Method and system for verifying influence of shale formation on fracture propagation and flow conductivity

Abstract

The invention discloses a method and system for verifying the influence of a shale formation on fracture propagation and flow conductivity, and relates to the field of geological engineering, and the method comprises the steps: obtaining a shale sample, and carrying out the preprocessing; carrying out fracturing control experiments on the pretreated shale sample under different experiment conditions; detecting the crack data of the sample after the experiment, and analyzing the influence of the shale formation on the expansion of the fracturing crack; detecting the oil distribution condition of the shale sample after the experiment, and analyzing the influence of the shale formation on the fracturing transformation flow conductivity; multiple pieces of sample data are obtained according to experimental data arrangement, a support vector machine model is constructed, a sample data set is used for training, and a shale flow conductivity prediction model is obtained. According to the invention, through a fracturing simulation experiment, the influence of the shale formation on the fracturing crack and the flow conductivity can be analyzed; and meanwhile, a machine learning model is trained based on experimental data, so that technical support can be provided for how to carry out fracturing oil extraction better.

Description

Technical Field

[0001] The present invention relates to the field of geological engineering technology, and more particularly to a method and system for verifying the influence of shale laminations on the expansion and conductivity of hydraulic fractures. Background Art

[0002] Shale is a dense rock with microscopic fracture structure and high organic matter content, and is generally considered to be a high-quality oil and gas reservoir. Shale usually includes laminar shale and massive shale. Laminar shale has a distinct bedding structure, while massive shale has a uniform massive structure. Shale lamination refers to the microcracks and fracture systems with a certain scale and connectivity in shale. Shale lamination increases the conductivity of the matrix on the one hand, and hinders the expansion of fractures on the other hand. It is a double-edged sword for shale oil development and exploitation.

[0003] In the existing shale oil and shale gas production process, hydraulic fracturing technology is usually used to fracture shale to increase the oil and gas production of shale. However, in the specific fracturing operation, the comprehensive influence of shale strata and fracturing conditions and parameters is often ignored or cannot be accurately based on the historical conventional operations and experience. The shale layer is fractured and transformed; there are problems such as poor comprehensive conductivity after transformation or high fracturing operation costs, which are not conducive to improving the comprehensive benefits of shale oil and gas production.

[0004] Therefore, how to analyze the comprehensive impact of shale laminations, fracturing conditions and shale transformation conductivity to provide accurate guidance for fracturing operations is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0005] In view of this, the present invention provides a verification method and system for the influence of shale laminations on fracture expansion and conductivity. Shale samples with different physical properties are selected to conduct fracturing experiments under different conditions, and the oil distribution of the shale samples before and after the experiment is analyzed, so as to determine the influence of shale laminations and transformation conditions on the shale conductivity.

[0006] In order to achieve the above object, the present invention provides the following technical solutions:

[0007] The present invention discloses a method for verifying the influence of shale lamination on fracture expansion and conductivity, and the specific steps are as follows:

[0008] Obtain laminar shale samples and massive shale samples with different physical properties and perform pretreatment;

[0009] The pretreated shale samples were subjected to fracturing control experiments under different experimental conditions;

[0010] The crack data of the samples after the test were tested, and the influence of shale lamination on the expansion of hydraulic fractures was analyzed; the oil distribution of the shale samples after the test was tested, and the influence of shale lamination on the conductivity of hydraulic fracturing transformation was analyzed;

[0011] A plurality of sample data are obtained according to the experimental data, and then a sample data set is obtained; a support vector machine model is constructed, and the sample data set is used for training to obtain a shale conductivity prediction model.

[0012] Further, the physical properties include: density, grain size, tensile strength, elastic modulus, permeability, laminar spacing, and laminar bonding strength of laminar shale samples, and density, grain size, tensile strength, elastic modulus, and permeability of massive shale samples;

[0013] The experimental conditions include: sample loading temperature, sample loading pressure, fracturing fluid pressure, fracturing flow rate, fracturing time, fracturing fluid proppant composition and concentration;

[0014] The crack data include: maximum crack length, maximum crack width, and maximum crack depth.

[0015] Further, the preprocessing includes:

[0016] The shale samples are treated with oil washing, and the cores are preliminarily cleaned to remove surface oil and impurities during the coring process; the cores are soaked with cleaning agents to decompose and dissolve the oil in the core pores; the oil in the cores is removed by ultrasonic cleaning; the residual oil and cleaning agents are washed away and then dried;

[0017] The shale sample after cleaning and drying is subjected to oil saturation treatment, the core is placed in a high-pressure oil saturation device, pressurized to a set pressure at a set rate; the pressure is maintained, and oil saturation injection is performed to allow the experimental oil to penetrate into the core; the core is vented to discharge free gas in the core pores; the pressure is reduced to normal atmospheric pressure at a set rate, and the surface is cleaned to complete the pretreatment.

[0018] Furthermore, the fracturing control experiment is specifically as follows:

[0019] The laminar shale sample and the massive shale sample having the same physical properties are selected, and the first laminar sample, the second laminar sample, the first massive sample, and the second massive sample are obtained by equally dividing the samples into a group of experimental samples;

[0020] A fracturing experiment is performed on the first laminae sample and the first block sample in the experimental samples, and a displacement treatment is performed; and a displacement treatment is directly performed on the second laminae sample and the second block sample in the experimental samples.

[0021] Furthermore, the analysis of the influence of shale lamination on the expansion of hydraulic fractures includes:

[0022] Respectively detecting the maximum crack length, maximum crack width, and maximum crack depth of the first laminar sample and the first block sample in each group of the experimental samples after the experiment, and calculating the difference;

[0023] Taking the laminae spacing value and laminae bonding strength value of the first laminae sample as the X and Y axes, and taking the difference of the maximum crack length, the difference of the maximum crack width, and the difference of the maximum crack depth as the Z axis, respectively, the first influence diagram, the second influence diagram, and the third influence diagram are obtained by drawing and fitting, and the influence of shale laminae on the expansion of hydraulic fractures is intuitively analyzed according to the three-dimensional surface of the influence diagram;

[0024] The analysis of the influence of shale lamination on conductivity includes:

[0025] After the experiment of each group of the experimental samples, the oil distribution of the first laminae sample, the second laminae sample, the first block sample, and the second block sample is detected respectively;

[0026] The difference in oil content between the first laminae sample and the second laminae sample is used as the fracturing conductivity of the laminae shale samples in the group of experimental samples, and the difference in oil content between the first block sample and the second block sample is used as the fracturing conductivity of the block shale samples in the group of experimental samples;

[0027] The fourth influence diagram is drawn with the laminar spacing value and laminar bonding strength value of the laminar shale sample as the X and Y axes, and the difference between the fracturing conductivity of the laminar shale sample and the fracturing conductivity of the massive shale sample as the Z axis, so as to intuitively analyze the influence of shale lamination on the fracturing conductivity.

[0028] Furthermore, the oil distribution is determined by detecting and drawing a three-dimensional oil-containing model of the sample through nuclear magnetic resonance. The oil content is calculated as follows:

[0029] Meshing the oil-bearing three-dimensional model to transform the oil-bearing three-dimensional model into a plurality of independent and connected cubes;

[0030] marking the cubes that are completely oil-containing and the cubes that are partially oil-containing among the cubes;

[0031] The oil-containing volume of the full oil-containing cube and the oil-containing volume of the partial oil-containing cube are calculated respectively, and the oil content is obtained by adding them up;

[0032] The oil-containing volume of the partial oil-containing cube is calculated as follows:

[0033] Determining the interface, non-interface, and oil body curved surface between the oil body and the cube surface in the partial oil-containing cube;

[0034] When the number of interfaces is 3, 4 or 5, the oil-containing volume is directly calculated;

[0035] When the number of interfaces is 6, the non-oil volume is first calculated, and the oil-containing volume is obtained by subtracting the non-oil volume from the cubic volume;

[0036] Furthermore, the direct calculation of the oil-containing volume is specifically as follows:

[0037] Step 1: Generate a triangulated network model according to the key points of the oil body surface;

[0038] Step 2: Determine the intersection of the normal vectors of the center points of all triangles in the current triangulated network model and the oil body surface;

[0039] Step 3: Determine three sub-triangles by connecting the intersection point of the triangle with the vertex of the triangle and the sides of the triangle;

[0040] Step 4: Update the triangulated network model with all the sub-triangles to determine whether the set update times are met, and if not, return to step 2;

[0041] Step 5: Taking the opposite surface of any non-intersection interface as the bottom surface, projecting all triangles in the triangulated mesh model onto the bottom surface to generate subdivided prisms of the oil body;

[0042] Step 6: Calculate the volume of each prism and add them together to obtain the oil-containing volume.

[0043] Furthermore, the construction of the support vector machine model includes:

[0044] Set the first strategy, the second strategy, and the third strategy as optimization strategies;

[0045] Setting a dynamic weight factor, optimizing individual update positions by the dynamic weight factor, and obtaining a dynamic adaptive weight population algorithm;

[0046] Combining the dynamic adaptive weighted population algorithm and the optimization strategy to obtain an improved population algorithm;

[0047] A support vector machine model is constructed based on the improved population algorithm, and the kernel function and penalty coefficient of the support vector machine model are corrected by the improved population algorithm.

[0048] Furthermore, the first strategy, the second strategy, and the third strategy are respectively a firefly perturbation strategy, a Cauchy mutation strategy, and a reverse learning strategy;

[0049] The dynamic weight factor formula is:

[0050]

[0051] Among them, ω represents the weight factor, e is the base of the natural logarithm, t represents the current number of iterations, iter max Indicates the maximum number of iterations;

[0052] The dynamic adaptive weighted population algorithm satisfies the following formula:

[0053]

[0054] in, represents the new position of the jth dimension of the ith individual, X i,j represents the current position of the i-th individual, p i,j represents the prey position of the i-th individual, I represents a random integer of 1 or 2, The objective function value F represents the prey position of the i-th individual i represents the fitness value of the ith individual and F represents the objective function vector of the population. Represents the objective function value of the i-th individual after updating based on the first stage;

[0055] The improved population algorithm satisfies the following formula:

[0056]

[0057] in, Indicates the population's latest position update parameter, X′ i,j represents the inverse solution of the optimal solution of the tth iteration, b1 represents the information exchange control coefficient, Indicates the fitness corresponding to the latest individual, F i ′ represents the firefly position fitness value of the i-th individual, Cauchy(0,1) satisfies the standard Cauchy distribution, and Cauchy represents The Cauchy mutation operator of .

[0058] The present invention also discloses a verification method system for the influence of shale lamination on fracture expansion and conductivity, which is characterized by comprising:

[0059] Sample acquisition module: obtain laminar shale samples and massive shale samples with different physical properties and perform preprocessing;

[0060] Control experiment module: conduct fracturing control experiments under different experimental conditions on pretreated shale samples;

[0061] Detection and analysis module: detect the fracture data of samples after the experiment, analyze the influence of shale lamination on the expansion of fractures; detect the oil distribution of the shale samples after the experiment, and analyze the influence of shale lamination on the conductivity of fracturing transformation;

[0062] Prediction module: multiple sample data are obtained according to the experimental data, and then a sample data set is obtained; a support vector machine model is constructed, and the sample data set is used for training to obtain a shale conductivity prediction model.

[0063] It can be seen from the above technical scheme that, compared with the prior art, the present invention discloses a verification method and system for the influence of shale lamination on fracture expansion and conductivity. By conducting fracturing simulation experiments under different conditions on shales with different physical properties, the influence of shale lamination on fracture expansion and conductivity can be analyzed. At the same time, a sample data set is constructed by obtaining experimental data under various conditions through experiments, and a conductivity prediction model is constructed using a machine learning model, which can provide technical support for how to better carry out fracturing oil production. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0065] Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] The embodiment of the present invention discloses a method for verifying the influence of shale lamination on fracture expansion and conductivity, as shown in FIG. Figure 1 As shown, the specific steps are as follows:

[0068] Obtain laminar shale samples and massive shale samples with different physical properties and perform pretreatment;

[0069] The pretreated shale samples were subjected to fracturing control experiments under different experimental conditions;

[0070] Detect the fracture data of the samples after the experiment and analyze the influence of shale lamination on the expansion of fractures; detect the oil distribution of shale samples after the experiment and analyze the influence of shale lamination on the conductivity of fracturing transformation;

[0071] A plurality of sample data are obtained by collating the experimental data, and then a sample data set is obtained; a support vector machine model is constructed, and the sample data set is used for training to obtain a shale conductivity prediction model.

[0072] In a specific embodiment, the physical properties include: density, grain size, tensile strength, elastic modulus, permeability, laminar spacing, and laminar cohesion strength of laminar shale samples, and density, grain size, tensile strength, elastic modulus, and permeability of massive shale samples;

[0073] The experimental conditions included: sample loading temperature, sample loading pressure, fracturing fluid pressure, fracturing flow rate, fracturing duration, fracturing fluid proppant composition and concentration;

[0074] Crack data include: maximum crack length, maximum crack width, and maximum crack depth.

[0075] In a specific embodiment, the preprocessing includes:

[0076] The shale samples are treated with oil washing, and the cores are preliminarily cleaned to remove surface oil and impurities during the coring process; the cores are soaked with cleaning agents to decompose and dissolve the oil in the core pores; the oil in the cores is removed by ultrasonic cleaning; the residual oil and cleaning agents are washed away and then dried;

[0077] The cleaned and dried shale samples are subjected to oil saturation treatment. The core is placed in a high-pressure oil saturation device and pressurized to a set pressure at a set rate; the pressure is maintained and oil saturation injection is performed to allow the experimental oil to penetrate into the core; the core is vented to discharge free gas in the core pores; the pressure is reduced to normal atmospheric pressure at a set rate, and the surface is cleaned to complete the pretreatment. Through oil washing and supersaturated oil saturation treatment, a representative and unified experimental oil is selected to replace the reservoir oil in the sample, which can effectively unify the reservoir oil composition in each sample, ensure that the reservoir oil in the sample is saturated, and reduce the error of the comparative experiment.

[0078] In a specific embodiment, the fracturing control experiment is specifically as follows:

[0079] Select laminar shale samples and massive shale samples with the same physical properties, divide them equally to obtain the first laminar sample, the second laminar sample, the first massive sample, and the second massive sample as a group of experimental samples;

[0080] A fracturing experiment is performed on the first laminae sample and the first block sample in the experimental samples, and a displacement treatment is performed; a displacement treatment is directly performed on the second laminae sample and the second block sample in the experimental samples;

[0081] Analyze the impact of shale lamination on fracture propagation, including:

[0082] After the experiment, the maximum crack length, maximum crack width and maximum crack depth of the first laminar sample and the first block sample in each group of experimental samples were detected respectively, and the difference was calculated;

[0083] With the laminae spacing value and laminae bonding strength value of the first laminae sample as the X and Y axes, and the difference in maximum crack length, maximum crack width and maximum crack depth as the Z axis, the first, second and third influence diagrams were plotted and fitted. The influence of shale laminae on the expansion of hydraulic fractures was intuitively analyzed based on the three-dimensional surface of the influence diagram.

[0084] In a specific embodiment, analyzing the effect of shale lamination on conductivity includes:

[0085] After testing each group of experimental samples, the oil distribution of the first laminae sample, the second laminae sample, the first block sample, and the second block sample was detected;

[0086] The difference in oil content between the first laminar sample and the second laminar sample is used as the fracturing conductivity of the laminar shale samples in the group of experimental samples, and the difference in oil content between the first block sample and the second block sample is used as the fracturing conductivity of the block shale samples in the group of experimental samples;

[0087] The fourth influence diagram is drawn with the laminar spacing value and laminar bonding strength value of the laminar shale sample as the X and Y axes, and the difference between the fracturing conductivity of the laminar shale sample and the fracturing conductivity of the massive shale sample as the Z axis, and then the influence of shale lamination on the fracturing conductivity is intuitively analyzed.

[0088] In a specific embodiment, the oil content of the sample is detected and a three-dimensional model is drawn by nuclear magnetic resonance to determine the oil distribution; the oil content is calculated as follows:

[0089] Meshing the oil-bearing three-dimensional model, and converting the oil-bearing three-dimensional model into a number of independent and connected cubes;

[0090] Mark the cubes that are completely oil-filled and the cubes that are partially oil-filled;

[0091] The oil-containing volume of the full oil-containing cube and the oil-containing volume of the partial oil-containing cube are calculated separately, and the oil content is obtained by adding them together;

[0092] The oil-containing volume of a partially oil-containing cube is calculated as follows:

[0093] Determine the interface, non-interface, and oil body surface between the oil body and the cube surface in some oil-containing cubes;

[0094] Usually, the cube obtained by division is a cube. By selecting the size of the divided cube, the oil body contained in each cube is a whole. Then, there may be 3-6 interfaces between the oil body and the six surfaces of the cube. When there are 3-5 interfaces, the oil body can be directly projected onto the surface of the cube and divided into several prisms. Therefore, when there are 3, 4 or 5 interfaces, the oil-containing volume can be directly calculated.

[0095] When there are 6 interfaces, the non-oil volume is calculated first, and the oil-containing volume is obtained by subtracting the non-oil volume from the cubic volume;

[0096] In a specific embodiment, the oil-containing volume is directly calculated as follows:

[0097] Step 1: Generate a triangulated network model based on the key points of the oil body surface;

[0098] Step 2: Determine the intersection of the normal vectors of the center points of all triangles in the current triangulated network model and the oil body surface;

[0099] Step 3: Determine three sub-triangles using the intersection of the triangles, the lines connecting the triangle vertices, and the sides of the triangles;

[0100] Step 4: Update the triangulated network model with all sub-triangles to determine whether the set update times are met. If not, return to step 2.

[0101] Step 5: Take the opposite surface of any non-interface as the bottom surface, project all triangles in the triangulated mesh model onto the bottom surface, and generate subdivided prisms of the oil body;

[0102] Step 6: Calculate the volume of each prism and add them together to obtain the oil-containing volume.

[0103] Among them, the key points are obtained by detecting the corners of the points in the oil body surface and determining the inflection points of the surface. By iteratively updating and generating a more subdivided triangulated mesh model, the error in estimating the volume of the oil body with triangulated meshes can be minimized. Excessive iterative subdivision will significantly increase the amount of calculation. Here, the number of iterations is 4, thus achieving a balance between calculation error and calculation complexity.

[0104] The calculation method of the non-oil body volume is similar to that of the oil body volume. The triangulated mesh model is divided and subdivided through steps 1-4, and the triangles are projected to the plane where any non-oil body is located to generate non-oil body subdivision prisms. The volume of each prism is then calculated and added to obtain the volume of the non-oil body.

[0105] In a specific embodiment, constructing a support vector machine model includes:

[0106] Set the first strategy, the second strategy, and the third strategy as optimization strategies;

[0107] Set a dynamic weight factor, optimize the individual update position through the dynamic weight factor, and obtain a dynamic adaptive weight population algorithm;

[0108] Combining the dynamic adaptive weighted population algorithm and optimization strategy, an improved population algorithm is obtained;

[0109] A support vector machine model is constructed based on the improved population algorithm, and the kernel function and penalty coefficient of the support vector machine model are corrected by the improved population algorithm.

[0110] The physical properties and experimental conditions of the shale sample are used as sample feature data, and the corresponding conductivity is used as a sample label to obtain sample data. Multiple sample data are obtained according to the experimental data, and then a sample data set is obtained; a support vector machine model is constructed, and the sample data set is used for training to obtain a shale conductivity prediction model. By improving the population algorithm, the parameters of the modified support vector machine model can be effectively found, so as to better fit the relationship between the fracturing conditions and the backflow capacity, thereby obtaining an accurate shale conductivity prediction model, providing effective guidance for actual fracturing operations.

[0111] In a specific embodiment, the first strategy, the second strategy, and the third strategy are respectively a firefly perturbation strategy, a Cauchy mutation strategy, and a reverse learning strategy;

[0112] The dynamic weight factor formula is:

[0113]

[0114] Among them, ω represents the weight factor, e is the base of the natural logarithm, t represents the current number of iterations, iter max Indicates the maximum number of iterations;

[0115] Dynamic adaptive weighted population algorithm satisfies the following formula:

[0116]

[0117] in, represents the new position of the jth dimension of the ith individual, X i,j represents the current position of the i-th individual, p i,j represents the prey position of the i-th individual, I represents a random integer of 1 or 2, The objective function value F represents the prey position of the i-th individual irepresents the fitness value of the ith individual and F represents the objective function vector of the population. Represents the objective function value of the i-th individual after updating based on the first stage;

[0118] Improve the population algorithm to meet the following formula:

[0119]

[0120] in, Indicates the population's latest position update parameter, X′ i,j represents the inverse solution of the optimal solution of the tth iteration, b1 represents the information exchange control coefficient, Indicates the fitness corresponding to the latest individual, F i ′ represents the firefly position fitness value of the i-th individual, Cauchy(0,1) satisfies the standard Cauchy distribution, and Cauchy represents The Cauchy mutation operator of .

[0121] In a specific embodiment, the improved population algorithm further includes comparing the fitness value of the latest individual's unupdated position with the fitness value of the new position to obtain a better position, and the better position satisfies the following formula:

[0122]

[0123] in, f(X i,j ) represent the fitness values ​​of the latest position and the unupdated position respectively;

[0124] Said The Cauchy mutation operator satisfies the following formula:

[0125]

[0126] Among them, ub represents the upper boundary, lb represents the lower boundary, rand obeys the random matrix of the standard uniform distribution [0,1], η represents the function generating the random variable of the Cauchy distribution, and ξ represents a random number between 0 and 1; when Less than F i 'When the random number generated by rand is less than the reverse learning rate, the reverse learning rate satisfies the following formula:

[0127] P s =-exp(1-t / iter max ) 20 +θ

[0128] Among them, P s represents the reverse learning rate, and θ represents the adjustment parameter value of 0.08.

[0129] In a specific embodiment, the support vector machine model satisfies the following formula:

[0130]

[0131] The Lagrangian function is introduced as:

[0132]

[0133] The kernel function is expressed as The support vector machine model is expressed as:

[0134]

[0135] Where w represents the normal vector of the hyperplane, C represents the penalty parameter, ξ p represents the slack variable, b represents the bias; M is the total number of samples, p and q represent the pth and qth samples respectively, and m represents the total number of sample pairs; x represents the sample feature vector, y represents the sample label, and α is the Lagrange multiplier; K(x p ,x q ) represents the kernel function, represents the mapping function; σ represents the expansion constant of the radial basis function, and gamma represents the penalty coefficient.

[0136] The embodiment of the present invention also discloses a verification method system for the influence of shale lamination on fracture expansion and conductivity, which is characterized by comprising:

[0137] Sample acquisition module: obtain laminar shale samples and massive shale samples with different physical properties and perform preprocessing;

[0138] Control experiment module: conduct fracturing control experiments under different experimental conditions on pretreated shale samples;

[0139] Detection and analysis module: detect the fracture data of samples after the experiment, analyze the influence of shale lamination on the expansion of fractures; detect the oil distribution of shale samples after the experiment, and analyze the influence of shale lamination on the conductivity of fracturing transformation;

[0140] Prediction module: multiple sample data are obtained according to the experimental data, and then a sample data set is obtained; a support vector machine model is constructed and trained using the sample data set to obtain a shale conductivity prediction model.

[0141] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0142] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may 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 the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for verifying the influence of shale lamination on fracture expansion and conductivity, characterized in that: The specific steps are as follows: Obtain laminar shale samples and massive shale samples with different physical properties and perform pretreatment; The pretreated shale samples were subjected to fracturing control experiments under different experimental conditions; The crack data of the samples after the test were tested, and the influence of shale lamination on the expansion of hydraulic fractures was analyzed; the oil distribution of the shale samples after the test was tested, and the influence of shale lamination on the conductivity of hydraulic fracturing transformation was analyzed; A plurality of sample data are obtained according to the experimental data, and then a sample data set is obtained; a support vector machine model is constructed, and the sample data set is used for training to obtain a shale conductivity prediction model.

2. A method for verifying the influence of shale lamination on fracture expansion and conductivity according to claim 1, characterized in that: The physical properties include: density, grain size, tensile strength, elastic modulus, permeability, laminar spacing, and laminar cohesion strength of laminar shale samples, and density, grain size, tensile strength, elastic modulus, and permeability of massive shale samples; The experimental conditions include: sample loading temperature, sample loading pressure, fracturing fluid pressure, fracturing flow rate, fracturing time, fracturing fluid proppant composition and concentration; The crack data include: maximum crack length, maximum crack width, and maximum crack depth.

3. The method for verifying the influence of shale lamination on fracture expansion and conductivity according to claim 1, characterized in that: The pre-processing comprises: The shale samples are treated with oil washing, and the cores are preliminarily cleaned to remove surface oil and impurities during the coring process; the cores are soaked with cleaning agents to decompose and dissolve the oil in the core pores; the oil in the cores is removed by ultrasonic cleaning; the residual oil and cleaning agents are washed away and then dried; The shale sample after cleaning and drying is subjected to oil saturation treatment, the core is placed in a high-pressure oil saturation device, pressurized to a set pressure at a set rate; the pressure is maintained, and oil saturation injection is performed to allow the experimental oil to penetrate into the core; the core is vented to discharge free gas in the core pores; the pressure is reduced to normal atmospheric pressure at a set rate, and the surface is cleaned to complete the pretreatment.

4. The method for verifying the influence of shale lamination on fracture expansion and conductivity according to claim 1, characterized in that: The fracturing control experiment is specifically as follows: The laminar shale sample and the massive shale sample having the same physical properties are selected, and the first laminar sample, the second laminar sample, the first massive sample, and the second massive sample are obtained by equally dividing the samples into a group of experimental samples; A fracturing experiment is performed on the first laminae sample and the first block sample in the experimental samples, and a displacement treatment is performed; and a displacement treatment is directly performed on the second laminae sample and the second block sample in the experimental samples.

5. A method for verifying the influence of shale lamination on fracture expansion and conductivity according to claim 4, characterized in that: The analysis of the influence of shale lamination on the expansion of hydraulic fractures includes: Respectively detecting the maximum crack length, maximum crack width, and maximum crack depth of the first laminar sample and the first block sample in each group of the experimental samples after the experiment, and calculating the difference; Taking the laminae spacing value and laminae bonding strength value of the first laminae sample as the X and Y axes, and taking the difference of the maximum crack length, the difference of the maximum crack width, and the difference of the maximum crack depth as the Z axis, respectively, the first influence diagram, the second influence diagram, and the third influence diagram are obtained by drawing and fitting, and the influence of shale laminae on the expansion of hydraulic fractures is intuitively analyzed according to the three-dimensional surface of the influence diagram; The analysis of the influence of shale lamination on the conductivity of fracturing transformation includes: After the experiment of each group of the experimental samples, the oil distribution of the first laminae sample, the second laminae sample, the first block sample, and the second block sample is detected respectively; The difference in oil content between the first laminae sample and the second laminae sample is used as the fracturing conductivity of the laminae shale samples in the group of experimental samples, and the difference in oil content between the first block sample and the second block sample is used as the fracturing conductivity of the block shale samples in the group of experimental samples; The fourth influence diagram is drawn with the laminar spacing value and laminar bonding strength value of the laminar shale sample as the X and Y axes, and the difference between the fracturing conductivity of the laminar shale sample and the fracturing conductivity of the massive shale sample as the Z axis, so as to intuitively analyze the influence of shale lamination on the fracturing conductivity.

6. A method for verifying the influence of shale lamination on fracture expansion and conductivity according to claim 5, characterized in that: The oil distribution is determined by detecting and drawing a three-dimensional oil-containing model of the sample through nuclear magnetic resonance. The oil content is calculated as follows: Meshing the oil-bearing three-dimensional model to transform the oil-bearing three-dimensional model into a plurality of independent and connected cubes; marking the cubes that are completely oil-containing and the cubes that are partially oil-containing among the cubes; The oil-containing volume of the full oil-containing cube and the oil-containing volume of the partial oil-containing cube are calculated respectively, and the oil content is obtained by adding them up; The oil-containing volume of the partial oil-containing cube is calculated as follows: Determining the interface, non-interface, and oil body curved surface between the oil body and the cube surface in the partial oil-containing cube; When the number of interfaces is 3, 4 or 5, the oil-containing volume is directly calculated; When the number of interfaces is 6, the non-oil volume is calculated first, and the oil-containing volume is obtained by subtracting the non-oil volume from the cubic volume.

7. A method for verifying the influence of shale lamination on fracture expansion and conductivity according to claim 6, characterized in that: The direct calculation of the oil-containing volume is specifically as follows: Step 1: Generate a triangulated network model according to the key points of the oil body surface; Step 2: Determine the intersection of the normal vectors of the center points of all triangles in the current triangulated network model and the oil body surface; Step 3: Determine three sub-triangles by connecting the intersection point of the triangle with the vertex of the triangle and the sides of the triangle; Step 4: Update the triangulated network model with all the sub-triangles to determine whether the set update times are met, and if not, return to step 2; Step 5: Taking the opposite surface of any non-intersection interface as the bottom surface, projecting all triangles in the triangulated mesh model onto the bottom surface to generate subdivided prisms of the oil body; Step 6: Calculate the volume of each prism and add them together to obtain the oil-containing volume.

8. The method for verifying the influence of shale lamination on fracture expansion and conductivity according to claim 1, characterized in that: The support vector machine model is constructed, comprising: Set the first strategy, the second strategy, and the third strategy as optimization strategies; Setting a dynamic weight factor, optimizing individual update positions by the dynamic weight factor, and obtaining a dynamic adaptive weight population algorithm; Combining the dynamic adaptive weighted population algorithm and the optimization strategy to obtain an improved population algorithm; A support vector machine model is constructed based on the improved population algorithm, and the kernel function and penalty coefficient of the support vector machine model are corrected by the improved population algorithm.

9. A method for verifying the influence of shale lamination on fracture expansion and conductivity according to claim 8, characterized in that: The first strategy, the second strategy, and the third strategy are respectively the firefly perturbation strategy, the Cauchy mutation strategy, and the reverse learning strategy; The dynamic weight factor formula is: Among them, ω represents the weight factor, e is the base of the natural logarithm, t represents the current number of iterations, iter max Indicates the maximum number of iterations; The dynamic adaptive weighted population algorithm satisfies the following formula: in, represents the new position of the jth dimension of the ith individual, X i,j represents the current position of the i-th individual, p i,j represents the prey position of the i-th individual, I represents a random integer of 1 or 2, The objective function value F represents the prey position of the i-th individual i represents the fitness value of the ith individual and F represents the objective function vector of the population. Represents the objective function value of the i-th individual after updating based on the first stage; The improved population algorithm satisfies the following formula: in, Indicates the population's latest position update parameter, X′ i,j represents the inverse solution of the optimal solution of the tth iteration, b1 represents the information exchange control coefficient, F i t+1 Indicates the fitness of the latest individual, F i ′ represents the firefly position fitness value of the i-th individual, Cauchy(0,1) satisfies the standard Cauchy distribution, and Cauchy represents F i t+1 The Cauchy mutation operator of .

10. A verification method system for the effect of shale lamination on fracture expansion and conductivity, characterized in that: include: Sample acquisition module: obtain laminar shale samples and massive shale samples with different physical properties and perform preprocessing; Control experiment module: conduct fracturing control experiments under different experimental conditions on pretreated shale samples; Detection and analysis module: detect the fracture data of samples after the experiment, analyze the influence of shale lamination on the expansion of fractures; detect the oil distribution of the shale samples after the experiment, and analyze the influence of shale lamination on the conductivity of fracturing transformation; Prediction module: multiple sample data are obtained according to the experimental data, and then a sample data set is obtained; a support vector machine model is constructed, and the sample data set is used for training to obtain a shale conductivity prediction model.

Images