A 3D modeling method of dominant seepage channel based on multi-scale fusion

Abstract

The present invention discloses a multi-scale fused three-dimensional modeling method for advantageous seepage channels, which belongs to the field of oil and gas field development. Based on the vertical high-precision characteristics of well logging data, a comprehensive evaluation index capable of identifying and evaluating advantageous seepage channels formed by long-term water injection development of sandstone oil layers is constructed, and the lateral continuity characteristics of seismic data are utilized to extract geological structure tensors that can reflect the characteristics of the formation, and then the elliptic partial differential equation interpolation method is used to form a three-dimensional model of advantageous seepage channels, providing a basis for the efficient development of high water-cut oil reservoirs. The present invention has high precision and wide applicability, and solves the problem of existing methods for identifying and evaluating regional advantageous seepage channels, and provides a technical basis for the next step of oil reservoir water injection development adjustment and residual oil potential tapping.

Description

Technical Field

[0001] The present invention relates to the technical field of oil and gas geological exploration, and in particular to a three-dimensional modeling method for preferential seepage channels with multi-scale fusion. Background Art

[0002] Preferential seepage channels refer to interconnected high-permeability intervals formed in sandstone reservoirs during long-term water injection development. Their existence can cause ineffective circulation of injected water, resulting in reduced oil recovery rate of the oilfield, increased production cost, and decreased development efficiency. Therefore, based on reasonably determining the development of preferential seepage intervals in a single well, establishing a data model that can accurately reflect the distribution of preferential seepage channels and visually presenting it with relevant software can provide a basis for the next step of plugging and profile control, and contribute to the efficient development of reservoirs in the new stage of old oilfields.

[0003] Currently, the commonly used research methods for preferential seepage channels are mainly based on dynamic data, such as inter-well dynamic monitoring, well testing technology, etc. Logging only provides static pore permeability parameters of a single well, and finally forms a qualitative or semi-quantitative evaluation of preferential seepage channels in a small area on the plane. The above methods are obviously not applicable to the evaluation of regional preferential seepage channels, and do not utilize the advantages of the basic and high-precision nature of logging data. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a three-dimensional modeling method for preferential seepage channels with multi-scale fusion. On the basis of fully utilizing core experiment data, a comprehensive evaluation method for preferential seepage intervals based on logging data is constructed, combined with a spatial modeling method for logging reservoir parameters with seismic constraints. For the first time, a method of multi-scale fusion of core, logging, and seismic data is used to realize high-precision three-dimensional spatial modeling of preferential seepage channels.

[0005] To solve the above technical problem, the technical solution adopted by the present invention is: a three-dimensional modeling method for preferential seepage channels with multi-scale fusion, including the following steps:

[0006] Step 1, according to core experiment data and logging data, calculate the microscopic pore structure coefficient ε of the sandstone reservoir based on logging data through multiple regression

[0007] ε = -53.3016GRB + 118.2481CNB - 54.0836DENB + 50.20352. In the formula, ε is the microscopic pore structure coefficient of the sandstone reservoir; GRB, CNB, and DENB are the normalized output values of the natural gamma ray, compensated neutron, and compensated density of the conventional logging curves, respectively;

[0008] Step 2, combining logging geological theory, calculate the macroscopic seepage homogeneity coefficient ρ that can effectively identify preferential seepage intervals through core experiment data analysis

[0009]

[0010] Where ρ is the seepage homogeneity coefficient; V k , T k , J k are the coefficient of variation of permeability, the coefficient of permeability breakthrough, and the permeability gradient, respectively;

[0011] Step 3: Comprehensively construct the evaluation index Sd of the dominant seepage section

[0012] Sd = 0.75×ε + 0.25×ρ

[0013] Step 4: Using the structure tensor extracted from seismic data as a constraint framework, and applying the elliptic equation interpolation method, conduct 3D modeling on the comprehensive evaluation index Sd of the dominant seepage section constructed from well logging data.

[0014] Specifically, Step 4 includes:

[0015] ① According to the seismic data in the depth domain, extract the geological structure tensor data volume that can reflect the spatial variation characteristics of the formation. The structure tensor of the formation satisfies the following formula:

[0016]

[0017] Where D(x) is the geological structure tensor, x is the three-dimensional data of the target point in the formation, s is the scaling scale coefficient, C -1 (x) is the inverse matrix of C(x), and a(x) is the structure coefficient;

[0018] ② Based on the comprehensive evaluation index Sd of the dominant seepage section and the geological structure tensor data, use the elliptic partial differential equation algorithm to perform spatial domain data interpolation and construct a 3D model of the dominant seepage section;

[0019] The elliptic partial differential equation satisfies the following formula:

[0020]

[0021] Where x is the three-dimensional data of the target point in the formation, v(x) is the geological data of the target point, and v i is the geological data of the discrete point closest to the target point; represents the gradient of the vector v(x), g(x) is the minimum geological distance, represents the gradient of the vector g(x), represents the divergence of the vector.

[0022] In Step 2, according to the permeability values in the well logging data, the coefficient of variation of permeability, the coefficient of permeability breakthrough, and the permeability gradient are calculated layer by layer.

[0023] The beneficial effects of the present invention are as follows: It can quantitatively display the development degree of preferential seepage channels in the formation in three dimensions, provide a basis for water injection profile control during the development of old oilfields in the high water cut stage, and thus improve the development efficiency of the reservoir. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a flowchart of the three-dimensional modeling method for preferential seepage channels with multi-scale fusion of the present invention.

[0025] Figure 2 It is a map of the preferential seepage index of well logging discrete points.

[0026] Figure 3 It is a geological structure tensor map extracted from seismic data.

[0027] Figure 4 It is a three-dimensional model of the preferential seepage layer section.

[0028] Figure 5 It is a map of the calculation results of the comprehensive evaluation index of the preferential seepage layer section of a single well. DETAILED DESCRIPTION OF THE INVENTION

[0029] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0030] The three-dimensional modeling method for preferential seepage channels with multi-scale fusion of the present invention includes the following steps:

[0031] Step 1: According to the core experiment data and well logging data, through multiple regression, calculate the microscopic pore structure coefficient ε of the sandstone reservoir based on well logging data

[0032] ε = -53.3016GRB + 118.2481CNB - 54.0836DENB + 50.20352. In the formula, ε is the microscopic pore structure coefficient of the sandstone reservoir; GRB, CNB, and DENB are the normalized output values of the conventional well logging curves natural gamma ray, compensated neutron, and compensated density, respectively;

[0033] Step 2: Combining well logging geological theory, through the analysis of core experiment data, calculate the macroscopic seepage homogeneity coefficient ρ that can effectively identify the preferential seepage layer section

[0034]

[0035] In the formula, ρ is the seepage homogeneity coefficient; V k , T k , Jk They are permeability variation coefficient, permeability breakthrough coefficient and permeability difference respectively;

[0036] Step 3: Comprehensively construct the evaluation index Sd of the dominant seepage layer

[0037] Sd=0.75×ε+0.25×ρ

[0038] Step 4: Using the structural tensor extracted from the seismic data as a constraint framework and the elliptic equation interpolation method, a three-dimensional model is constructed for the comprehensive evaluation index Sd of the dominant seepage layer constructed in the logging data.

[0039] Specifically, step 4 includes:

[0040] ①According to the depth domain seismic data, a geological structure tensor data volume that can reflect the spatial variation characteristics of the stratum is extracted, and the structure tensor of the stratum satisfies the following formula:

[0041]

[0042] Wherein, D(x) is the geological structure tensor, x is the three-dimensional data of the target point in the stratum, s is the expansion scale coefficient, C -1 (x) is the inverse matrix of C(x), a(x) is the structural coefficient;

[0043] ②Based on the comprehensive evaluation index Sd of the dominant seepage layer and the geological structure tensor data, the elliptic partial differential equation algorithm is used to interpolate the spatial domain data and construct a three-dimensional model of the dominant seepage layer;

[0044] The elliptic partial differential equation satisfies the following formula:

[0045]

[0046] Where x is the three-dimensional data of the target point in the stratum, v(x) is the geological data of the target point, v i is the geological data of the discrete point closest to the target point; represents the gradient of vector v(x), g(x) is the minimum geological distance, represents the gradient of the vector g(x), represents the divergence of a vector.

[0047] In step 2, the permeability variation coefficient, permeability breakthrough coefficient and permeability difference are obtained in layers according to the permeability values ​​in the logging data.

[0048] Based on the longitudinal high-precision characteristics of logging data, the present invention constructs a comprehensive evaluation index capable of identifying and evaluating the preferential flow channels formed by long-term water injection development in sandstone oil reservoirs, and utilizes the lateral continuity characteristics of seismic data to extract the geological structure tensor that can reflect the formation characteristics. Then, through the elliptic partial differential equation interpolation method, a three-dimensional model of the preferential flow channels is formed, providing a basis for the efficient development of oil reservoirs in the high water cut period. The three-dimensional modeling method of the preferential flow channels of the present invention has high precision and wide applicability, solves the problem of the identification and evaluation of regional preferential flow channels in existing methods, and provides a technical basis for the subsequent adjustment of oil reservoir water injection development and the tapping of remaining oil. The three-dimensional modeling method of the preferential flow channels of the present invention is mainly aimed at oil reservoir blocks with long-term water injection development, and generally requires a certain number of oil wells and corresponding logging data. The more the number of wells, the stronger the regularity shown by the three-dimensional model.

[0049] The following further describes the present invention in conjunction with the accompanying drawings and specific embodiments:

[0050] (1) Taking 3 wells in the study area as an example, the process of calculating and obtaining the comprehensive evaluation index of the preferential flow layer section of the sandstone formation from logging data is described.

[0051] First, the logging data of each well should at least include the measured natural gamma ray GR, compensated neutron CN, and compensated density DEN curves. Generally, in order to eliminate the measurement errors of different instruments and the order of magnitude differences of different curve values, the logging curves are standardized, so as to ensure the consistency of data in subsequent data modeling. The curve values GRB, CNB, and DENB used here are the standardized curve values.

[0052] Secondly, substituting the above standardized curve values into the formula in step 1, the microscopic pore structure coefficient ε of the vertically continuous sandstone reservoir can be obtained.

[0053] Then, according to the permeability curve K in the processing results of the logging data of each well, the coefficient of variation Vk, the breakthrough coefficient Tk, and the range Jk are calculated for each layer of the sandstone reservoir. These coefficients can be used to quantitatively describe the heterogeneity of the sandstone reservoir. It should be noted in this step that for the calculation of the permeability of the sandstone reservoir, the conventional calculation method is to use Darcy's formula to obtain it through the correlation with porosity. This algorithm often ignores the influence of rock mineral composition and pore connectivity, and the calculation accuracy of permeability is relatively low. Therefore, when there are a large number of nuclear magnetic resonance logging data in the logging data, the permeability parameters calculated by nuclear magnetic resonance logging can be preferentially considered. According to the calculated permeability coefficients, substituting them into the formula in step 2, the macroscopic flow homogeneity coefficient ρ of each sandstone reservoir is obtained.

[0054] Through the above calculation process, the microscopic pore structure coefficient ε and the macroscopic seepage homogeneity coefficient ρ of each set of sandstone reservoirs in each well are obtained. That is, the parameters for measuring the development degree of the preferential seepage section in the sandstone reservoir from both microscopic and macroscopic perspectives are obtained. Finally, these two parameters are substituted into the formula in Step 3 to calculate the comprehensive evaluation index of the preferential seepage section. After processing all the wells in the study area through the above steps, a longitudinal continuous Sd curve will be formed in the data of each well, which is the data basis for the next three-dimensional modeling.

[0055] Figure 5 It is the calculation result of the comprehensive evaluation index of the preferential seepage section of a single well in the study area. The figure includes GR, CN, DEN curves and the permeability curve. During the processing, the first three curves are first standardized and then substituted into the formula in Step 1 to calculate the microscopic pore structure coefficient ε. Then, according to the permeability curve, the variation coefficient Vk, the breakthrough coefficient Tk and the differential range Jk of each sandstone reservoir are calculated, and then substituted into the formula in Step 2 to calculate the macroscopic seepage homogeneity coefficient ρ. Finally, the comprehensive evaluation index Sd of the preferential seepage section is calculated using the formula in Step 3. Figure 5 The 7th trace in it is the calculation result of the comprehensive evaluation index of the preferential seepage section, including Sd calculated from core experiment data, Sd calculated from conventional logging data, and Sd calculated from nuclear magnetic resonance logging data. Usually, the calculation result from core experiment data is taken as the standard. From the above calculation results, it can be seen that the Sd result calculated from logging data is highly consistent with the core data result. Among them, the calculation result of nuclear magnetic resonance logging has a higher degree of coincidence with the core data than that of conventional logging data. This also shows that in the actual processing process, nuclear magnetic resonance logging data can be preferably selected to improve the calculation accuracy of the comprehensive evaluation index of the preferential seepage section.

[0056] Three-dimensional data modeling generally forms a spatially continuous data volume through mathematical interpolation such as Kriging using spatially discrete point data. This traditional interpolation method is an isotropic uniform interpolation, and only the average value of two points is considered during the process. The multi-scale fusion three-dimensional modeling method implemented in the present invention utilizes the spatial continuity of seismic data. Through the proposal of the geological structure tensor, the factors that can reflect geological characteristics are transformed into a quantifiable parameter, and then the interpolation method of the elliptic partial differential equation is introduced. Starting from the logging discrete data (Sd) and with the geological structure tensor as the constraint, spatial interpolation is carried out, and finally a three-dimensional data volume that integrates the data characteristics of different scales of core, logging and seismic is formed, which can quantitatively display the development degree of the preferential seepage channels in the formation three-dimensionally. The specific implementation is as follows:

[0057] ①Convert the discrete well logging data points into a spatial domain data volume. The implementation of this step is to determine the relationship between the discrete well logging data points in the depth domain and the three-dimensional spatial data, spatialize the well logging data at each point, assign it a coordinate position, and then determine the spatial domain data volume of each discrete point.

[0058] ②Extract the geological structure tensor from the depth domain seismic data. The geological structure tensor of the formation is used to reflect the structural continuity of the formation. The structure tensor of the formation satisfies the following formula:

[0059]

[0060] Among them, C(x) is the structure tensor, x is the three-dimensional data of the target point in the formation, Ix is the horizontal gradient of the target point in the depth domain seismic data, Iy is the vertical gradient of the target point in the depth domain seismic data, and Ixy is the gradient of the target point in the depth domain seismic data in the direction of the horizontal and vertical connection line.

[0061] It should be noted that Ix, Iy, and Ixy are all obtained through image processing.

[0062] The geological structure tensor of the formation satisfies the following formula:

[0063]

[0064] Among them, D(x) is the geological structure tensor, x is the three-dimensional data of the target point in the formation, s is the dilation scale coefficient, C-1(x) is the inverse matrix of C(x), and a(x) is the structure coefficient.

[0065] In the above implementation, since the geological structure tensor can reflect the spatial continuity change information of the formation, the spatial continuity change of the formation can be reflected by the obtained geological structure tensor, which is convenient for subsequent constraint of the elliptic partial differential equation. It should be noted that the geological structure tensor at any point in the formation can be regarded as a set of parameters reflecting the spatial heterogeneity change of the formation, which contains the spatial change information of the geological image.

[0066] It should be noted that the numerical range of a(x) is [0,1], which describes the relationship between the geological tensor field and the credibility of geological structure information. Both s and a(x) are fixed constants. Generally speaking, at places where the geological horizon continuity is good, the value of a(x) is large; while at places where the geological horizon continuity is poor, the value of a(x) is small. In addition, the numerical range of C(x) is [0,1].

[0067] ③Based on the well logging spatial domain data volume and the geological structure tensor, use the elliptic partial differential equation for interpolation to construct a three-dimensional data model.

[0068] The elliptic partial differential equation satisfies the following formula:

[0069]

[0070] Wherein, x is the three-dimensional data of the target point in the formation, v(x) is the geological data of the target point, and vi is the geological data of the discrete point closest to the target point; represents the gradient of the vector v(x), and g(x) is the minimum geological distance, represents the gradient of the vector g(x), represents the divergence of the vector.

[0071] In the above interpolation process, the Sd values of discrete points can be quickly formed into a three-dimensional data model with spatial continuity under the constraint of the geological structure tensor, and then the development of the dominant seepage channels in the formation of the study area can be visually displayed.

[0072] Figure 2 This is the spatial domain data volume of the comprehensive evaluation index Sd of the logging dominant seepage intervals of 62 wells in the study area targeted by the present invention. Figure 3 It is a schematic diagram of the geological result tensor extracted from seismic data. Figure 4 It is a three-dimensional model of the dominant seepage channels formed after interpolation processing by the elliptic partial differential equation. The dark areas in the figure represent a higher degree of development of the dominant seepage intervals. Therefore, this three-dimensional model can quickly and effectively display the development degree of the dominant seepage channels in the reservoir under long-term water injection development, providing a basis for the next step of plugging and profile control and improving development efficiency.

[0073] In summary, the content of the present invention is not limited to the above embodiments. Those skilled in the same field can easily propose other embodiments within the technical guiding ideology of the present invention, but such embodiments are all included within the scope of the present invention.

Claims

1. A multi-scale fusion dominant seepage channel three-dimensional modeling method, characterized in that: The following steps are involved: Step 1: According to the core experimental data and logging data, the microscopic pore structure coefficient ε of the sandstone reservoir based on the logging data is calculated by multivariate regression. ε=-53.3016GRB+118.2481CNB-54.0836DENB+50.20352, where ε is the microscopic pore structure coefficient of the sandstone reservoir; GRB, CNB and DENB are the standardized output values ​​of natural gamma, compensated neutron and compensated density of conventional logging curves respectively; Step 2: Combined with logging geological theory, through core experimental data analysis, calculate the macroscopic seepage homogeneity coefficient ρ that can effectively identify the dominant seepage layer segment Where, ρ is the homogeneity coefficient of seepage; V k , T k , J k They are permeability variation coefficient, permeability breakthrough coefficient and permeability difference respectively; Step 3: Comprehensively construct the evaluation index Sd of the dominant seepage layer segment Sd=0.75×ε+0.25×ρ Step 4: Using the structural tensor extracted from the seismic data as a constraint framework and the elliptic equation interpolation method, a three-dimensional model is constructed for the comprehensive evaluation index Sd of the dominant seepage layer constructed in the logging data.

2. According to the multi-scale fusion dominant seepage channel three-dimensional modeling method of claim 1, it is characterized in that: Step 4 includes: ① According to the depth domain seismic data, a geological structure tensor data volume that can reflect the spatial variation characteristics of the stratum is extracted, and the structure tensor of the stratum satisfies the following formula: Where D(x) is the geological structure tensor, x is the three-dimensional data of the target point in the stratum, s is the expansion scale coefficient, C -1 (x) is the inverse matrix of C(x), a(x) is the structural coefficient; ② Based on the comprehensive evaluation index Sd of the dominant seepage layer and the geological structure tensor data, the elliptic partial differential equation algorithm is used to interpolate the spatial domain data and construct a three-dimensional model of the dominant seepage layer; The elliptic partial differential equation satisfies the following formula: Where x is the three-dimensional data of the target point in the stratum, v(x) is the geological data of the target point, and v i The geological data of the discrete point closest to the target point; represents the gradient of vector v(x), g(x) is the minimum geological distance, represents the gradient of the vector g(x), Represents the divergence of a vector.

3. According to the multi-scale fusion dominant seepage channel three-dimensional modeling method of claim 1, it is characterized in that: In the step 2, the permeability variation coefficient, the permeability breakthrough coefficient and the permeability difference are obtained in layers according to the permeability values ​​in the logging data.

Images