Single-well reservoir classification rapid evaluation method based on principal component analysis method

Abstract

The invention discloses a single well reservoir classification rapid evaluation method based on a principal component analysis method, and the method aims at the rapid classification processing of a work area reservoir, and comprises the steps: firstly carrying out the correlation calculation of the geological parameters of a block oil and gas reservoir through a Spearman correlation coefficient method and a Kendall correlation coefficient method based on the principal component analysis method; taking a variable coefficient method as a processing means, performing weighting processing on calculation results of the Kendall and Spearman correlation to obtain comprehensive correlation, and determining geological main control factors according to the comprehensive correlation degree; based on the screened geological main control factors, the score of each component is obtained by using a principal component analysis method, and then the weight ratio of each component is obtained according to the variance interpretation rate, so that the calculation method of the principal component in the matrix and the weight of each component are determined, and finally the single well geological score is obtained, and the C-type reservoir (0-0.85), the B-type reservoir (0.85-1.25) and the A-type reservoir (gt; 1.25), and the type of the single well reservoir can be accurately, efficiently and rapidly evaluated.

Description

Technical Field

[0001] The invention relates to the field of petroleum engineering, and in particular to a single well reservoir classification rapid evaluation method based on principal component analysis. Background Art

[0002] The lithofacies of various oil and gas reservoirs are rich and varied, the pore structure is complex, the reservoir type classification is difficult, and the evaluation parameters cannot be unified. Reservoir classification is a key parameter for evaluating the storage capacity and seepage capacity of the reservoir, and it is also an important factor affecting the reservoir production capacity. Its related characteristic indicators are mainly based on the rock physical properties, geological structure and fluid characteristics of the reservoir. The screening of these characteristic indicators is a complex and systematic process, which involves multiple geological factors, which jointly determine the storage capacity and seepage performance of the reservoir.

[0003] At present, the methods for evaluating the classification of single well reservoirs are often based on nuclear magnetic resonance logging, hierarchical analysis method and cluster analysis. These methods have a long cycle, a small scope of application, and classify reservoirs in a general way, making it difficult to quickly and accurately identify reservoir classification on site.

[0004] The problems faced by single-well reservoir classification are: there are controversial intervals in the conventional reservoir classification standards, and some wells in the work area cannot accurately match the classification standards; the range of some characteristic parameters is not clear enough, and there are certain overlapping intervals; the parameters selected in each block are difficult to determine. These difficulties have led to the evaluation of single-well reservoir classification, and at the same time affected the smooth progress of the later transformation, which is not conducive to the rational and efficient development of oil and gas reservoirs. Summary of the invention

[0005] In order to solve the above problems, the present invention provides a single well reservoir classification rapid evaluation method based on principal component analysis. The method is aimed at rapid classification processing of reservoirs in a work area. Based on principal component analysis, the Spearman correlation coefficient method and the Kendall correlation coefficient method are first used to calculate the correlation between drilling and logging display, basic physical properties, natural fracture imaging interpretation and geomechanical parameters of the oil and gas reservoir in the block. The coefficient of variation method is used as a processing means to weight the calculation results of Kendall and Spearman correlation to obtain a comprehensive correlation. The main geological controlling factors are clarified according to the comprehensive correlation degree, and the main Component matrix; Based on the selected geological main controlling factors, the principal component analysis method is used to multiply the correlation between each main controlling factor and each component and then add them up to get the score of each component. Then, the weight proportion of each component is obtained according to the variance explanation rate, so as to determine the calculation method of the five main components in the matrix and the weight of each component. The total score of the component is positively scaled, and finally the single well geological score is obtained, which stipulates Class C reservoirs (0-0.85), Class B reservoirs (0.85-1.25), and Class A reservoirs (>1.25), and the single well reservoir type is accurately and efficiently evaluated. .

[0006] A single well reservoir classification rapid evaluation method based on principal component analysis method, characterized in that the method comprises:

[0007] Step S1, using the Spearman correlation coefficient method and the Kendall correlation coefficient method to calculate the correlation of drilling and logging display, basic physical properties, natural fracture imaging interpretation and geomechanical parameters of the oil and gas reservoir block;

[0008] (1) Spearman correlation coefficient calculation model:

[0009] r=1-(6*∑(d*2)) / (n * (n * 2-1)) (1)

[0010] Where n is the sample size; d is the difference between the corresponding ranks of the two variables; the Spearman correlation coefficient r ranges from -1 to 1: when r = 1, it means that the two variables are completely monotonically increasing; when r = -1, it means that the two variables are completely monotonically decreasing; when r = 0, it means that there is no monotonic relationship between the two variables.

[0011] (2) Kendall correlation coefficient calculation model:

[0012]

[0013] In the formula: x and y represent two sorted lists; C represents the number of consistent element pairs among any two element pairs in the two sorted lists; D represents the number of inconsistent element pairs in the two sorted lists, and N is the length of the sorted lists.

[0014] Step S2, using the coefficient of variation method as a processing means, weighting the Kendall and Spearman correlation calculation results to obtain a comprehensive correlation, and selecting parameters with a comprehensive correlation degree greater than a critical comprehensive correlation degree as geological main controlling factors;

[0015] (1) Coefficient of variation calculation model:

[0016]

[0017] In the formula is the geometric mean of each index; r ij is the value of the correlation of the jth parameter of the ith well; σ i is the standard deviation of the indicator; x ij For each indicator value, is the average value of each index; V i is the coefficient of variation of the i-th indicator, also known as the standard deviation coefficient; σ i is the standard deviation of the ith indicator; is the average value of the i-th indicator; ω iis the weight of the i-th indicator; W i The final score of the i-th indicator.

[0018] Step S3: According to the selected geological main controlling factors, the original geological main controlling data are standardized using the principal component analysis method to obtain a simple standardized score Z for each geological factor. j , the calculation model is:

[0019]

[0020]

[0021] Where x ij is the jth parameter of the i-th well; For standardized indicators; is the sample mean of the jth parameter; s j is the sample standard deviation of the jth parameter; x j is the value of the jth parameter; Z j The jth parameter simply normalizes the score.

[0022] Step S4, calculating the correlation coefficient matrix C and the eigenvalues ​​and eigenvectors of the matrix C from the geological main controlling factor data, determining the number of principal components, calculating the score of each principal component, and constructing a principal component confusion matrix;

[0023] (1) Correlation coefficient matrix C calculation model:

[0024]

[0025] Where r ij It is the correlation coefficient of the comprehensive correlation between the i-th well and the j-th parameter.

[0026] (2) Calculation model of each principal component score:

[0027]

[0028] Where y i is the vector of the i-th principal component; u nm is the eigenvector; X′ nm The normalized result of the mth parameter for the nth well; F n The nth principal component score.

[0029] Step S5: Calculate the weight of each component using the variance explanation rate according to the scores of each principal component of each parameter calculated in step S4.

[0030] Step S6, based on the calculation results of step S4 and step S5, the total score of the single well is obtained by summing up the weight of each component and the product of each component, and the reservoir type of the single well is accurately and quickly evaluated based on the classification indicators of Class C reservoir (0-0.85), Class B reservoir (0.85-1.25), and Class A reservoir (>1.25).

[0031] Furthermore, the geological parameters included in the embodiment should include the span of the transformation section, the leakage of the transformation section, the maximum total hydrocarbon, the formation pressure coefficient, the thickness of the oil and gas layer + the differential oil and gas layer, the porosity, the saturation, the fracture density, the proportion of 0-15 angle fractures, the proportion of 15-30 angle fractures, the proportion of 30-45 angle fractures, the proportion of 45-60 angle fractures, the proportion of >60 angle fractures, the proportion of 0-30 inclination angle fractures, the proportion of 30-60 inclination angle fractures, the proportion of 60-90 inclination angle fractures, the average minimum horizontal principal stress, the average horizontal stress difference, the Young's modulus and the Poisson's ratio, etc., and the data can be increased or decreased according to the actual situation.

[0032] As a further improvement of the present invention, when deciding on appropriate geological parameters, the Spearman correlation coefficient method and the Kendall correlation coefficient method are used to calculate the correlation results of each geological parameter, and the results are weighted added using the variable coefficient method to obtain the final correlation coefficient of the geological parameters.

[0033] As a further improvement of the present invention, a critical comprehensive correlation degree is established, and the critical comprehensive correlation degree is calculated according to the following formula:

[0034]

[0035] Where p is the critical comprehensive correlation degree (generally, the number of geological parameters greater than this value does not exceed 10); p i The comprehensive correlation degree of each geological parameter; S is the comprehensive correlation coefficient, which is generally 1.2 to 1.5.

[0036] Furthermore, the standardization process is for the oil and gas reservoirs in a certain block, and the score is the score of each geological parameter of each well, in preparation for the subsequent principal component analysis.

[0037] Furthermore, the variance explanation rate is calculated according to the following formula:

[0038]

[0039] Where S i is the variance explanation rate of the i-th principal component; i is the i-th eigenvalue; W i is the weight of the i-th principal component.

[0040] As a further improvement of the present invention, the total score of a single well is calculated according to the following formula:

[0041] F * =W1F1+W2F2+W3F3+…+W i F i (17)

[0042]

[0043] Where F * is the proposed total score of a single well; F is the total score of a single well.

[0044] As a further improvement of the present invention, it is defined as Class C reservoir (0-0.85), Class B reservoir (0.85-1.25), and Class A reservoir (>1.25), wherein the predicted open flow rate after compression of Class A reservoir is >2 million cubic meters / day, the predicted open flow rate after compression of Class B reservoir can reach 1-2 million cubic meters / day, and the predicted open flow rate after compression of Class C reservoir is <1 million cubic meters / day.

[0045] Furthermore, single-well reservoir classification based on principal component analysis significantly improves classification accuracy while being highly operational and not restricted by geographical location. It has promotional value and can quickly and accurately evaluate single-well reservoir types. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic flow chart of a method for rapid evaluation of single well reservoir classification based on principal component analysis provided by the present invention. DETAILED DESCRIPTION

[0047] The present invention is described in detail below in conjunction with embodiments. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0048] A rapid evaluation method for single well reservoir classification based on principal component analysis is proposed. The specific process is as follows:

[0049] Taking a well area with known oil and gas reservoir drilling and logging displays, basic physical properties, natural fracture imaging interpretation and geomechanical parameters as an example, 10 single wells were selected to carry out rapid reservoir classification evaluation.

[0050] S1. The Spearman correlation coefficient method and the Kendall correlation coefficient method were used to calculate the correlation between the drilling and logging display, basic physical properties, natural fracture imaging interpretation and geomechanical parameters of oil and gas reservoirs. The calculation results are shown in Table 1.

[0051] Table 1 Correlation statistics of various parameters calculated by Spearman correlation coefficient method and Kendall correlation coefficient method

[0052]

[0053] S2. Using the coefficient of variation method as a processing method, weighted processing of the Kendall and Spearman correlation calculation results can obtain a comprehensive correlation, and the critical comprehensive correlation degree is calculated to be 0.15. Eight geological parameters, including leakage in the reformed section, maximum total hydrocarbon, formation pressure coefficient, porosity, proportion of 45-60 angle fractures, proportion of 0-30 angle fractures, and proportion of 60-90 angle fractures, are selected as the main controlling factors of this block.

[0054] Table 2 Comprehensive correlation statistics calculated by coefficient of variation method

[0055]

[0056]

[0057] S3. Based on the selected geological main controlling factors, the principal component analysis method is used to standardize the original geological main controlling data to obtain a simple standardized score Z for each geological factor. j The calculation results are shown in Table 3.

[0058] Table 3 Standardized statistics of main controlling factors in a well area

[0059]

[0060] Table 4 Scores of main controlling geological factors of 10 single wells in a well area

[0061]

[0062] (Note: Red numbers represent negative numbers)

[0063] S4. Calculate the correlation coefficient matrix C and the eigenvalues ​​and eigenvectors of the matrix C based on the geological main controlling factor data. Here, it is calculated that there are 5 principal components. Calculate the score of each principal component and construct the principal component confusion matrix. The calculation results are shown in Table 5.

[0064] Table 5 The five principal component scores of the eight main controlling geological factors in a well area

[0065]

[0066]

[0067] S5. According to the scores of each principal component of each parameter calculated in step S4, the weight of each component is calculated using the variance explanation rate. The calculation results are shown in Table 6.

[0068] Table 6 Weights of five principal components of eight main controlling geological factors in a well area

[0069] name Variance explained (%) Cumulative variance explained (%) Weight Principal component 1 34.10％ 34.064 36.27％ Principal component 2 24.60％ 58.695 26.22％ Principal component 3 13.80％ 72.509 14.71％ Principal component 4 11.60％ 84.122 12.37％ Principal component 5 9.80％ 93.924 10.44％

[0070] S6. According to the calculation results of step S4 and step S5, the total score of the single well is obtained by adding the weight of each component and the product of each component. The scores of the 10 single wells are shown in Table 7.

[0071] Table 7 Geological scores of 10 single wells in a well area

[0072]

[0073] By comparing the open-flow rate predicted by the total geological score of a single well with the actual open-flow rate at set production, it was found that the accuracy was extremely high, and the reservoir type of a single well was accurately and quickly evaluated.

[0074] Although the embodiments disclosed in the present invention are as above, the above contents are only embodiments adopted for facilitating the understanding of the present invention and are not intended to limit the present invention. Any technician in the technical field to which the present invention belongs can make any modifications and changes in the form and details of the implementation without departing from the spirit and scope disclosed in the present invention, but the patent protection scope of the present invention shall still be subject to the scope defined in the attached claims.

Claims

1. A rapid evaluation method for single well reservoir classification based on principal component analysis, characterized in that: The method comprises: Step S1, using the Spearman correlation coefficient method and the Kendall correlation coefficient method to calculate the correlation of drilling and logging display, basic physical properties, natural fracture imaging interpretation and geomechanical parameters of the oil and gas reservoir block; (1) Spearman correlation coefficient calculation model: r=1-(6*∑(d*2)) / (n * (n * 2-1)) (1) Where: n is the sample size; d is the difference between the corresponding ranks of the two variables; the Spearman correlation coefficient r ranges from -1 to 1: when r = 1, it means that the two variables are completely monotonically increasing; when r = -1, it means that the two variables are completely monotonically decreasing; when r = 0, it means that there is no monotonic relationship between the two variables; (2) Kendall correlation coefficient calculation model: In the formula: x and y represent two sorted lists; C represents the number of consistent prime pairs in any two element pairs in the two sorted lists; D represents the number of inconsistent element pairs in the two sorted lists, and N is the length of the sorted lists; Step S2, using the coefficient of variation method as a processing means, weighting the Kendall and Spearman correlation calculation results to obtain a comprehensive correlation, and selecting parameters with a comprehensive correlation degree greater than a critical comprehensive correlation degree as geological main controlling factors; (1) Coefficient of variation calculation model: In the formula is the geometric mean of each index; r ij is the value of the correlation of the jth parameter of the ith well; σ i is the standard deviation of the indicator; x ij For each indicator value, is the average value of each index; V i is the coefficient of variation of the i-th indicator, also known as the standard deviation coefficient; σ i is the standard deviation of the ith indicator; is the average value of the i-th indicator; ω i is the weight of the i-th indicator; W i The final score of the ith indicator; Step S3: Based on the selected geological main controlling factors, the principal component analysis method is used to standardize the original geological main controlling data to obtain a simple standardized score Z for each geological factor. j , the calculation model is: Where x ij is the jth parameter of the i-th well; For standardized indicators; is the sample mean of the jth parameter; s j is the sample standard deviation of the jth parameter; x j is the value of the jth parameter; Z j The jth parameter simply normalizes the score; Step S4, calculating the correlation coefficient matrix C and the eigenvalues ​​and eigenvectors of the matrix C from the geological main controlling factor data, determining the number of principal components, calculating the score of each principal component, and constructing a principal component confusion matrix; (1) Correlation coefficient matrix C calculation model: Where r ij It is the correlation coefficient of the comprehensive correlation between the i-th well and the j-th parameter. (2) Calculation model of each principal component score: Where y i is the vector of the i-th principal component; u nm is the eigenvector; X′ nm The normalized result of the mth parameter for the nth well; F n The nth principal component score; Step S5, according to the scores of each principal component of each parameter calculated in step S4, the weight of each component is calculated using the variance explanation rate; Step S6, based on the calculation results of step S4 and step S5, the total score of the single well is obtained by summing up the weight of each component and the product of each component, and the reservoir type of the single well is accurately and quickly evaluated based on the classification indicators of Class C reservoir (0-0.85), Class B reservoir (0.85-1.25), and Class A reservoir (>1.25).

2. The method according to claim 1, characterized in that The geological parameters included in step S1 should include the span of the reformed section, the leakage of the reformed section, the maximum total hydrocarbon, the formation pressure coefficient, the thickness of the oil and gas layer + the differential oil and gas layer, the porosity, the saturation, the fracture density, the proportion of 0-15 angle fractures, the proportion of 15-30 angle fractures, the proportion of 30-45 angle fractures, the proportion of 45-60 angle fractures, the proportion of >60 angle fractures, the proportion of 0-30 inclination angle fractures, the proportion of 30-60 inclination angle fractures, the proportion of 60-90 inclination angle fractures, the average minimum horizontal principal stress, the average horizontal stress difference, the Young's modulus and the Poisson's ratio, and the data can be increased or decreased according to the actual situation.

3. The method according to claim 1, characterized in that: In step S2, the correlation of geological data is calculated by the Spearman correlation coefficient method and the Kendall correlation coefficient method to obtain the correlation result of each geological parameter, and the results are weighted added by the variable coefficient method to obtain the final correlation coefficient of the geological parameters.

4. The method according to claim 1, characterized in that: The calculation model of the critical comprehensive correlation degree in step S2 is: Where p is the critical comprehensive correlation degree (generally, the number of geological parameters greater than this value does not exceed 10); p i the comprehensive relevance of each geological parameter; S is the comprehensive correlation coefficient, which is generally between 1.2 and 1.

5.

5. The method according to claim 1, characterized in that The standardization process in step S3 is for the oil and gas reservoirs in a certain block, and the score is the score of each geological parameter of each well, in preparation for the subsequent principal component analysis.

6. The method according to claim 1, characterized in that The number of principal components of each geological parameter in step S4 should be less than the number of eigenvalues ​​of the correlation coefficient matrix C.

7. The method according to claim 1, characterized in that The variance explanation rate calculation model described in step S5 is: Where S i is the variance explanation rate of the i-th principal component; i is the i-th eigenvalue; W i is the weight of the i-th principal component.

8. The method according to claim 1, characterized in that: Calculation model for the total score of a single well: F * =W1F1+W2F2+W3F3+…+W i F i (17) Where F * is the estimated total score of a single well; F is the total score of a single well.

9. The method according to claim 1, characterized in that: The reservoir classification index is defined as Class C reservoir (0-0.85), Class B reservoir (0.85-1.25), and Class A reservoir (>1.25), among which the predicted open flow rate after compression of Class A reservoir is >2 million cubic meters / day, the predicted open flow rate after compression of Class B reservoir can reach 1-2 million cubic meters / day, and the predicted open flow rate after compression of Class C reservoir is <1 million cubic meters / day.

Images