Quantitative characterization and prediction method for heterogeneity of marine clastic rock stratum

Through machine learning algorithms and core principal component analysis, the comprehensive characterization factor of heterogeneity of reservoirs was calculated, which solved the shortcomings of quantitative calculation of heterogeneity of marine clastic rock reservoirs in the Cretaceous Napo Formation in the Orente Basin of South America, and achieved The fine quantitative characterization of reservoir characteristics and the improvement of oil and gas recovery rate were carried out.

CN119001875AInactive Publication Date: 2024-11-22SOUTHWEST PETROLEUM UNIV

Patent Information

Application Number
CN202410901443.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-05
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

There is insufficient quantitative calculation of heterogeneity characteristics of marine clastic rock reservoirs in the Cretaceous Napo Formation in the Orient Basin in South America, resulting in unclear fluid distribution rules and unclear single well pressure drop patterns, making it difficult to accurately lay wells and lift oil and gas Recovery rate.

Method used

By acquiring core and logging data, microscopic heterogeneity characteristics such as core particle size standard deviation, flow band index and pore throat radius are calculated, and dimensionality reduction fusion of multivariate information is combined with core principal component analysis (KPCA). Form a comprehensive heterogeneous characterization factor for reverse reconstruction of sensitive factor curves and improve quantitative characterization of reservoir heterogeneous characteristics.

Benefits of technology

The fine quantitative characterization of the heterogeneity characteristics of marine clastic rock reservoirs is realized, the reservoir description accuracy and oil and gas recovery are improved, and it is suitable for complex and changeable oil field environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119001875A_ABST
    Figure CN119001875A_ABST
Patent Text Reader

Abstract

The invention discloses a quantitative characterization and prediction method for heterogeneity of a marine clastic rock stratum. Under the comprehensive influence of sea level rising and falling and material sources, a series of prominent problems such as uneven granularity of clastic rock sediments, complex rock structure, different qualities of sand bodies in reservoirs, unclear reservoir fluid distribution rules and unknown single well pressure drop modes caused by strong heterogeneity characteristics are solved; and the next step of accurate well spacing, water injection development potential evaluation and recoverable reserve improvement are not facilitated. Therefore, aiming at the defects of the existing characterization method, the invention provides a new method for extracting a multivariate heterogeneous characteristic sensitive factor by using core experiment data and logging data and fusing and characterizing the heterogeneity of the reservoir based on a KPCA information dimension reduction method. Practice proves that the method can be used for finely representing the heterogeneous characteristics of the high-heterogeneity clastic rock reservoir, the result calculated by using the new method is matched with the actual production of the oil reservoir, and the method has wide application and popularization prospects.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the fields of oil and gas reservoir exploration and development and machine learning, and in particular to a method for quantitatively characterizing and predicting heterogeneity of marine clastic rock formations. Background Art

[0002] In recent years, there have been major exploration discoveries in the Cretaceous Napo Formation of the P oilfield in the Oriente Basin in South America. The source-reservoir configuration relationship is good and favorable reservoir facies belts are developed, laying the foundation for high-quality reservoirs. At present, the block has initially formed a high-yield well deployment model and achieved good results. However, due to the frequent and irregular changes in sea level and the influence of the Brazilian Shield source, the reservoir sediments have uneven grain size, complex rock components, and different sand bodies in the reservoir. The strong heterogeneity characteristics have led to a series of prominent problems such as unclear reservoir fluid distribution and unclear single well pressure drop patterns, which are not conducive to the next step of accurate well layout, water injection development potential evaluation and increase in recoverable reserves. It is urgent to improve the quantitative characterization method of reservoir heterogeneity and implement the distribution law of favorable reservoir areas and remaining oil. However, the current quantitative characterization methods of heterogeneity are mostly concentrated in continental clastic rocks and shale reservoirs, and the flow unit mathematical method for characterizing heterogeneity characteristics does not meet the requirements of "minimum difference in reservoir properties within the flow unit and maximum difference in reservoir properties between different flow units". Therefore, the present invention is mainly based on core experimental data and conventional logging data, and makes appropriate improvements to the shortcomings of the existing methods. The improved method is used to calculate the sensitive factors of reservoir heterogeneity characteristics. On this basis, the multivariate information is integrated and characterized based on kernel principal component analysis (KPCA), providing technical support for the next step of accurate well layout, oil well productivity evaluation, remaining oil favorable zone prediction, and water injection development potential evaluation.

[0003] Reservoir heterogeneity refers to the fact that during the long geological evolution of oil and gas reservoirs, the physical properties of reservoirs and rocks at different scales have extremely uneven changes due to the combined influence of sedimentation, diagenesis and structure. As an important part of reservoir description and evaluation, heterogeneity characteristics not only need to fully consider the geological factors that affect fluid seepage inside the reservoir, but also apply these factors to conduct in-depth evaluation and research on the reservoir. Therefore, the detailed characterization of reservoir heterogeneity is of great significance and applicable value for further subdividing reservoirs, exploring wells and improving oil and gas recovery. Therefore, since the concept of detailed reservoir description was proposed, geologists have proposed many characterization methods for the description of heterogeneity characteristics. They can be roughly summarized into the following four categories: (1) Heterogeneous characterization methods based on geological theory (CN108008464A (Zhang Zhaohui et al., 2018), CN107144566B (Liu Keyu et al., 2017), CN114076991A (He Faqi et al., 2022)); (2) Heterogeneous characterization methods based on core experimental analysis data (CN110702580A (Dong Fengjuan et al., 2020), CN103353462 (Ge Xinmin et al., 2013), CN117649036A (Song Benbiao et al., 2024)); (3) Multi-parameter heterogeneous characterization methods based on mathematical methods (CN110242291A (Liu Jinhua et al., 2019), CN108241785A (Yang Xipu et al., 2018)).

[0004] The advantage of the heterogeneous characterization method based on geological theory is that it has sufficient theoretical basis, but its disadvantage is that it can only roughly characterize the heterogeneous characteristics on a macro scale. The accuracy of the division of strata with strong heterogeneity is not high, and it is difficult to meet the needs of fine description of oil reservoirs. The advantage of the method of analyzing data division based on cores is that it uses reliable data core experimental data, the heterogeneity characterization basis is sufficient, and the characterization results are accurate, but its disadvantages are also obvious. Due to the constraints of coring costs, drilling engineering technology and other factors, the current oil fields cannot meet the experimental requirements of continuous coring of a large number of wells. Therefore, it is impossible to carry out extensive application research in the oil field area, resulting in low accuracy of heterogeneous characterization at the regional scale. The multi-parameter heterogeneous characterization method based on mathematical methods combines the more popular artificial intelligence and machine learning methods today, bringing new advantages to the cross-integration of multidisciplinary and multi-scale information. The research methods and technologies of heterogeneity have developed rapidly from early qualitative descriptions to quantitative, refined, and intelligent directions, and have been verified by the actual production of oil fields. Its accuracy is higher than the first two categories and is more suitable for complex and changing environments.

[0005] Therefore, taking the highly heterogeneous Napo Formation T sandstone reservoir in the P oil field in Ecuador as an example, based on the existing core experiments and conventional logging data of the reservoir, combined with the current popular machine learning algorithms, this paper proposes a quantitative characterization method for heterogeneous characteristics based on machine learning. Summary of the invention

[0006] In order to overcome some shortcomings of the above-mentioned heterogeneous characterization method in practical application, the purpose of the present invention is to provide a method for quantitative characterization of heterogeneous characteristics by means of a machine learning algorithm to solve the problem of quantitative calculation of heterogeneous characteristics of marine clastic reservoirs similar to the Cretaceous Napo Formation in the Oriente Basin in South America.

[0007] Therefore, the technical solution of the present application discloses a method for predicting the heterogeneity of marine clastic rock formations by quantitative characterization agents, comprising:

[0008] Obtain the microscopic heterogeneity characteristics of the target well, determine the target well core particle size standard deviation σ, core flow zone index FZI, Por, Perm and the pore throat radius data R corresponding to the target well when the non-wetting phase saturation is 35% 35 ;

[0009] The core flow zone index FZI is improved by using the rock electrical parameter m to obtain FZIm;

[0010] Obtain the macroscopic heterogeneity characteristics of the target well, analyze the factors affecting the physical properties, and use the optimization model to construct Face, Glau, Kaot, and Cal respectively;

[0011] KPCA is used to analyze σ and R 35 , FZIm, Face, Glau, Kaot, Cal, Por and Perm are fused by dimensionality reduction to obtain the heterogeneity comprehensive characterization factor HCF;

[0012] HCF is used to reversely reconstruct the sensitivity factor curve and test its effect. If the effect is good, the HCF parameters will be applied to the entire area to obtain the heterogeneous quantitative characterization results of each well in the study area; if the effect is not good, it will return to the KPCA dimension reduction step and re-integrate the sensitive factor dimension reduction to form a new heterogeneous comprehensive characterization factor until the comprehensive characterization factor can better reconstruct the sensitivity factor curves, that is, the comprehensive characterization factor formed by dimension reduction fusion can represent the important characteristics of each sensitive factor.

[0013] Furthermore, the target well microscopic heterogeneity characteristic parameters include σ, FZI, R 35 , Por and Perm, and the required relevant data are obtained by experimental tests, including core mercury injection, grain size analysis and core physical properties.

[0014] Furthermore, the standard deviation σ of the core particle size of the target well is calculated using the standard deviation formula proposed by Falk and Ward, and the calculation formula is:

[0015]

[0016] In the formula, φ 84 -φ 16 is the concentrated part of rock grain size, φ 95 -φ 5 They respectively represent the particle size sorting of the coarse tail and fine tail, which are most sensitive to hydrodynamics.

[0017] Furthermore, the calculation formula of the core flow zone index FZI is:

[0018]

[0019] Among them, Φ e is the core porosity; K is the core permeability, mD; Φz is the ratio of pore volume to particle volume, and RQI is the reservoir quality index.

[0020] Furthermore, the core flow zone index FZI is improved by using the rock electrical parameter m to obtain the FZIm formula:

[0021] LqCy m =LgRQI-LgΦ Z -(m-1)LgΦ e ;

[0022] Among them, RQI is the reservoir quality index, Φz is the ratio of pore volume to particle volume, and m is the rock electrical parameter.

[0023] Furthermore, the kernel principal component analysis dimensionality reduction fusion method is obtained by the following expression:

[0024]

[0025] In the formula, is a D×D dimensional covariance matrix; X is a K×N dimensional sensitive factor matrix; η is a nonlinear mapping; η(X) is a D×N dimensional matrix; k(x, y) is a radial basis function kernel; p is a weight vector; λ is an eigenvalue; α is an eigenvector; ζ is an eigenvalue threshold condition.

[0026] Furthermore, the reverse reconstruction of the sensitivity curve is specifically as follows: using the kernel principal component analysis algorithm to learn the HCF curve, so as to predict each sensitive factor.

[0027] Compared with the prior art, the present invention has at least the following beneficial technical effects:

[0028] The present invention calculates R 35, FZI, σ, Por and Perm are used as sensitive factors to characterize the microscopic heterogeneity of the reservoir. Considering the limitations of FZI, the influencing factors such as lithology and cementation degree are combined with FZI, and the FZI parameters are improved to FZIm based on the influencing factors. Based on the main controlling factors of reservoir physical properties and conventional logging curves, Face, Glau, Kaot and Cal are constructed as sensitive factors of macroscopic heterogeneity using the optimization model. On this basis, kernel principal component analysis (KPCA) is used to analyze Face, Kaot, Glau, Cal, Por, Perm, FZIm, R 35 The 9 parameters such as σ and σ are simplified by dimensionality reduction to form the heterogeneity comprehensive characterization factor HCF. The HCF is used to reversely reconstruct the sensitive factor curve to verify the excellence of the KPCA dimensionality reduction fusion model. After meeting the accuracy requirements, it is promoted and applied. Practice has proved that the HCF formed by the dimensionality reduction and fusion characterization of multivariate sensitive factors by KPCA conforms to the regional geological laws and production practices, and has a wide range of applications and promotion prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is the technical flow chart of this invention;

[0030] Figure 2 This is the sensitivity factor parameter curve of the T segment of the Napo Formation in the Oriente Basin. DETAILED DESCRIPTION

[0031] The present invention is described in detail below in conjunction with the embodiments and drawings, but it should be understood that the embodiments and drawings are only used to exemplify the present invention and do not constitute any limitation on the protection scope of the present invention. All reasonable changes and combinations within the scope of the inventive concept of the present invention fall within the protection scope of the present invention.

[0032] In the description of the present invention, it should be noted that the structures, proportions, sizes, etc. shown in this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the conditions under which the present invention can be implemented. Any structural modification, change in proportional relationship or adjustment of size should still fall within the scope of the technical content disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0033] At the same time, the terms such as "center", "lateral", "upper", "lower" and "one" cited in this specification are only for the convenience of description and are not used to limit the scope of implementation of the present invention. Changes or adjustments in their relative relationships should be regarded as the scope of implementation of the present invention without substantially changing the technical content.

[0034] The specific implementation of the present invention is further described below by taking the Cretaceous Napo Formation T sandstone reservoir of the P oil field in the Oriente Basin in South America as an example.

[0035] It should be noted that the corresponding data related to the present application for the target well in the Cretaceous Napo Formation T sandstone reservoir of the P oil field in the Oriente Basin of South America described in this application have been tested and obtained, including core physical property experimental data after depth regression of the core data (the data include core porosity Φe / CPor, core permeability parameter CPerm, etc.), mercury injection experimental data (mercury injection experimental parameters are the numerical values ​​of the water saturation change of the rock under different displacement pressures) and particle size analysis experiments (the experiment can obtain relevant data for calculating the standard deviation).

[0036] like Figure 1 As shown, the method for quantitatively characterizing and predicting heterogeneity of marine clastic rock formations described in this embodiment includes:

[0037] 1. Obtain the microscopic heterogeneity characteristics of the target well, determine the core particle size standard deviation σ, core flow zone index FZI, Por, Perm and the pore throat radius data R corresponding to the target well when the non-wetting phase saturation is 35% 35 ; The core flow zone index FZI is improved by using the rock electrical parameter m to obtain FZIm, where Por and Perm are the logging porosity and logging permeability obtained by fine logging interpretation, and are sensitive factors representing reservoir properties.

[0038] In a specific implementation, taking the Cretaceous Napo Formation T sandstone reservoir of the P oil field in the Oriente Basin in South America as an example, the microscopic heterogeneity characteristics include core data such as core mercury injection and core physical properties, and the above core data are deeply relocated to the logging depth, thereby obtaining indicators such as formation characterization lithology, cementation degree, core physical properties, and rock electrical parameters. Its rock electrical parameter m is shown in Table 1.

[0039] Table 1 Rock electrical parameters in the study area

[0040]

[0041] In a further embodiment, the pore throat radius data R corresponding to the non-wetting phase saturation of 35% is 35 The method for obtaining the pore throat radius is as follows: the mercury injection experimental data of the Cretaceous Napo Formation T sandstone reservoir in the P oil field in the Oriente Basin of South America are sorted out to obtain the mercury injection curve, and then the pore throat radius data corresponding to the non-wetting phase saturation of 35% on the mercury injection curve is read, that is, R 35 , as shown in Table 2.

[0042] Table 2 Statistics of some mercury injection parameters

[0043]

[0044] In a further embodiment, the target well core particle size standard deviation σ is calculated using the standard deviation formula proposed by Fock and Ward, and the calculation formula is:

[0045]

[0046] In the formula, φ 84 -φ 16 is the concentrated part of rock grain size, φ 95 -φ 5 They respectively represent the particle size sorting of the coarse tail and fine tail, which are most sensitive to hydrodynamics.

[0047] In a further embodiment, the core flow zone index FZI is calculated as:

[0048]

[0049] Among them, Φ e is the core porosity; K is the core permeability, mD; Φz is the ratio of pore volume to particle volume, and RQI is the reservoir quality index.

[0050] Furthermore, the core flow zone index FZI is improved by using the rock electrical parameter m to obtain the FZIm formula:

[0051] LqCy m =LgRQI-LgΦ Z -(m-1)LgΦ e ;

[0052] Among them, RQI is the reservoir quality index, Φz is the ratio of pore volume to particle volume, and m is the rock electrical parameter. In this embodiment, the purpose of using the rock electrical parameter m to improve the core flow zone index is that FZI has limitations, so this application combines influencing factors such as lithology and cementation degree with FZI, and modifies the FZI parameter to FZIm based on the influencing factors.

[0053] 2. Obtain the macroscopic heterogeneity characteristics of the target well, analyze the factors affecting the physical properties, and use the optimization model to construct Face, Glau, Kaot, and Cal respectively;

[0054] The main influencing factors of the reservoir properties of the Cretaceous Napo Formation T sandstone reservoir in the P oilfield in the Oriente Basin of South America are sedimentary microfacies, glauconite, kaolinite and calcite. The optimization model is used to construct Face, Glau, Kaot and Cal respectively, among which Face is the sensitive factor representing sedimentary microfacies, Glau is the sensitive factor representing glauconite, and Kaot is the sensitive factor representing kaolinite.

[0055] The optimal model of constructing Face, Glau, Kaot, and Cal is as follows Figure 2 shown.

[0056] It should be understood that the optimization model is a commonly used interpretation method in the well logging profession. This method is based on the rock physics model. By optimizing the well logging parameters, the optimal model is established, and different weights are matched to different parameters to obtain the optimal solution of the equation (i.e., the well logging interpretation result). For example, this patent uses the optimization well logging method to calculate the glauconite content (Glau) and kaolinite mineral content (Kaot). The relevant optimization model formula is:

[0057]

[0058] Where GR is the natural gamma curve, GR glau , GR kaot is the natural gamma value corresponding to glauconite and kaolinite, API; V glau 、V kaot is the volume of glauconite and kaolinite, %; Pe glau 、Pe kaot is the photoelectric cross section index value corresponding to glauconite and kaolinite, b / electron; Pe is the photoelectric cross section index, b / electron; ρ is the rock density, g / cm 3 ρ glau , kaot is the density value corresponding to glauconite and kaolinite, g / cm 3 ; Vsh is the mud content, %.

[0059] 3. Use KPCA to analyze σ and R 35 , FZIm, Face, Glau, Kaot, Cal, Por and Perm are fused by dimensionality reduction to obtain the heterogeneity comprehensive characterization factor HCF;

[0060] The dimensionality reduction method is kernel principal component analysis, and its method and code can be obtained from CSDN or Github forum. Its expression is:

[0061]

[0062] In the formula, is a D×D dimensional covariance matrix; X is a K×N dimensional sensitive factor matrix; η is a nonlinear mapping; η(X) is a D×N dimensional matrix; k(x, y) is a radial basis function kernel; p is a weight vector; λ is an eigenvalue; α is an eigenvector; ζ is an eigenvalue threshold condition.

[0063] 4. Use HCF to reversely reconstruct the sensitivity factor curve and test its effect. If the effect is good, the HCF parameters will be applied to the entire area to obtain the heterogeneous quantitative characterization results of each well in the study area; or if the effect is not good, it will return to the KPCA dimension reduction step and re-integrate the sensitive factor dimension reduction to form a new heterogeneous comprehensive characterization factor until the comprehensive characterization factor can better reconstruct the curves of each sensitivity factor, that is, the comprehensive characterization factor formed by dimension reduction fusion can represent the important characteristics of each sensitive factor. Further, the reverse reconstruction of the sensitivity curve is specifically: using the kernel principal component analysis algorithm to learn the HCF curve, so as to predict each sensitive factor. Figure 2 As shown in the figure, the pink color is the sensitivity factor curve reconstructed by the kernel principal component algorithm. Compared with the black original sensitivity factor curve, the reconstructed sensitivity factor curve has a high degree of overlap with the original sensitivity factor curve, indicating good correlation. The dimensionality reduction fusion HCF curve can characterize the important characteristics of each sensitive factor. The method of quantitatively characterizing heterogeneity through HCF can be extended to subsequent wells.

[0064] The present invention calculates R based on the core physical property experimental data and mercury injection experimental data after homing. 35 , FZI, σ, Por and Perm are used as sensitive factors to characterize the microscopic heterogeneity of the reservoir. Considering the limitations of FZI, the influencing factors such as lithology and cementation degree are combined with FZI, and the FZI parameters are improved to FZIm based on the influencing factors. Based on the main controlling factors of reservoir physical properties and conventional logging curves, Face, Glau, Kaot and Cal are constructed as sensitive factors of macroscopic heterogeneity using the optimization model. On this basis, kernel principal component analysis (KPCA) is used to analyze Face, Kaot, Glau, Cal, Por, Perm, FZIm, R 35 The nine parameters such as σ and σ are simplified by dimensionality reduction to form the comprehensive characterization factor HCF of heterogeneity. The sensitive factor curve is reversely reconstructed by HCF to verify the excellence of the KPCA dimensionality reduction fusion model. After meeting the accuracy requirements, it is promoted and applied.

[0065] Practice has proved that the HCF formed by dimensionality reduction and fusion characterization of multivariate sensitive factors using KPCA is in line with regional geological laws and production practices, and has broad application and promotion prospects.

[0066] Taking the core well in the study area as an example, the heterogeneity comprehensive characterization factor HCF obtained by the technical method of the present invention has achieved good results in quantitative evaluation of reservoir heterogeneity characteristics. In the T sandstone reservoir section, the HCF curve shows an obvious "box-shaped" high value in the high-quality reservoir section. For the interlayers inside the sand body, HCF also has a sensitive low-value feature. The change in HCF value can clearly indicate the spatial configuration relationship between the high-quality reservoir and the interlayer, as shown in Figure 2. Figure 2 .

[0067] The above embodiments are only preferred implementations of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A method for predicting the heterogeneity of marine clastic rock formations by quantitative characterization agents, characterized in that: include: Obtain the microscopic heterogeneity characteristics of the target well, determine the target well core particle size standard deviation σ, core flow zone index FZI, Por, Perm, and the pore throat radius data R corresponding to the target well when the non-wetting phase saturation is 35% 35 ; The core flow zone index FZI is improved by using the rock electrical parameter m to obtain FZIm; Obtain the macroscopic heterogeneity characteristics of the target well, analyze the factors affecting the physical properties, and use the optimization model to construct Face, Glau, Kaot, and Cal respectively; KPCA is used to analyze σ and R 35 , FZIm, Face, Glau, Kaot, Cal, Por and Perm are fused by dimensionality reduction to obtain the heterogeneity comprehensive characterization factor HCF; HCF is used to reversely reconstruct the sensitivity factor curve and test its effect. If the effect is good, the HCF parameters will be applied to the entire area to obtain the heterogeneous quantitative characterization results of each well in the study area; if the effect is not good, it will return to the KPCA dimension reduction step and re-integrate the sensitive factor dimension reduction to form a new heterogeneous comprehensive characterization factor until the comprehensive characterization factor can better reconstruct the sensitivity factor curves, that is, the comprehensive characterization factor formed by dimension reduction fusion can represent the important characteristics of each sensitive factor.

2. The prediction method according to claim 1, characterized in that: The microscopic heterogeneity characteristic parameters of the target well include σ, FZI, Por, Perm and R 35 The required relevant data are obtained from experimental tests, including core mercury injection, particle size analysis and core physical properties.

3. The prediction method according to claim 1, characterized in that: The standard deviation σ of the core particle size of the target well is calculated using the standard deviation formula proposed by Falk and Ward, and the calculation formula is: In the formula, φ 84 -φ 16 is the concentrated part of rock grain size, φ 95 -φ5 represent the particle size sorting of the coarse tail and fine tail which are most sensitive to hydrodynamics.

4. The prediction method according to claim 1, characterized in that: The calculation formula of the core flow zone index FZI is: Among them, Φ e is the core porosity; K is the core permeability, mD; Φz is the ratio of pore volume to particle volume, and RQI is the reservoir quality index.

5. The prediction method according to claim 1, characterized in that: The FZIm formula obtained by improving the core flow zone index FZI using the rock electrical parameter m is: LgFZI m =LgRQI-LgΦ Z -(m-1)LgΦ e ; Among them, RQI is the reservoir quality index, Φz is the ratio of pore volume to particle volume, and m is the rock electrical parameter.

6. The prediction method according to claim 1, characterized in that: The kernel principal component analysis dimensionality reduction fusion method is obtained by the following expression: In the formula, is a D×D dimensional covariance matrix; X is a K×N dimensional sensitive factor matrix; η is a nonlinear mapping; η(X) is a D×N dimensional matrix; k(x, y) is a radial basis function kernel; p is a weight vector; λ is an eigenvalue; α is an eigenvector; ζ is an eigenvalue threshold condition.

7. The prediction method according to claim 1, characterized in that: The specific method of reverse reconstruction of sensitivity curve is: using kernel principal component analysis algorithm to learn HCF curve, so as to predict various sensitive factors.

Citation Information

Patent Citations

  • Rock heterogeneous quantitative evaluation method based on magnetic resonance imaging

    CN103353462A

  • Method for representing pulveryte anisotropism by using chemical-sedimentary facies

    CN107144566A

  • Fracture heterogeneity quantitative characterization method and system

    CN108008464A

  • Heterogeneous reservoir saturation field fine representation method

    CN108241785A

  • Characterization method of heterogeneity of super-strong heterogeneous oil and gas reservoir

    CN110242291A

Cited By

  • Clastic rock clastic particle size quantitative characterization method based on CT scanning structure image

    CN120510312A