and accuracy of shear compliance factor solving, the least squares method is used to solve the established objective function (12). However, there are often a lot of noise in well logging data, therefore, in order to improve the stability and uniqueness of the solution, it is necessary to regularize the solution process.

[0109] Regularization refers to in linear algebra theory, ill-posed problems are usually defined by a set of linear algebraic equations, and this set of equations usually comes from ill-posed inverse problems with a large condition number. When the linear equation solution does not exist or is not unique, it is called ill-posed problem. Many times, the solution of ill-posed problem is needed. For the case of non-existent solution, the solution is to add some conditions to find an approximate solution, and for the case of non-unique solution, the solution is to add some restrictions to narrow the range of solution. This method of solving ill-posed problem by adding conditions or restrictions is called regularization method.

[0110] L2 norm regularization least squares method is used to solve the established objective function, which can realize the constrained solving of shear compliance factor. L2 norm refers to the square sum of each element in the vector and then taking the square root. For L2 norm regularization, in terms of statistics, it can be understood as prior knowledge, i.e. XShould not be too large; in terms of prediction, larger X values will cause AX larger changes when the model is applied to new instances, and can not achieve higher prediction performance, on the contrary, smaller X values can achieve higher prediction performance; in terms of optimization, L2 norm regularization gives a compromise between solving the system and having a small X , the objective function formula (12) is extended as follows by L2 norm regularization:

[0111] (13)

[0112] Wherein: , that is, the L2 norm; D =log( V s 2 p ) - log µ m ; A =log( 1- ); X = g µ ; V s represents the shear wave velocity; p represents the density; µ m represents the shear modulus of rock matrix; represents the porosity, g µ represents the shear compliance factor; l is a constant; x is the scalar form of the vector X .

[0113] Derivation of formula (13) is zero, and the final form of the constraint solution of the shear compliance factor is:

[0114] (14)

[0115] Wherein, D =log( V s 2 p ) - log µ m ; A =log( 1- ); X = gµ ; A T Representation matrix A Transpose of; µ m Indicates the shear modulus of the rock matrix; V s Indicates the transverse wave velocity; p Indicates density; Indicates porosity. g µ Indicates the shear compliance factor; I It is a diagonal matrix; l It is a constant.

[0116] Comparison of the effects of different methods on solving the shear compliance factor Figure 7 As shown, compared to directly solving for the shear compliance factor, the constrained solution for the shear compliance factor exhibits stronger local variation contrast, more stable overall variation, and improved resolution. A comparison of the correlation between different shear compliance factor solutions and permeability is also provided. Figure 8 As shown, the correlation coefficient between the directly calculated shear compliance factor and permeability is 0.23, while the correlation coefficient between the constrained shear compliance factor and permeability is 0.52, indicating a significant increase in correlation and demonstrating the accuracy of the constrained shear compliance factor. Figure 9 The cross plot of permeability obtained by constraining the linear fitting of the shear compliance factor and the original permeability shows a correlation of 0.86, indicating that the constrained shear compliance factor improves the accuracy of permeability fitting and prediction.

[0117] Those skilled in the art should understand that the above embodiments are merely illustrative and are not intended to imply that the scope of the invention is limited to these examples. Within the framework of this invention, technical features of the above embodiments or different embodiments can be combined, steps can be implemented in any order, and many other variations of the different aspects of the invention as described above exist, which are not provided in detail for the sake of brevity. Any omissions, modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A constraint-based solution-based method for determining shear flexibility factors, comprising: Comprising the following steps: Step 1, based on the rock physics theory, shear flexibility factor modeling is carried out, and the theoretical formula of shear flexibility factor is derived: (1) wherein, V p represents the longitudinal wave velocity, V s represents the transverse wave velocity, K s represents the bulk modulus, µ s represents the shear modulus, Rho represents the density; (2) wherein, K d represents the dry rock bulk modulus, K m represents the rock matrix bulk modulus, µ d represents the dry rock shear modulus, µ m represents the rock matrix shear modulus, represents the porosity, Gamma represents the bulk skeleton flexibility factor, Gamma µ represents the shear flexibility factor; logarithm of both sides: logarithm of both sides: (3) wherein, µ d Gdrepresents the dry rock shear modulus, µ m Gmrepresents the rock matrix shear modulus, φrepresents the porosity, Gamma µ Gdrepresents the dry rock shear modulus, Gmrepresents the rock matrix shear modulus, The formula (3) is arranged as follows: (4) wherein, µ d Gdrepresents the dry rock shear modulus, µ m Gmrepresents the rock matrix shear modulus, φrepresents the porosity, Rho µ Gdrepresents the dry rock shear modulus, Gmrepresents the rock matrix shear modulus, φrepresents the porosity, Gdrepresents the dry rock shear modulus, Gmrepresents the rock matrix shear modulus, φrepresents the porosity, Gdrepresents the dry rock shear modulus, Gmrepresents the rock matrix shear modulus, φrepresents the porosity, Gd Gamma (5) µ d Gdrepresents the dry rock shear modulus, µ s G represents the shear modulus; The final solution for the shear flexibility factor Because the fluid has no effect on the shear modulus of the rock, therefore: µ The theoretical formula: (6) wherein, Rho µ denotes the shear compliance factor; V s denotes the transverse wave velocity; Gamma denotes the density; Rho m denotes the rock matrix shear modulus; denotes the porosity; Mu Step 2, the logging curve is spliced, completed and singular value is removed; Step 3, by deducing the theoretical formula of shear flexibility factor, the objective function of constraint solving shear flexibility factor is established; (10) wherein: V s represents the shear modulus of the rock matrix, The formula (3) is further arranged as follows: represents the density, µ m represents the shear modulus of the rock matrix, represents the porosity, Rho µ represents the shear compliance factor; Gamma (11) wherein D = log( V s 2 The above formula is simplified as follows: ) - log µ m ; A = log( 1- ) ; X = Rho µ ; V s represents a shear wave velocity; Gamma represents a density; µ m represents a shear modulus of a rock matrix; represents a porosity, Rho µ represents a shear compliance factor; n represents noise; Solving by objective function optimization X An objective function is established as follows: (12) wherein D = log( V s 2 Gamma ) - log µ m ; A = log( 1- ) ; X = Rho µ ; V s denotes the shear wave velocity; Gamma denotes the density; µ m denotes the shear modulus of the rock matrix; denotes the porosity, Rho µ denotes the shear compliance factor; Gamma (13) wherein: i.e., the L2 norm; D = log( V s 2 Step 4, the L2 norm regularization least square method is used to solve the established objective function, and the constraint solving shear flexibility factor is obtained, and the objective function formula (12) is expanded as follows: ) - log µ m ; A = log( 1- ); X = Rho µ ; V s denotes the S-wave velocity; Gamma denotes the density; µ m denotes the shear modulus of the rock matrix; denotes the porosity, Rho µ denotes the shear compliance factor; Gamma is a constant; x is the scalar form of the vector X ; Lambda (14) wherein D = log( V s 2 The derivative of formula (13) is zero, and the form of the constraint solution of shear flexibility factor is as follows: ) - log µ m ; A = log( 1- ) ; X = log( Rho µ ; A T denotes the transpose of the matrix A ; µ m denotes the shear modulus of the rock matrix; V s denotes the S-wave velocity; Gamma denotes the density; denotes the porosity, Rho µ denotes the shear compliance factor; I is a diagonal matrix; Gamma is a constant.

2. The method of claim 1, wherein, Rock matrix shear modulus in step 1 Lambda m Solutions are obtained using Voigt-Reuss-Hill averaging method.

3. The method of claim 1, wherein, Mu Step 2 adopts the method of combining median filtering and mean filtering to remove singular values of logging curve, and adopts the method of random forest curve fitting to realize missing curve completion.

Images